This is a thread for discussion of November's favorites experiment. November 27, 2009 8:15 PM   Subscribe

This is a thread for discussion of November's favorites experiment.

The favorites experiment is ending in three days, and the old thread is kind of long. I thought it would be nice to have a new thread for people to post their impressions of how the experiment affected the tenor of discussion, the usage of favorites, and their general usage of the site.
posted by Anonymous to MetaFilter-Related at 8:15 PM (398 comments total)

Just to head people off at the pass, we're all still adjusting to mathowie being better and the end of the month holidays in the US, so there's no particular plan or announcement from Team Mod [except that things will go back to normal, as we sid they would, on Dec 1]. We just thought it might be nice to have a thread that loaded in under a minute. Thanks lalex.
posted by jessamyn (staff) at 8:21 PM on November 27, 2009 [1 favorite]


Or maybe we could wait for the people holding the experiment to post something.

You're a rat in a maze. You don't get to question the scientists!

Oh, maybe I'm wrong.
posted by cjorgensen at 8:25 PM on November 27, 2009


I tried it the new way for a week or so. It drove me batty, so I changed my preferences back to the old way. I don't know why it drove me batty, it just did.

I prefer seeing the favorite count. I logged in on a different computer where my preferences weren't changed. I clicked on a "has favorites" to just see the amount of favorites . . . and the list was a mile long. For some reason I really wanted to see a number - but dang I was too lazy to count them myself.

On my own comments, I like to see how many favorites it gets - not for a popularity contest, but maybe because I'm a little insecure and I like to see that the things I wrote meant something to someone, that I said the right thing, and that my ideas aren't completely crazy.

And that's all I have to say about that.
posted by Sassyfras at 8:25 PM on November 27, 2009 [35 favorites]


When the experiment ends, will we get to keep the option to not have the count? Because it has grown on me and I'd like to keep it if possible.
posted by Lentrohamsanin at 8:27 PM on November 27, 2009 [10 favorites]


Can I take my last comment back?

My take, I didn't like it. I felt like my totals were inflated, and felt like I (ironically) gave less. Maybe I actually didn't, but the balance of favorite to favoriting was closer going in than coming out.

I like the visual feedback of seeing how much something was liked.

I want the numbers back. That's my vote. But this isn;t a democracy either, so I'll live with it either way.
posted by cjorgensen at 8:28 PM on November 27, 2009 [1 favorite]


My data point: it didn't make much difference to me, but I was a little annoyed at how a self-favouriter could get the same amount of eyes to their post as someone with lots of favourites. Not annoyed enough to change it from the new way for the month, though.
posted by gaspode at 8:29 PM on November 27, 2009 [3 favorites]


I preferred either seeing the favorite count or not knowing if something had been favorited at all. The new way seemed like just enough info to make you wonder about the details, but not enough to satisfy.
posted by dilettante at 8:37 PM on November 27, 2009 [10 favorites]


I still don't like it. I also gave less favourites but that may be be cause my new toy is an iPhone and that little plus sign can be hard to hit.
posted by saucysault at 8:40 PM on November 27, 2009


I thought I'd HATE it but made the pledge to not turn the counts back on. You know, to see if I'd learn anything. As the month went on I realized that I didn't HATE it, but I'll be glad to see it go.

I'm not sure if I learned anything.
posted by dirtdirt at 8:41 PM on November 27, 2009 [1 favorite]


I think I like the new count-less way a little better. I was afraid it would discourage snarky one-liners (which I actually like having around), but I didn't notice any effect in that department. On the other hand, I think it's made it a little easier to read threads with disagreements and arguments, since there's less of a sense of one side winning or losing. Makes the atmosphere less tense, I guess.
posted by equalpants at 8:44 PM on November 27, 2009 [7 favorites]


I stuck with the new setup for only a day or three before switching back to the old. For me, it diminished the perceived utility of the site. Sure, there is a tiny bit of ego-boo when something I say interests ten or twelve people enough to favorite it (and the counterbalancing bafflement and gloom when a trenchant bon mot vanishes without a ripple), but mostly I find seeing the numbers of favorites steers me to things I might have overlooked otherwise. A post that ratchets the favorites count up into three digits is almost always an interesting read, even if I disagree or don't have a strong opinion.

I think the people who might shed some light on this in December are the users who have joined in the last month and have only experienced MeFi without favorites visible. What will they think when [faved] turns into [17 favorites] overnight?
posted by ricochet biscuit at 8:52 PM on November 27, 2009 [3 favorites]


In the interests of experimentation, I paid a lot more attention to favorites this month than I normally do (I'm a bookmarker, who didn't care about the counts) I had never realized you got a gold star if a comment got more than 15 favorites before (I don't have any contacts other than myself), which set me back on my heels a bit, and I spent more time checking out my own favorites, and consciously looking at the lists of favoritors for comments I loved and hated once I finished reading a post's comments.

I did discover a good half dozen people who - aside from having the good taste to appreciate what I had to say - are really interesting people whom I never consciously noticed before. So, in addition to honing my secret quonsar info-gathering skills, I've found some people who I now keep an eye out for. That was a really nice unexpected bonus for me.

I still don't think a post/comment having favorites really impacts how I normally read a thread - I just skip over that nugget of information, while reading from the top down to the bottom. I still abandon a lot of threads if I get bored or annoyed halfway through. HOWEVER, I'm a little discomfitted by how I went from reminding myself to check my favorite count to doing it automatically to doing it a little obsessively in a month. I don't want to be someone who posts for favorites; I liked it better when I posted simply because I had something useful or relevant to say without thinking about how much love it was going to get. So I'm cutting myself off! Cold Turkey! (without cranberry sauce, though, I'm saving that for my toast in the morning.)
posted by julen at 8:58 PM on November 27, 2009


On my way to Thanksgiving dinner yesterday, I saw a sign that said: "CORN MAZE." So I said, "Isn't that redundant?"

Yeah, it wasn't funny then, either.
posted by Eideteker at 8:59 PM on November 27, 2009 [8 favorites]


I preferred either seeing the favorite count or not knowing if something had been favorited at all

That's sort of what I thought. It was a bit of a fake experiment because it just presented favourites a different way. They should have been turned completely off to get any real value, experiment-wise. I think maybe 2 or 3 times I saw "has favorites" and thought, "Oh really? Who?" and clicked on it; otherwise I didn't really care. I still think favourites as voting is gross.
posted by chococat at 9:00 PM on November 27, 2009 [3 favorites]


I want to know if the strident anti-favoritists perceive an improvement in comment quality. It all seems the same to me (which isn't a bad thing IMO).

Disclosure: I opted out from day one.
posted by cj_ at 9:00 PM on November 27, 2009 [1 favorite]


I tried it for a few days and ultimately I opted out. I think the "YES!"-ness of favorites was diminished when you couldn't see the number, especially in AskMe. I want to know when people are saying "YES!". If that is wrong then let me be damned for it.
posted by hermitosis at 9:10 PM on November 27, 2009 [2 favorites]


I installed the GreaseMonkey script as soon as it was posted in the old thread.

I changed my prefs as soon as that was an option.

I installed the script on my work computer, because I don't log in on it and I wanted the normal version of things.

As long as I can see things the old way, I don't really care if things change; I do think some set of phrases to distinguish between "tiers" of favorite activity would be a good idea, though, if that is going to stick around.
posted by paisley henosis at 9:10 PM on November 27, 2009 [1 favorite]


The net effect of the experiment on my experience of MetaFilter was minimal. I can't say that I noticed a substantial change in the tone of discussion on the site - though maybe that's just because I chose to follow quite a few contentious threads this month. I didn't notice a significant change in the way I favorite, either. I wouldn't say I prefer this system, but I didn't hate it.
posted by EvaDestruction at 9:10 PM on November 27, 2009


I found it annoying for about 2 days then turned it off. Favorite counts are both useful information about other people's posts and a fun ego stroke when I get them on mine. I missed them while they were gone.
posted by jacquilynne at 9:19 PM on November 27, 2009 [3 favorites]


The original thread caused me to make a contact finally. Suddenly the contacts sidebar appeared. It was a revelation! I added a few more over the month, and now I see why people sometimes might get so panty-bunched about favorites. The contacts *really* changed it for me. But it drove me nuts how far I had to scroll for the sidebar announcements, and I realized that most of my contacts only occasionally had something interesting (to me) to say. Maybe a week and a half ago, I closed my contacts sidebar, and it's much better for me. I still have favorite counts displayed - I've flipped it off about three or four times this month and I always can only handle it for about 18 hours before I have to flip the numbers back on.

I felt like I received less favorites than I otherwise would have, but I gave just as many as I always would. In AskMe, I often reiterated other people, and then added my own tweak to their answer, when before I would have favorited their answer and just mentioned my notes sans-ditto. It made me feel inefficient and like I was trying to take credit for the other person's idea, sometimes.

In short: discovery of contacts causes brief clique environment to form, followed by disillusionment as to content provided by contacts. No guarantee of others seeing my individual favorite cause repetitive commenting in AskMe. Suspicion that "has favorites" next to a comment causes a person to be less likely to favorite it, rather than a number they can add to.

And those were my thoughts! Thrilling, I know.
posted by Mizu at 9:22 PM on November 27, 2009 [2 favorites]


I turned it off as soon as I was able. I like seeing counts, mainly to be sure I take care to be sure I spend a little extra time investigating heavily favored comments; there is often gold there.

I wonder, though, how much of my initial distaste was simply the horrible "faved" thing.
posted by maxwelton at 9:24 PM on November 27, 2009


What I've found is that the current system makes me more conscious of favoriting than it otherwise would.

If I'm considering favoriting a comment, I'm going to check to see how many other favorites it has. But, given that this requires clicking on a link and opening a new page, it means that I'm paying more attention to the information than I otherwise would. So, instead of the mental process being, "Oh, 5 favorites? Now it's six," it's more like, "Hm, maybe I want to favorite this. I wonder how many..... Oh, five, huh? And those are the people who favorited?"

I don't really like it. I feel more like I'm participating in a popularity contest, because I just can't help but pay attention to who has favorited something, and the whole discovery process of how many favorites is so much longer.

It's not the most important shift in the world, no, but I prefer it the original way.
posted by Ms. Saint at 9:25 PM on November 27, 2009 [1 favorite]


When the experiment ends, will we get to keep the option to not have the count?

If you use Greasemonkey, yes. I first wrote that to remove "faved" (then the current "has favorites"), but it's flexible enough to just hide counts.

(Comments & suggestions welcome on the project page.)
posted by Pronoiac at 9:26 PM on November 27, 2009


I was afraid it would discourage snarky one-liners (which I actually like having around)

pansy
posted by mannequito at 9:29 PM on November 27, 2009


I think the much more interesting experiment would be to remove the favorites count on your profile page. Or to change the name to, simply, "bookmarks," which could take this useful tool out of the realm of crass popularity contest.

I have literally no interest in whether someone else's comments are popular, so whether a comment "has favorites" or "has 39 favorites" (or whatever the old version used to say), my reading does not change. I do enjoy seeing my comments favorited, but that is pure foolish vanity, and I'd just as soon not know, or not have a global tally. It's a waste of time.

Can we have an option to leave the experiment on? I don't care to go back to the old system.
posted by Admiral Haddock at 9:29 PM on November 27, 2009 [1 favorite]


I found "has favorites" distracting. Take out the word "has" and it would be perfect.
posted by water bear at 9:30 PM on November 27, 2009


I second what Sassyfras said.
posted by amyms at 9:36 PM on November 27, 2009


I've said it before but will say it again here: I've enjoyed the no-numbers way. I would heartily endorse having an preference-setting to have the no-numbers display.
posted by LobsterMitten at 9:40 PM on November 27, 2009


(ie, I'd like to have the option to "keep the experiment on" in my own case)
posted by LobsterMitten at 9:41 PM on November 27, 2009


cj_: “I want to know if the strident anti-favoritists perceive an improvement in comment quality. It all seems the same to me (which isn't a bad thing IMO).

Disclosure: I opted out from day one.”

I will take up the "strident anti-favoritists" side, but unfortunately my computer was in the shop for most of the month and I wasn't really on here enough to partake in a few epic discussions and test it out.

Frankly, though, my biggest reasons for hating favorites are all actually in that original epic favesperiment thread: score upon score of users came there to complain that they suddenly had no way to communicate which comments they thought were right – and, more tellingly, at least half a dozen people told me personally there that they were disappointed that they were suddenly unable to tell me I was wrong by favoriting someone else. I don't want to continue the shitfest that happened over there, but the sheer volume of people unashamed and unafraid to admit openly that they use favorites as simple votes demonstrated to me that favorites have clearly gone beyond the neat little bookmarking and highlighting tool they were intended as to become part of the conversation - and in some cases even to eclipse the conversation. Bookmarking and highlighting are fine things, but I dislike favorites because they bring voting into it, and voting tends to end conversation by increasing exponentially the social disapproval of those who are perceived as unpopular.

We all - myself included - subconsciously approve more highly of people and things which we perceive as being more popular. I think that's a subconscious action that often leads us to make bad choices.

If there's anything this experiment did teach me about, however, it's the resilience of the people who run the site. My initial reaction was, ah, well, Metafilter's too far gone down that road to change now anyway. But the mods here have shown, I think, that they can institute a change and see it through without a hitch, even if a huge amount of heat is on them. I commend them for this, and I'd like them to know that beyond simple respect from me they've earned my happy surprise and a certain amount of hopefulness as I come to realize that Metafilter is actually even more within their control and malleable than I'd thought. That is pretty cool, and it'll be neat to see where things move in the future.

I know, though, that Matt & co don't mind favorites, so I don't expect to see this change stay permanent. That's fine. I don't mind the presence of favorites, but I do hope people stop and think at least every once in a while about the fact that the favorited comments are not necessarily the most interesting or the truest, and the least-favorited comments might sometimes be the best ones.

I also know that I'm going to greasemonkey favorites away from my own view of Metafilter permanently, so that I don't see a count or even a "has favorites" marker under comments. I'd rather not know what was popular, or what comment is getting the most votes in the poll. People can go ahead and tell me that I'm using metafilter wrong (as they all did in that thread) but it's the way I like it.

Finally, here's hoping Matt continues on the road to recovery.
posted by koeselitz at 9:41 PM on November 27, 2009 [13 favorites]


equalpants: “I think I like the new count-less way a little better. I was afraid it would discourage snarky one-liners (which I actually like having around), but I didn't notice any effect in that department. On the other hand, I think it's made it a little easier to read threads with disagreements and arguments, since there's less of a sense of one side winning or losing. Makes the atmosphere less tense, I guess.”

That's really the best statement of everything I've had against favorites that I've read anywhere.
posted by koeselitz at 9:43 PM on November 27, 2009


I agree with sassyfrass when he says that I like to see the amount of favorites on my comments without having to click an extra thing or go to my profile. On the other hand, the new system isn't very annoying, and I actually thought people who made a big deal about in the other thread were being ridiculous.

I think I decreased the amount of favorites I give out, but that may be completely false. I do know that I tended to do this:

Let's say if there were two borderline favoritable comments (meaning they're the same quality/ say the same thing). If comment A already had favorites, I probably wouldn't favorite it, but if comment B didn't have favorites I'd favorite it over comment A. So I guess I used it to even out the favorites on a thread.
posted by azarbayejani at 9:49 PM on November 27, 2009


cjorgensen: “I want the numbers back. That's my vote. But this isn;t a democracy either, so I'll live with it either way.”

NO NO NO NO

I don't mean "no" you shouldn't get numbers back. I mean NO, LET'S NOT HAVE THIS DISCUSSION UNDER THE IMPRESSION THAT WHAT WE SAY HERE CHANGES ANYTHING. It's not to say that the mods aren't open to suggestions, but this thread is not about deciding what to do. The mods have already stated very, very clearly that FAVORITE COUNTS WILL BE RESTORED AT THE END OF THIS MONTH and we will go back to business as usual on December 1, just as things were before November 1 when the experiment started. There is no option on this table or any other table to keep things the way they are. (I'm with LobsterMitten - I'd like a little checky-box-thingy on my user prefs page which let me keep things like this - but I'm not too het up about things going back, since I'm looking forward to figuring out some Greasemonkey.)

I'm actually glad that we can talk about things already knowing what's going to happen. So let's all stay chill and not stress about having to convince the mods, ourselves, or everybody else of something that's already decided, eh?
posted by koeselitz at 9:53 PM on November 27, 2009


koeselitz, OK, but that wasn't my question. I know -- in great detail -- why you don't like favorites from the last thread. My question is, do you think taking them away changed the quality of comments in the past month? Is this bandwagon effect, which you assert is caused by favorite counts, now gone? Do we have less "favorite bait" snark? Are contentious threads going differently now that you can't see who's "winning?"

I mean, I get you don't like to see them. But you can always script that away. The actual issue here is whether they effect the quality of the site at all, and if so, for better or worse. Er, right? I'd like to think your issue here isn't just that it pisses you off how other people use the site if it has no effect on anything.

Seems like that's what people should be talking about at this point rather than why they hate/love favorites. That isn't just a dead horse, it's a bloody pulp. My personal take is nothing whatsoever has changed in the substance of comment threads, but this is pretty subjective. I'm willing to hear some arguments that things are way better now. I am so certain that nothing has changed that I don't even care if they don't switch it back, as long as I can opt out so I can use the site how I like to use it.

I'd also like to here some number crunchy stuff. Has total favorites given this month dropped? Has comments-per-thread in AskMe increased? What % of users opted out? These numbers are crucial to making sense of the experiment. I'm guessing we'll get that data when the experiment actually ends.
posted by cj_ at 9:58 PM on November 27, 2009 [2 favorites]


I was someone very pro-favorites.

I liked metafilter without favorites okay, actually.

What I liked: I favorited things because I liked them, and not because other people did or didn't. Often I favorited someting and was surprised to see only one or two others had. I'm not sure I would have favorited it in the first place if I had known that.

What I didn't like: I always found myself assuming "has favorites" meant "has a decent amount of favorites." This confused me when it was a really obnoxious comment. A few times it gave legitimacy to hateful/sketchy comments in my mind.

Conclusion: I like the idea of keeping the option. I really wish I could mouseover to show # of favorites, or something - clicking to open a new window is so tedious and awkward.
posted by Solon and Thanks at 10:00 PM on November 27, 2009 [3 favorites]


Also in the spirit of chillness and respect for the mods maybe people can post their thoughts once - maybe twice - and keep this thread from becoming a giant back-and-forth argument. Since this is just a reaction, not "convince other people" thread.
posted by Solon and Thanks at 10:04 PM on November 27, 2009 [3 favorites]


I thought the Dec. experiment was kind of pointless. It ended up being divisive and so easily circumvented that it may as well have never taken place.

But I'm an anti-favorites person, and every time I see a comment like "my numbers are inflated" or "I was expecting more faves from that comment" or "I use favorites to agree/disagree" with posts I remember why.

I'm over it, though, because it's never going back to how it used to be, and how it used to be had plenty of problems too.

Happy Thanksgiving USAians (yesterday? day before? dunno anymore)

You can all be my favorites today.
posted by Joseph Gurl at 10:10 PM on November 27, 2009 [1 favorite]


I've been a reader of Metafilter since before favorites, but I've only been a poster since relatively recently. So my history as a contributor has always involved favorites and the attendant 'yardstick of quality' sense you get from them, especially favorite counts on my own comments and posts. The attention I pay to favorites is frankly embarrassing to admit.

At the same time, there's almost nothing I can't put up with for one measly month, and the most surprising thing about this experiment for me has undoubtedly been seeing how many people couldn't stand browsing the site without visible favorite counts. I wonder if the people in the original thread (and this one) who say they opted out are really a representative sample of the whole population here. I don't mean to sound judgmental but c'mon, where's your sense of adventure? One part of me thinks the people who reacted most negatively to the change are the ones who would have benefitted the most from toughing it out.

And I think "toughing it out" is a pretty fair characterization. It definitely made my favorite site on the internet more annoying than it was before -- especially at first -- and as much as I try to remember it's just a website, Jesus, calm down, it really had an effect. It was like losing a tooth; it's really hard to concentrate on anything else for a while because your tongue keeps finding the hole where it used to be.

I think most of the backlash / bad experience anecdotes with the experiment are going to deal with how it affects the consumption of the site, and that rings true to me. It definitely affects how I read the site more than it affects how I write on the site. No question I've scrolled right past a fantastic-but-longish comment or two because there wasn't a [96 Favorites] at the bottom that compelled me to scroll back to the top and dig in. Don't get me wrong; there's always threads in which I really take the time to read every comment, but there's also plenty of threads I'll skim, trawling for one-liners and other snackable quips, in which my eyes will automatically glaze over on a 4-to-8-paragraph comment unless something about it grabs me or I recognize the commenter's name.

But here's the biggest change in how I interact with the site, and this totally blindsided me: I know who favorites what comments. Because when you want to know how many favorites your clever comment has -- that is, when you want to ascertain the 'missing' information -- you click on "has favorites" and instead of just a number, you also get the names in a list.

And this is kind of a revelation, because it replaces an abstract number with a much more 'tangible' thing. I've probably learned/remembered more usernames in the last month than any other month since I signed up. And it's because in order to see how many favorites something has -- whether it's a quip I wrote, or a post I was curious about, or a comment I thought was funny or brilliant or referenced something I thought was ultra-obscure -- I have to click and see who favorited it. And then I notice the person who favorited super-narrow-interest-but-cool FPP x also favorited my comment y, and I get curious and check out what FPPs that user has made, and who they like to favorite, and so on.

It's a subtle change in my 'click-flow' that ended up having a surprising effect on how I discover other users, and how often. If/when it changes back, I think it's actually going to seem really impersonal to me to see [{number} favorites], and I'm going to be much more compelled to click on that number, whether it's my comment/FPP or someone else's. It's nice to think that "how many" is a little less important, and "who" is a little more.
posted by churl at 10:28 PM on November 27, 2009 [4 favorites]


Prior to the experiment I tended to use favorite counts differently depending on the situation. In silly SLYT posts, all I really wanted was snarky one liners and favorite counts enabled me to scan threads for those delightful jokes. In those threads I feel that people frequently fish for favorites, hoping their one liner would resound with the community. However, this behavior harms the signal to noise ratio, making it difficult to find those gems without the favorite counts.

In pithy debate threads, I tended to read more of the whole thread, rather than just skimming for heavily favorited posts. This was good for me, as long as I had the time to sort out the thread.

I agree with equal pants that no longer having the count removes the aspect of clear winners and losers in discussions. I'm not sure this is a good thing. I think that we as a community should welcome different perspectives, but visible favorite counts reinforce ways of thinking and writing that we approve of. The visible peer moderation through favorite counts helps to create the culture that we all hold so dear.

I've see instances where someone will drop a snarky unproductive comment and get a heap of favorites. Someone will then inevitably post a witty, well researched, personal and well written rebuttal that quickly gains hundreds more favorites. I think this pattern is one of my favorite things about MeFi. We can be some sassy folks, but we do reward someone who puts thought into what they post. That visible count symbolized an affirmation of what the community values. I made it the whole time without turning the counts back on, but I almost did for the Schrodinger's rapist thread. I kept clicking on the [has favorites] of the amazing comments by women that were brave enough to share. When I reflected on why I cared, I think I wanted to make sure that these people were rewarded for their bravery publicly.

Overall, it didn't affect my use too much. I just enjoy seeing brilliance rewarded.
posted by JimmyJames at 10:28 PM on November 27, 2009


cj_: “koeselitz... I mean, I get you don't like to see them. But you can always script that away. The actual issue here is whether they effect the quality of the site at all, and if so, for better or worse. Er, right? I'd like to think your issue here isn't just that it pisses you off how other people use the site if it has no effect on anything.”

Well, like I said: I wasn't here at all this month, and therefore I haven't read anything that happened here this month; so it'll take me some time to sift through the last few weeks, see what happened here, and decide how I feel about it. No, favorites don't just piss me off. I just can't talk about the effects of the experiment yet – all I have to discuss now is the rational reasonings behind my perspective – so, yeah, maybe I should not have commented here until I review the last month of Metafilter.

[The issue has never been that it pisses me off how other people use the site. In fact, I spent most of the last thread on this being told I was using Metafilter the wrong way. I don't think it's worthwhile to immediately bring this to the point where we all assume that everybody else is telling us what to do.]

Off I go...
posted by koeselitz at 10:33 PM on November 27, 2009


As I mentioned in the prior thread, I reluctantly participated in the "experiment" by hiding favorite counts for a few weeks, but I felt that I was missing part of the texture and subtexture of the ongoing conversation that is Metafilter, so I turned them back on. I don't regret either decision, but I am glad that Metafilter will be returning to a single way of reading and viewing the site.
posted by dersins at 10:34 PM on November 27, 2009


After six days of noticing absolutely zero difference in the flow and tone of MeFi discussion w/ invisicounts, I switched back to the proper mode. Losing the fun and utility of visible counts was pointless, as it derived no benefit whatsoever. I'm looking forward to the end of the experiment.
posted by EatTheWeek at 11:28 PM on November 27, 2009 [3 favorites]


i add up all my favorites
about a hundred times a day
i keep a tab, a running count
now tell me what'd I say...
i said i gots to count 'em up
i'm gonna count 'em til i die
and when i reach that great beyond
i'm gonna count 'em in the sky
i'm gonna count 'em in the sky
posted by flapjax at midnite at 11:33 PM on November 27, 2009 [4 favorites]


I stopped noticing it after a few days, really. I still obsess over how much I get favourited, and occasionally I will click through to see who favourited another comment/post, at about the same rate I would when the number was displayed.

I have noticed in the past that if a comment has a bunch of favourites I am reluctant to favourite it myself, because I hate being part of a crowd. I no longer have to worry about that and will [+] away if I feel like it, without feeling like I may have been dirtily influenced by existing popularity.
posted by jacalata at 11:36 PM on November 27, 2009 [1 favorite]


Is it possible to see how many people opted out?

Of course, the fact that we could means that the 'experiment' didn't really test out what life would be like without favourites, as those who enjoy them opted out and continued to act as they normally would.
posted by twirlypen at 11:37 PM on November 27, 2009 [1 favorite]


Didn't really enjoy knowing something had 1+ favourites. That seems needlessly coy.
But give me the option of banishing favourites visibility altogether and I'd take it.
posted by Durn Bronzefist at 11:39 PM on November 27, 2009 [3 favorites]


Didn't like the new way, changed it back to the old way as soon as it was an option.
1) I like knowing how many favorites something has. I don't think if affects my choice of whether or not to favorite something very much, if at all.
2) Something about it bothered me aesthetically.
posted by ishotjr at 11:39 PM on November 27, 2009


koeselitz writes "I also know that I'm going to greasemonkey favorites away from my own view of Metafilter permanently, so that I don't see a count or even a 'has favorites' marker under comments. I'd rather not know what was popular, or what comment is getting the most votes in the poll. People can go ahead and tell me that I'm using metafilter wrong (as they all did in that thread) but it's the way I like it."

I'll probably be doing this too. I've liked the lack of attention drawing that invisible favourite counts enabled when I'm reading posts. If I never see 35 favourites on that stupid Simpsons monorail bit it'll be too soon.
posted by Mitheral at 11:44 PM on November 27, 2009 [1 favorite]


I like it without the numbers and hope to keep them turned off for this reason:

I think it's made it a little easier to read threads with disagreements and arguments, since there's less of a sense of one side winning or losing. Makes the atmosphere less tense, I guess.

During the argumentative thread I was involved in, I could listen for the merits of the arguments, without having to know that when someone said something snarky and mean, 13 other people were saying "yeah!" Metafilter has felt more peaceful to me this month.

I do sometimes turn them on when I want to see where the opinion is falling out or what statements are most popular, but I still find the most favorited statements to be funny and well-said more often than well-argued, so it's not a good metric for me if I'm wanting to get the meaning out of the thread.

So, count me with the "invisible favorites" crowd. I like marking my own, I like knowing who favorited mine, but I don't care about the score. I read for my own enjoyment and prefer not to be inflicted with others' opinions. When I read newspapers, I don't know how many of my neighbors agree with a particular article, and it's fine with me if Metafilter is like that most of the time, too.
posted by salvia at 11:59 PM on November 27, 2009 [1 favorite]


I left the experiment on because I wanted to see if I'd notice the difference, and because I am too lazy to switch it off. I like favourites and I like numbers, so my opinion is thumbs down.
posted by sgrass at 11:59 PM on November 27, 2009


Oh, and one last bit:

I love metafilter. I love all you favorites-loving people. So this isn't a battle for the soul of it all; just something I believe.
posted by koeselitz at 12:29 AM on November 28, 2009


It was a good experiment and we should continue with obscuring the favorite counts on comments.
posted by caddis at 12:32 AM on November 28, 2009


I think you jumped the gun here.

My vote? Turn off favorite counts for new users, but let them turn the counts on if they want. Settings for current users stay the same.
posted by dunkadunc at 12:38 AM on November 28, 2009 [1 favorite]


The experiment is annoying. I look forward to it ending.
posted by orthogonality at 12:39 AM on November 28, 2009 [1 favorite]


It wasn't broke. I haven't, from day one, understood one reasonable reason for this whole snafu.
posted by allkindsoftime at 12:45 AM on November 28, 2009 [13 favorites]


Disclaimer: I've never been a fan of the term 'favorites' - to me, it's always had connotations of being like tokens in a popularity contest, and popularity contests in large groups either tend to reinforce groupthink or be divisive. Given that, I've always tended to use favourites here as a tag denoting anything from the occasional "couldn't have said it better myself" through "I must think about that some more and come back to it" to the much more likely "there's an interesting looking link, I might come back and follow it up sometime (but I know damned well I probably never will)*".

(And, just looking now, I've favourited much more often over the last 5 or so years than I thought I had - an average of once every 5 days or so. Hmmm…)

The interesting thing I've noticed for myself over the last month - and backed up by that quick skim of my recent favouriting behaviour - has been that I've been more likely to favourite comments for "good comment" or "I agree" reasons - the thinking persons AOLian "me too!" - than anything else. That surprised me; I really do try to limit my MeFi favouriting to things for future follow-up or reference reasons. I'm not sure why that should be, because I'm also aware that I've deliberately not favourited comments in the lat month because I didn't know if they'd already had 1 favourite or 100.

In short, my favouriting behaviour has changed, but it seems to the opposite of what I thought would happen. Interesting, & I might informally analyse it more sometime soon.

(* I've already got hundreds of assorted .url files scattered around my hard drive, and my browser bookmarks too, full of links like that. I favourite things on MeFi to avoid filling both up even more with stuff I intend to look at but know I never will…)
posted by Pinback at 12:48 AM on November 28, 2009


I liked the in-line counts off and I'll probably keep it that way. When I was curious about a comment it was not hard to check.

I would be interested in a display that's quantitative but not numerical. I care more about difference between "many" and "few" than I care about "ten" versus "twelve," or "89" versus "208."
posted by fantabulous timewaster at 1:14 AM on November 28, 2009


I turned it off as soon as possible, even to the point of installing a user script thing before the user option was put in. So I haven't really noticed it personally, but it does seem like people are favoriting less.
posted by delmoi at 1:17 AM on November 28, 2009


I joined during November, and almost immediately switched favourite counts back on as soon as I had the option. I didn't dislike not having the count, as such, but MeFi is a community as well as a collection of interesting links, and part of that is that people will like/agree with/vote for/bookmark things that they find good/agreeable/interesting. That's information that I like to see, personally, so put me down for "prefer having the numbers".

I don't think there's anything wrong with using favourites as a "me too" or "I like this" either. Speaking as someone for whom favourites as a whole are something of a novelty, I think it's really nice to see the tiny bit of goodwill that someone sent me, and to be able to do the same for others. It makes my day that little bit happier.
posted by ZsigE at 1:24 AM on November 28, 2009 [1 favorite]


I liked the change and the effect it had on my commenting behavior. I hope we will still have the option to keep it turned on.
posted by painquale at 1:30 AM on November 28, 2009


I'd like to keep my count off I can!
posted by ageispolis at 1:42 AM on November 28, 2009


*if I can.

B e e r.
posted by ageispolis at 1:43 AM on November 28, 2009


I said in the original thread that I was in favour of The Hot New Shit. For the first day I have to say that I did not like it. (Such hypocrisy.)

Strangely though, I have become accustomed to its face and have not actually disabled the option. Consequently, I'm actually now rather fond of the New Way; It forces one to read more and apply more critical thinking (If I have a fault - WHICH I AM NOT PREPARED TO CONCEDE - it is that I tend to believe everything I read.)

So. Yeah. You know?
posted by Jofus at 2:20 AM on November 28, 2009 [1 favorite]


Having a score next to everyone's contributions to the conversation was vulgar and horrible, and I'm sad the mods ruled out making the change permanent so early on in the experiment.

Is everyone who doesn't want counts going to be required to switch them back off on December 1st? I'd hope not.
posted by cillit bang at 2:25 AM on November 28, 2009 [1 favorite]


I don't necessarily need to see favorites all the time, but between contact activity sidebar, and clicking on the 'has favorites' to get the details, I'm not sure this experiment made me any less aware of them.
posted by BrotherCaine at 2:32 AM on November 28, 2009


It tells me that this is an alphabet thread.

Anyways, I just wanted to put the notion out there that the spike in favorites this month so far, as compared to last month, doesn't tell us anything other than that there were more favorites this month than last month, so far. This is because we cannot assess why the spike occurred. For example, here are some potential reasons:

(Not an exhaustive list. List items may be compounding and/or combined. Some items may be bullshit.)

A) A random increase in activity
B) Consciousness about favorites causes people to pay attention to them more, and thus, favorite more
C) Interest in epic threads, such as Schroedinger's Rapist, Mefi Study, Raffle, Mathowie's Get Well
D) Holiday season/winter/November
E) Outside factors, such as political events, disasters, news
F) Inside factors, such as site changes, flameouts, culture shifts
G) The Favorites thread, awareness of
H) Internet factors, such as activity spikes, decreases, or changes on other sites such as Digg, Reddit, 4chan, etc.
I) Low favorites activity in October
J) Rising unemployment rates = more people with time and internet
K) Pie
L) Increased involvement post Cortex Sees MetaFilter
M) Removal of favorite counts
N) Misplaced vitriol for the term 'faved'
O) Increase in user signups
P) Bookmarkers Revolt!
Q) Favoriter Underground Uprising!
R) Tehloki
S) People fucking with teh Mefi
T) Trolls and data skewers
U) Accidental clicking
V) Cats on keyboards
W) Increase in AskMe questions
X) Increase in a specific type of AskMe question, preferential to a popular demographic on Metafilter
Y) Fewer flameouts, less snark, and/or more cooperation
Z) Mod love
posted by iamkimiam at 2:59 AM on November 28, 2009 [3 favorites]


Is everyone who doesn't want counts going to be required to switch them back off on December 1st? I'd hope not.

Nobody knows the trouble I seen.
posted by flapjax at midnite at 3:28 AM on November 28, 2009 [1 favorite]


Decided to turn the experiment off. It just doesn't look right. If I see "favorites", I want to know how many without extra clicking.

I agree that if numerical counts are cut out, there should be a vague definition of how popular a post is. While tastes here are usually predictable, that's not always the case. Especially considering that one was posted years before favorites.

On second thought, sometimes ignorance truly is bliss.
posted by Saydur at 3:29 AM on November 28, 2009


> so, yeah, maybe I should not have commented here until I review the last month of Metafilter.

Sorry, I didn't mean to shout you down, I'm genuinely curious if anyone thinks the site has changed at all in the past month rather than rehashing all the same arguments from the original thread. My own feeling is that the good, bad, and the ugly of Metafilter is influenced by things other than favorite counts. But I'm open to evidence otherwise. For example, I feel the ability to downvote people has a huge impact on the tenor of sites like Reddit and Digg, so the idea isn't alien to me. It's just, I think there are a factors contributing to this that are more relevant than the actual number sitting next to the post -- such as how comments with lower scores get pushed to the bottom of the thread, and eventually made invisible, while posts with lots of favorites are moved to the top. None of that happens here, it's just a number, and most people only see it after reading the post.

I myself did not take a break in November. I feel like nothing changed with regards to people's concerns: early thread-shitting, snarky one-liners, pile-ons in contentious threads, etc., is still there. Maybe this stuff has nothing to do with favorites so much as there are some users who behave that way? I was not convinced before that they behaved this way just to see their favorite count rise, and I'm even less convinced now that I haven't seen that behavior abate.

I do think that a setting to remove favorites from the site should be added for people that really don't like them, though. Like, keep the "experiment" as an opt-in feature, but go all the way and just don't show them at all. I don't think it would hurt anything and if it made some people happier I'm all for it.
posted by cj_ at 4:14 AM on November 28, 2009


The fact that a "turn it off" hack was quickly cobbled and tacked on tells me two things:

1) This experiment was not sufficiently thought through prior to implementation.
2) (placeholder for future edit)
posted by Meatbomb at 4:31 AM on November 28, 2009 [1 favorite]


i didn't care for it and opted out - people's commenting behavior didn't seem any different that i could see
posted by pyramid termite at 4:32 AM on November 28, 2009


I've kept the new setting the entire month, and I can say with some confidence that I don't really give a damn either way.
posted by stavrosthewonderchicken at 4:59 AM on November 28, 2009 [2 favorites]


I didn't care for the new favoriting system. When I'm reading a thread just for the sake of reading or one I may relate to, I use the number of favorites each answer has to filter out all the other 23 answers that have none. It lets me skim and read particular gems.
posted by biochemist at 5:06 AM on November 28, 2009


Opted out as soon as there was an option to.
posted by fire&wings at 5:20 AM on November 28, 2009


we will go back to business as usual on December 1, just as things were before November 1 when the experiment started. There is no option on this table or any other table to keep things the way they are.

I know they said they'd change it back on Dec 1, but did they say they'd never go back to the November system? I don't think so, which means we can't really say that "there is no option on the table" to keep it the way it is now.
posted by Jaltcoh at 5:29 AM on November 28, 2009


Also, using bold and all caps doesn't necessarily make your comment more important.
posted by Jaltcoh at 5:30 AM on November 28, 2009 [2 favorites]


I'm not sure it was a good experiment, since there was an officially sanctioned opt-out button. I'll be interested to hear the stats - for example, were there more or less favourites overall this month? Did the top comments have as many or fewer favourites this month compared to other months?

But I doubt people would have changed their commenting habits or the way they read the site. Unless they were blindly unaware of the experiment (which most users would have been), but there's no way to get stats on their posting/favouriting trends, is there?

I opted out after a few days because I was too used to the old way. If it had been "has favourites" all along, that might have been something I was okay with - but seeing how many favourites a comment has does affect how I skim a long thread, so I couldn't stick with "has favourites" very long.
posted by crossoverman at 5:51 AM on November 28, 2009


I don't have a preference, but I will say that I personally have been favoriting less and that from checking my own profile stats - either I have been way less compelling or people have also been favoriting ME less.

Except for that weird bit about how I blarf out my medical history and it gets ASSTHOUSAND favorites. ?! That was bizarre.
posted by grapefruitmoon at 5:58 AM on November 28, 2009


Oh, huh, so I go abroad and this is what I miss? I was unaware of any "experiment". All I know is that many details of the world seem different after experiencing Amsterdam.
posted by little e at 6:00 AM on November 28, 2009 [1 favorite]


Baffled from the beginning by the reasons for this experiment. Turned it off as soon as that was an option. I am not in the habit of wanting to know less about a thread that I'm interested in, assuming that knowing more doesn't use up screen real estate, etc.
posted by game warden to the events rhino at 6:09 AM on November 28, 2009


It seems to me that the most common complaints are either "You can't tell the difference between a couple favorites and a hundred" and "It made me/others favorite more/less". The former could be fixed by just having a three value indicator ("has lotsa favorites", which, incidentally, I think I'd prefer even over numbers), and the latter is a red herring as I posted in the old thread. I'm away from my computer today, i'll rerun the stats later.
posted by Plutor at 6:10 AM on November 28, 2009


I saw no reason for ya'll to go fucking around with my interface without giving me an option to kept it as it was.
posted by Brandon Blatcher at 6:15 AM on November 28, 2009 [1 favorite]


I like the idea that my favorites are my own bookmarks rather than little brownie points I'm doling out. I'd like to keep the experiment conditions as my own default if I can figure out Greasemonkey.
posted by gladly at 6:23 AM on November 28, 2009


I'm not sure how to tell exactly (is there a place to click that gives you favorite totals by month?), but my hazy guess is that I've been receiving fewer than usual. The discussion in that previous thread did, however, make me realize some of the value of the "contacts" list, so I've been making a conscious effort to add contacts when someone's posts have been consistently catching my eye. I'd be really curious if overall there was a change in not just favoriting, but also in contacting, across the site, in the last month.

I preferred either seeing the favorite count or not knowing if something had been favorited at all. The new way seemed like just enough info to make you wonder about the details, but not enough to satisfy.

This captures my feelings exactly, or perhaps "has favorites" with a mouseover option. I really hated the pseudo-information of "has favorites" (though that was better than the awful "faved"), but a pop-up/mouseover thing that gave the number would return a lot of what I felt was lost.
posted by Forktine at 6:30 AM on November 28, 2009


Can someone make a Greasemonkey script that will do the one-couple-some-many-tons favorites options? Or just one-some and I bet I could extrapolate from there. I am all for coarser-grained knowledge, but one category does tend to feel a bit too coarse.

Though honestly I don't think it has made a lick of difference for me, and I am amused by people who get worked up about these sorts of things.
posted by that girl at 6:31 AM on November 28, 2009


I've always used favorites as bookmarks, so that use hasn't changed for me. I liked the new way, for the reasons already mentioned above - it made the discussions seem less fighty because you couldn't see how many people were furiously clicking "yeah, take that!" If there were an option to remove favorite counts permanently, I'd use it.
posted by misskaz at 6:46 AM on November 28, 2009


I always thought that the more data you have, the better, so I voted against the experiment in the previous thread. But I stayed with it because, well, it's an experiment.

(By the way, how can all the people who have opted out can talk so much about it? You didn't run the experiment. You are the "control group". "Nothing changed and I was very happy"? Or "Nothing changed and I am still angry you even suggested it"? Congrats! Every experiment needs a control group and you did your duty beautifully. But writing so much about doing nothing is so bizarre that your reactions about doing nothing should be the topic of a stand alone study.)

Finally, the experiment changed my mind. I like the absence of numbers. It kind of smooths out the reading rhythm. It removes spikes of excitement that have nothing to do with the content. I now understand why it was tried and I think it works as intended: it's the difference between watching a movie in a theater with strangers and alone in your home.

One is a collective experience (you hear people laughing, sighing, screaming or yawning) the other is more akin to reading, a solitary experience. Both are legitimate and I think now that both should remain available: sometimes you feel like reading in silence, other times you enjoy to be part of a crowd.

And I agree with water bear that "has favorites" is not the best choice: Take out the word "has" and it would be perfect.
posted by bru at 6:47 AM on November 28, 2009 [5 favorites]


This has been an interesting month.

All the attention on favorites/favoriting caused me to change my MetaFilter experience. I was happy to try the experiment because I've never been too thrilled with the whole popularity contest aspect, but I actually gave more attention to favorites. This past month I often clicked on the favorites to see both numbers and names; prior to this month I had never bothered with names.

I also added contacts for the first time. Go team Gravy!

For me the biggest difference (beside clicking to see who actually favorited something) was I didn't get distracted while reading threads by favorite counts. When the numbers are displayed I often find myself being jerked up by seeing [Simpson quote] 35 favorites, followed by thoughtful remark not favorited at all. That's irritating to me, and I know it is MY problem not a site problem so I have to get over it. I realize that people are free to use their favorites any way they want to, but it still bugs me.
posted by Secret Life of Gravy at 6:51 AM on November 28, 2009


I tried it all month (with a Greasemonkey script at home and nothing at work), and I'm very ready for things to go back to the way they were. Favorite Counts are my laugh track, my Greek chorus, they let me know how people are feeling about things and I didn't realize how much I enjoyed that part of the community until it was gone.
posted by ThePinkSuperhero at 6:56 AM on November 28, 2009 [4 favorites]


I don't see why the end state can't be that it remains a preference.

[X] show favorites
[_] completely hide favorites

If this was the case, would there still be a lot of strong feelings about it? Is it really about other people seeing how many people agree with you?

I was in the "hide favorites" camp, and still am, with the possible exception of Ask. It seems to be enough of an outlet for some folks to suppress "me too" comments.

I was amazed by the strong reaction in the last thread.
posted by popechunk at 6:59 AM on November 28, 2009 [2 favorites]


I like to see the number of times something has been favorited. Viva the old way, death to the experiment!
posted by Daddy-O at 7:11 AM on November 28, 2009


I've stuck with the experiment (as a good comrade able to accept collective discipline) and have found it suits me fine, as in the more substantive discussion threads I enjoy the popularity or otherwise of an argument doesn't sway me, and of course I have my own standards for what's a funny joke.
The caveat to that would be threads like this where we're discussing community preferences, where obviously it is relevant to know what views are most strongly endorsed.
posted by Abiezer at 7:13 AM on November 28, 2009


(has favorites) annoyed me so I happily went back to old style and stayed there.
posted by adamvasco at 7:13 AM on November 28, 2009


Favorites have always bothered me from the time they were implemented, so I was happy with the experiment and will leave the favorite count turned off if it is an option. There are plenty of other sites with a scoreboard that I post to, but I mostly lurk on Metafilter and like to read it without that taint.

I do think that a month is not long enough to start seeing any noticeable difference in commenting behavior. I think any changes would be more subtle and evolve as the initial contentiousness of the decision to remove favorites faded over time.
posted by Arch_Stanton at 7:16 AM on November 28, 2009


Viva the old way, death to the experiment!

Damn straight — down with favorites!

Oh, you meant the new old way.
posted by enn at 7:25 AM on November 28, 2009 [4 favorites]


When the numbers are displayed I often find myself being jerked up by seeing [Simpson quote] 35 favorites, followed by thoughtful remark not favorited at all.

Really? This never bothered me. I've always just favorited whatever thoughtful remarks were lurking around; I've also used contacts on the site since the beginning of my membership so it strikes me as a bit quaint that some are just discovering this functionality.

As for my usage, I turned it off from the moment I discovered I could. However, when I log on on a new computer/device (say, my husband's iPhone), I tended to be too lazy to go back to settings and reload. I favorite pretty liberally as-is, but it sure feels like I favorite just about everything under the "has favorites" system. I suspect this is due to being forced to read more comments in a situation where I would normally skim.
posted by PhoBWanKenobi at 7:47 AM on November 28, 2009


Sometimes there are really really really long comments. I mean really long. I sometimes skip over the really really long comments - perhaps I lack the patience and time to read a page-long comment.

However! I just noticed that I take into account the favorite count. If it has a decent number of favorites, I kinda assume that taking the time to read that novela may just be worth my time.
posted by Sassyfras at 7:48 AM on November 28, 2009


Last month I saw a thing quite rare -
A favorite that wasn’t there
It wasn’t there again today
Oh, how I wish it’d go away.

(That said, I prefer the old way, xx)
posted by prefpara at 7:52 AM on November 28, 2009


That said, I prefer the old way, xx
No, listen to TV Smith: the New Ways are Best.
posted by Abiezer at 8:02 AM on November 28, 2009


Favorites have always bothered me from the time they were implemented, so I was happy with the experiment and will leave the favorite count turned off if it is an option. There are plenty of other sites with a scoreboard that I post to, but I mostly lurk on Metafilter and like to read it without that taint.

+1

see what I did there?
posted by Joseph Gurl at 8:03 AM on November 28, 2009 [2 favorites]


Sometimes there are really really really long comments. I mean really long. I sometimes skip over the really really long comments - perhaps I lack the patience and time to read a page-long comment.

However! I just noticed that I take into account the favorite count. If it has a decent number of favorites, I kinda assume that taking the time to read that novela may just be worth my time.


Seriously? A page is an onerous burden? Maybe that's a bigger problem here than favorites. I may be tilting at the wrong windmill.
posted by Joseph Gurl at 8:05 AM on November 28, 2009 [1 favorite]


I stuck with it for most of the month and it's been fine. Got used to it much faster than I thought I would. I did turn it off for a couple of days when the Watcha reading? meTa was busiest, because I wanted to see if my experience of reading a thread like that was different with and without visible favorites. I learned that I use favorites in ways I didn't know I used them. I don't think I got or gave more or fewer favorites with the count turned off.
posted by rtha at 8:08 AM on November 28, 2009


I think a good compromise would be setting the default to show favorites but still have an option in our preferences to opt out of favorites.
posted by pwally at 8:25 AM on November 28, 2009 [2 favorites]


My guess is we'll do something like that, pwally. I've been enjoying the "has favorites" view of the site personally, I can see why it wouldn't be for everyone.
posted by jessamyn (staff) at 8:26 AM on November 28, 2009 [4 favorites]


Maybe one of the MeFi statisticians can look at a few things that I have observed with respect to favorites. The first is that earlier comments in any given thread seem to garner more favorites. That might be called tl;dr syndrome. The second is that longer comments, even if they are mostly gibberish, seem to get more favorites. It's like a pound of commentary is better than a few ounces.
posted by fixedgear at 8:36 AM on November 28, 2009


I prefer the "has favorites" view of Metafilter but I don't think it's worth messing with the site's default. I don't think it improves discussion that much, but I do think that it improves one's reading of the site in the more contentious threads. An opt-in option would be nice.
posted by anifinder at 8:36 AM on November 28, 2009


All I know is that many details of the world seem different after experiencing Amsterdam.

You have missed nothing, as this was also a part of the experiment. Thanks for participating.
posted by Meatbomb at 8:42 AM on November 28, 2009


I got used to it after a week or so. I don't think it did much to curb my own favouriting, as I've always just stuck to favouriting what I like, instead of what's popular, but I can see how it may have affected the habits of others. I would prefer to drop the 'has' though. I find it reads awkwardly.

I definitely wouldn't mind if this experiment became permanent, but I can't really see the benefit or drawback of either.

For those of you who want to go back to the old ways, I think you're just being contrary and inflexible. What's wrong with a new system? It doesn't hurt, and it might just 'help' in a way that I just don't fully understand.
posted by sunshinesky at 8:44 AM on November 28, 2009


I didn't find a comment I liked better than the one that's already my favorite, so nothing changed for me.
posted by mr_crash_davis mark II: Jazz Odyssey at 8:54 AM on November 28, 2009 [3 favorites]


The only thought I had was to suggest that if you don't want to turn it back on as it was, why not have the favourite count appear when you rollover [has favourites]

Apart from that I dont really mind one way or the other but if pushed I'd got back to the old way
posted by criticalbill at 8:56 AM on November 28, 2009


I liked it, as it may me more judicious with dispensing favorites and less obscenely fascinated with my own fave counts.

Keep it +1
posted by Potomac Avenue at 9:03 AM on November 28, 2009


I think you're just being contrary and inflexible

Perhaps those people just know what they like? Perhaps they just prefer to have favourites visible? Perhaps you're just being offensive to other people who have a different opinion to you? New doesn't always equal better.

I disabled the faved thing as soon as I figured out how. I also later installed the Greasemonkey script, and now have a lovely visual guide to what is worth checking out.

My problem wasn't that it didn't say [x favourites] any more. It could have said [x bananas] for all I care. The problem that I had was that there was no simple way to scroll through a thread, avoid the noise and get to the gold pretty quickly. If faved (or bananas, for that matter) was reinstated, I wouldn't mind, as long as there was still some way to see what was worth looking at, and at a deeper level, what was more/less worth looking at within the umbrella of "worth looking at". I think the word "favourites" is skewing the discussion, slightly. Favourites has a different connotation to bananas.

The greasemonkey script I have is invaluable for that. It's immediately obvious when entering the thread what the user base finds worthy of noting. Usually, I find that that's a pretty good guide to something being worth reading. I'd hate to lose that. I only found the script after enabling the favourites count visibility again. I don't know if it works with faved or not.
posted by Solomon at 9:06 AM on November 28, 2009


My guess is we'll do something like that, pwally.

Here is a vote for making the "don't show favorites" option completely obscure the favorite status of all comments, rather than displaying what I thought to be the compromise "has favorites".
posted by popechunk at 9:08 AM on November 28, 2009 [1 favorite]


My experience? I received a food pellet from my disk drive when I gave a favorite, and a mild shock from my mouse when I received one. I find as a result that I've gained a few pounds and am making fewer references to Hitler.
posted by found missing at 9:11 AM on November 28, 2009 [4 favorites]


Yes, sometimes a page-long comment is too long. Not all the time . . . but occasionally - when my time is limited . . . and those times I appreciate seeing that a long comment got 59 favorites. Kinda clues me in that perhaps that particular comment is on to something.
posted by Sassyfras at 9:12 AM on November 28, 2009


In the first few days of the experiment, I went back and forth between the two preference options. I thought I hated the "has favorites" approch at first, but I stuck with it to give the experiment a fair shot. I got over my discomfort quickly. I've been surprised to find that, really, I don't miss the "x favorites" approach.

I still favorite things, but I do it much the way I did before, and I frequently go through my favorites to cull the ones I don't need/like/want to support any longer. I'm also mostly unemployed right now, so I have a lot of time to do this kind of thing. I don't know what my behavior would be like if I were working a 9-5.
posted by runningwithscissors at 9:16 AM on November 28, 2009


fixedgear: "Maybe one of the MeFi statisticians can look at a few things that I have observed with respect to favorites. The first is that earlier comments in any given thread seem to garner more favorites. That might be called tl;dr syndrome. The second is that longer comments, even if they are mostly gibberish, seem to get more favorites. It's like a pound of commentary is better than a few ounces"

I have a bunch of datawankery stuff in mind for when the experiment is officially over, including something like the first part of your comment. I'm interested to see if the timing of comments and favorites changed during the experiment, among other things. Analysis of comment length vs. favorites would be interesting also, but the comment length data isn't in the Infodump for us to analyze.

I think, though, that the experiment isn't primarily about numbers and statistics, but about the quality of comments. I think posting a bunch of numbers and statistics here in this thread is going to derail it from what it should be about, or at least is going to make it even longer than I'm sure it's already going to be. The last Infodump thread I participated in exceeded 500 comments, just to illustrate the potential problem.

So I'd like to see a separate thread for stats and numerical analysis, and I'll volunteer to create that thread on Dec 1st. Then us data nerds can have fun with our stats, and if there is anything relevant that comes up with respect to the more important discussions over here, we can just link to it as appropriate.
posted by FishBike at 9:27 AM on November 28, 2009


I propose that for December, we no longer mark posts which have been favorited, but instead mark posts with the total number of people which have viewed it but have not favorited it.

for example:

posted by empath at 12:18 PM on November 28 [not favorited by 289 readers +] [!]
posted by empath at 9:28 AM on November 28, 2009 [3 favorites]


As an experiment the only thing it tested for was how many people would use a setting. With regard to comment quality/tone it tested nothing because members can and did opt-out. Any attempt to draw a conclusion about sitewide behavior is meaningless. Even if the experiment had been mandatory any conclusions drawn would have been member specific and subject to confirmation bias. (Overused term but seems appropriate in this context.) The resulting debate would have been an epic shitstorm with (more) useless bad feeling all around. It was worth a shot though.
That said I think options that affect only the personal preferences of each member are great.

Turkey soup with tarragon dumplings is going to be nommy.
posted by vapidave at 9:30 AM on November 28, 2009


Several times I found that a comment that had little to no redeeming qualities (didn't really say anything, was odd and didn't make sense, a complete tangent, was just stupid jokey in a context where that wasn't welcome) had favourites, and I was surprised. When I checked who had favourited it, it invariably turned out to be the writer of the comment. This bothered me. That someone who writes something amazing - some personal history of an amazing or tragic event/encounter and gets forty favorites, and someone who writes "Balls. Amirite?" and favourites it themselves, both get the exact same recognition. I wondered if the incidence of self-favourites had increased because of the experiement.

Ultimately, I turned it off on my home computer and left it on on my work computer, but just really found the lack of favourites count to be annoying. What is the point of knowing if something has favourites, if I can't easily know how many. If faves as votes is a problem (which I am not necessarily agreeing is so) then get rid of favourites entirely. All in or all out - this is one situation where a compromise loses the important elements of both sides and just keeps the irrelevant or annoying elements.
posted by arcticwoman at 9:33 AM on November 28, 2009 [2 favorites]


I think you're just being contrary and inflexible

You speak as if those are bad qualities in this situation. They're.

If you want to be part of an experiment, knock yourself out. I have no interest in doing so, particularly when there wasn't a way for me to opt out. Don't shovel crap at me and say I have to eat for a month.

The above language may sound harsh, but the choice of is deliberate and intentional, while no person animosity is meant towards the mods. There are times when the weight of the community is more important than any individual opinion and every person can't have everything go their special snowflake every single time. Still, it's important, as member of the community, to express your opinion when you believe it's important.

The more important question about this exercise is the whether to deal with such changes in the future. Should the mods should just make a change (after discussing it heavily amongst themselves) and see what happens? Should they implement a preference to turn off the change to those who don't like or want it? Will doing so fracture the community, perhaps in tiny ways that build over time? Has the size of Metafilter grown to the point where change is more and less impossible with the interface? Does the "backlash" even matter that much?
posted by Brandon Blatcher at 9:34 AM on November 28, 2009 [1 favorite]


I don't mind it at all -- if i'm interested enough in something, I can still see who favourited it, but I don't get that subconscious "this is Correct because so many people have faved it" feeling that I did before.
posted by ukdanae at 9:35 AM on November 28, 2009 [1 favorite]


I think you're just being contrary and inflexible

A useful feature that contributes mightily to my use and enjoyment of this site was nerfed in this experiment. I would be "contrary and inflexible" about the removal of my car's speedometer as well.

What's wrong with a new system?

It's crap. (please see above) It was a change that didn't need to happen which reduced the ease of using Metafilter. It was a downgrade that derived absolutely zero benefit in terms of discussion quality (go reread the TSA STOLE MY BABY thread if you don't believe me).

I propose that future MeTa threads that blame favorites for some imagined "reduction in MeFi conversation quality be immediately closed after links are posted to this thread, the original experiment thread and whatever greasemonkey script koeslitz cooks up to hide favorites completely from those who just can't stand'em.
posted by EatTheWeek at 9:37 AM on November 28, 2009 [5 favorites]


It sucked. I switched back to normal favorite counts as soon as I could because they're obviously superior and without need of examination, but my multifavorited greasemonkey script was still broken. This script highlights comments with more than seven favorites and is a necessity for making Mefi readable, especially in long threads. The visual pattern formed by favorites gives you a point of reference when scrolling through longboats. Without it, the past month has felt like reading with one eye missing.
posted by bunnytricks at 9:48 AM on November 28, 2009


Turned off the experiment on day one. I like seeing the numbers.
posted by Babblesort at 10:02 AM on November 28, 2009


All experiments, good and bad, come to an end. Thankfully this silliness will be over soon.
posted by Blazecock Pileon at 10:06 AM on November 28, 2009


I lasted three weeks in the experiment, and ended up turning it off around the 21st. I found it substantially less useful than the old method. My normal method of reading threads is to just read them, but sometimes I'll stumble into one I missed via the Popular links or what have you. If I don't feel like reading the whole thing, I'll search for 'favorites' to pick out comments that have more than one, and read those; this may or may not convince me to read everything. So that function was gone.

In regular reading, I'll often see interesting comments, and if I'm curious about how many favorites they got, the 'has favorites' thing doesn't stop me, I just click on 'has favorites' and look at the count. I get the same information either way, it's just less convenient, and increases your traffic a little.

If there's one lesson I've drawn from this, it's that favorites are indeed used by an awful lot of people, including me, to 'keep score'. If you give people something that's called "favorites" and accumulates in their profiles, it will be used as a score, period, no matter how frothy some people get about scorekeeping.

If you don't like that idea, this method isn't going to fix anything. It's just inconvenient. Either turn favorites into 'bookmarks', and make them visible only to the person who issued them, or get rid of them altogether. This halfway approach just causes damage without benefit. It's not fixing anything, it's just putting on a thin veil and pretending very hard.

If they do go away entirely, I won't care that much.
posted by Malor at 10:12 AM on November 28, 2009


Brandon Blatcher writes "I saw no reason for ya'll to go fucking around with my interface without giving me an option to kept it as it was."

Exactly. We should remove the display of favourite counts, which were put in place by fiat, until it's been well debated with the pros and cons outlined to everyone and then maybe make it opt in rather than the default.
posted by Mitheral at 10:14 AM on November 28, 2009 [1 favorite]


arcticwoman writes "Several times I found that a comment that had little to no redeeming qualities (didn't really say anything, was odd and didn't make sense, a complete tangent, was just stupid jokey in a context where that wasn't welcome) had favourites, and I was surprised. When I checked who had favourited it, it invariably turned out to be the writer of the comment."

This thread will be open for the month. I encourage anyone who finds one of these comments to link it here. I'd say I haven't seen it but then I never check who favourited stuff for the most part. It just seems like it's a problem that lends itself to confirmation bias.
posted by Mitheral at 10:17 AM on November 28, 2009


Following empath's joking suggestion we could replace "has favorites" with "stats". "Stats" would list any data that could become available over time: number of views, time spent, favorites, flagged as fantastic, number of of members commenting, number of your contacts commenting, money generated through contextual ads, number of tweets generated, number of retweets, exclusive to MetaFilter, also appearing in X other sites, number of snark attacks, index of overall likability generated by a secret cortex' formula. For starters.
posted by bru at 10:18 AM on November 28, 2009 [3 favorites]


I tired it for a week, didn't like it and switched back. It just makes the page look wrong and I started noticing the "has favorites" more and more until it made the site unreadable.
posted by The Whelk at 10:39 AM on November 28, 2009 [1 favorite]


This script highlights comments with more than seven favorites and is a necessity for making Mefi readable, especially in long threads.

A necessity? Huh. This is not my experience.
posted by rtha at 10:40 AM on November 28, 2009 [1 favorite]


lalex writes "There was a whole thread about it."

Where, IIRC, it was determined that it was a single user doing it with any sort of regularity. quonsar is tenth on the list and he only has 18 self favourites. And he's been banned since before the grand experiment. Hardly an epidemic requiring a policy change. tehloki is probably responsible for at least a couple order of magnitudes single favourite comments.
posted by Mitheral at 10:48 AM on November 28, 2009


Where, IIRC, it was determined that it was a single user doing it with any sort of regularity.

Not for nothing mitheral, but that list you linked to is from January. Of 2008. Two-year-old data isn't really very useful in this context.
posted by dersins at 11:01 AM on November 28, 2009 [1 favorite]


Didn't know that. So self favouriting behaviour is one other thing that should be crunched at the end of the month.
posted by Mitheral at 11:12 AM on November 28, 2009


We should remove the display of favourite counts, which were put in place by fiat, until it's been well debated with the pros and cons outlined to everyone and then maybe make it opt in rather than the default.

There's a big difference between adding a feature and taking away.

As to the great debate, it's already been done several times over the years, so we can just go ahead and skip to the opt in portion and be done with.

Alright, who wants mince meat pie?!
posted by Brandon Blatcher at 11:13 AM on November 28, 2009


I liked the bookmarklet, so I could by default have no count & yet pull up numbers when I wanted. It's been removed from its creator's page, but I still have a copy. Er. If it's helpful in the next couple of days.
posted by Pronoiac at 11:28 AM on November 28, 2009


Really quick count: as of this moment, there have been exactly* 6,528 self-favorites assigned since the feature was launched.

There are about 2.5 million favorites in the system. So self-favoriting accounts for about 0.25% of all favoriting activity, which, if we convert that value to Yiddish Metric, comes out to approximately bupkis.

I'll see if I can put together some slightly more interesting views of this data, but, yeah. It ain't much.

*Well, actually, this may not be true. What we can see is how many are currently assigned, but we keep no specific record of retracted favorites. So it stands to reason that there have probably been some number of additional self-favorites assigned and subsequently retracted. But I'm going to put on my Assumption Boots and declare that to be a likely very small sum that doesn't really affect the situation in any case.
posted by cortex (staff) at 11:34 AM on November 28, 2009 [4 favorites]


i didn't care for it and opted out - people's commenting behavior didn't seem any different that i could see
posted by pyramid termite


That pretty much sums up my experience.
posted by marxchivist at 11:35 AM on November 28, 2009


Not to derail the thread as it currently stands, but I just want to throw my two cents in. I'd been visiting MeFi long, long before I ever signed up and so I got used to the "# favorites" but whether or not the counts or shown, whatever the how or the why, the favorites are extremely helpful. It took some getting used to, but "has favorites" has grown on me as well. If I really want to know how many other people favorited a comment, it isn't that difficult to click on that link and see who else has. That comes with one benefit - it helps me find like-minded MeFites and posts I might've missed that I'll be interested in.

Alright, who wants mince meat pie?!

Got a nice sweet fruity one?
posted by neewom at 11:35 AM on November 28, 2009


So self-favoriting accounts for about 0.25% of all favoriting activity

How does that change if you look at only the last 27 days though?
posted by jessamyn (staff) at 11:38 AM on November 28, 2009


It was a downgrade that derived absolutely zero benefit in terms of discussion quality (go reread the TSA STOLE MY BABY thread if you don't believe me).

I don't understand what you're saying here. The TSA post was made on October 16th. The favorites change was implemented on October 31st. Out of the 653 comments in the TSA thread, 650 were posted before October 31st. What will rereading that thread tell me about favorites and discussion quality?
posted by enn at 11:39 AM on November 28, 2009 [1 favorite]


When the numbers are displayed I often find myself being jerked up by seeing [Simpson quote] 35 favorites, followed by thoughtful remark not favorited at all.

Really? This never bothered me. I've always just favorited whatever thoughtful remarks were lurking around;

Here is why it bothers me. First, because it makes me feel out of place. I have no great love for quotes from the Simpsons or Family Man or other pop culture quotes, I think they are lazy and cheap and stand in for original thought. Second, and far more important, we have had people say they study what gets favorited so they know how to comment. What message does a highly favorited Simpsons quote give? That we need more people to pull Simpsons quotes out of their ass? In fact over and over we have heard variations of this: "The problem that I had was that there was no simple way to scroll through a thread, avoid the noise and get to the gold pretty quickly." So quotes from TV shows are "the gold" and original thoughts are the noise. Great.

Ultimately, we cannot make people use favorites in any specific way, even though many members have confessed that they think of multiple favorited comments "the gold."
posted by Secret Life of Gravy at 11:42 AM on November 28, 2009 [3 favorites]


How does that change if you look at only the last 27 days though?

Dunno yet!
posted by cortex (staff) at 11:44 AM on November 28, 2009


Just another vote for "changed it back as soon as that was an option."

Let's be clear that the real issue remains: Everyone uses the favorites feature, and thus, the favorites count, in a different way. Some people use it to quickly identify posts of some merit, others ignore the count and use them to bookmark posts for later, and others ignore them entirely.

Removing functionality that has become so ingrained in the way we use something will just anger, but offering the option to hide the counts seems silly, since the change only "works" (in that it changes people's behavior as a whole) if it's no longer an option. Put another way, I don't really know why someone would opt to turn off the count, but since the code has already been written, might as well leave it in for those self-righteous of us who wish not to play a masturbatory numbers game with their perusal of comments. Or something.
posted by disillusioned at 11:45 AM on November 28, 2009 [1 favorite]


My data point: I left the [has favorites +] feature on for the entire month, and found that inevitably, on comments I was interested in favoriting that already had been favorited by others, I would click through to see who/how many people had favorited them. It didn't stop me from taking an interest in who/how many people had favorited something, it just put an intermediary step between me and the favorites.

What I wonder is how I would've reacted if I'd been unable to see even that a post or comment had favorites.
posted by limeonaire at 11:47 AM on November 28, 2009 [1 favorite]


I left it running. I also found I favorited fewer comments because I tend to use it as a "yes, I agree; take this advice" button. I have a few things favorited so I can find them again (or to indicate that the comment made me laugh), but I use it as a me-too button more than anything else, which I was less inclined to do when I couldn't see counts.

Generally, I don't look at a "has favorites" marker on a comment or post differently from "been favorited X times"
posted by crush-onastick at 11:54 AM on November 28, 2009


Calling it an experiment changed my view of what science is. Or maybe it was the FPP How to Think About Science.
posted by Obscure Reference at 11:57 AM on November 28, 2009 [1 favorite]


I'm sure cortex has the good data but looking at the infodump for November 1–November 22 (the latest available), 166 of 111456 favorites were self-favorites.
posted by enn at 12:05 PM on November 28, 2009


I like it. I don't care how many favorites something has unless I'm the one who wrote it.
posted by bingo at 12:10 PM on November 28, 2009 [1 favorite]


I would like to retain the ability to have the counts off, and would also prefer to have visible favorite counts turned off altogether for this option - but retaining the ability to click through to the list of favoriters using some fixed interface element would be useful.

I personally have not noticed any difference in the quality of discussions but I do prefer reading MeFi without the sense of every comment being accompanied by laughter, applause or silence. I have favorited more comments; I think because I'm not subconsciously thinking "I agree with this, but it has plenty of favorites already" or "this seems like a good point but it doesn't have much support, maybe I'm missing something?".

I've clicked through to the list quite a lot and been surprised at the number of times that I've only been the second person to favorite a comment. I would guess that if favorite counts were off altogether I might have been the first person to add a favorite more often.
posted by tomcooke at 12:12 PM on November 28, 2009


So no real surprise there, enn. Thanks for giving it a quick peek.

I'm going to go ahead and rerun the dump again ahead of the normal weekly schedule, and I'll run it again once the 1st rolls around as well. Anybody champing at the bit, feel free to snag a new dump once the index shows its updated. I'd give it ten or fifteen minutes here.
posted by cortex (staff) at 12:14 PM on November 28, 2009


I didn't see any noteworthy difference in how the site read, that I'm at all confident about noticing at least. No difference in how discussions flowed and gurgled and refluxed, no great decline or improvement in quality, etc.

I have a second-guessing-myself doubting notion that there were fewer sidebarred items during November than previously. It's a low confidence interval, on that, though. If that's right, maybe it's due to a somewhat lesser degree of piles-o'-favorites fantastic comments attracting fewer fantastic-flags. But it could also be due to completely unrelated factors. Or just not be a real pattern at all.
posted by Drastic at 12:14 PM on November 28, 2009


"We should remove the display of favourite counts, which were put in place by fiat, until it's been well debated with the pros and cons outlined to everyone and then maybe make it opt in rather than the default."

Wait, wait. I want to play.

We should remove the display of usernames, which were put into place by fiat, until it's been well debated with the pos and cons outlined to everyone and then maybe make it optin in rather than the default.


We should remove the display of the date posted, which were put into place by fiat, until it's been well debated with the pos and cons outlined to everyone and then maybe make it optin in rather than the default.


We should remove the display of live links, which were put into place by fiat, until it's been well debated with the pos and cons outlined to everyone and then maybe make it optin in rather than the default.


This is fun! When do I get my pony?
posted by mr_crash_davis mark II: Jazz Odyssey at 12:15 PM on November 28, 2009 [7 favorites]


I noticed the change in favorites, but thought little of it. Doesn't seem like a serious affair to begin with.
posted by attackthetaxi at 12:21 PM on November 28, 2009


I joined during November, .... but MeFi is a community as well as a collection of interesting links, and part of that is that people will like/agree with/vote for/bookmark things that they find good/agreeable/interesting. That's information that I like to see, personally, so put me down for "prefer having the numbers".

Hate 'em or Love 'em, favorites do enhance my feeling of community here; that sense that I'm not just sharing the same basic information as a bunch of (mostly) strangers, I'm also sharing in the ongoing feedback (and counter-feedback) to that information.

I dutifully went with the "experiment" up until a few days ago when I stumbled upon a particularly complicated thread (it was the Rushkoff thing) at a time when I simply didn't have the time to read everything in it. Not wanting to be just random in my skim-through, I adjusted my preferences and dove back in, quickly picked up what seemed to be the most "popular" talking-points ... and so on. This worked for me, even got me involved in the discussion in a fairly meaningful way.

I've had "favorites-on" ever since.
posted by philip-random at 12:26 PM on November 28, 2009


Calling it an experiment changed my view of what science is.

I know you're probably taking the piss here, but this sort of statement just makes me frustrated and sad.

We made a post about this and explained what we were doing and why. We did not say we were approaching this scientifically, just that cortex would do his usual number-crunching after the month was up. I would be hard to keep him from doing it, since it's what he does anyhow. We explained, probably a dozen times maybe more, that we were not attempting to do any rigorous hypothesis testing thing but we wanted to sort of experience the site without favorites and see if it felt different, how people perceived it, and what sort of changes might have happened as a result. Then the plan was to talk about it, with as much of the entire site as we could, and try to listen to people, see how we felt about stuff moving forward, and just try to examine what happened and see if any changes needed to be made in the long run.

We got called stupid, uninformed, disrespectful, cavalier and a bunch of other things. We were basically accused of practicing science without a license, and for a dumb reason and in a dumb way. By some people. When we changed some things to make it easier for people who were having such strong reactions to change it back, we were told that this made it even LESS scientific and even MORE stupid and even more laughable and even less worth paying attention to.

Well fuck it. You know what? It's not science and it wasn't supposed to be. Running this site isn't a "by the numbers" game and isn't supposed to be. If it was, you could plug in robots instead of cortex and mathowie and pb and vacapinta and me and they'd just delete stuff when it got to a certain threshhold and never answer your email when you said that something made you feel bad or something made you happy. Robots tend to not donate money to charity. Robots don't give a shit about you. And robots don't get their feelings hurt.

I'm probably just having some sort of fun-holiday hangover at this point, but I'd love it if the "experiment means one thing only and ur doin it rong" canard could go away. Given that we're running a site with about 10,000 active members in any weeklong period, if we want to change anything at all, we have to sort of forge ahead and try things because waiting for consensus is actually never going to happen. I'm sorry if November was unpleasant for you.
posted by jessamyn (staff) at 12:28 PM on November 28, 2009 [61 favorites]


I'm in the "get rid of them entirely or put them back the way they were because this in-between stuff doesn't mean squat" camp.

I haven't been giving out as many favorites. The [+] feels a little broken to me now and I stay away from it subconsciously.

Despite that, I still cravenly cherish the favorites I get, but I don't think it influences my comments, which were of such poor quality to begin with.

Also in the spirit of chillness and respect for the mods maybe people can post their thoughts once - maybe twice - and keep this thread from becoming a giant back-and-forth argument.

Good advice for any thread. I so despise the windbag style of argument, where one "wins" by ignoring criticism and reiterating one's views over and over and over, or by writing long posts so dense with framing and little logic errors that no one has the energy to refute them. Or yeah, the "NO NO NO NO" shit, too.
posted by fleacircus at 12:29 PM on November 28, 2009 [1 favorite]


MetaTalk: vehement, rhetorically charged reactions.
posted by UbuRoivas at 12:31 PM on November 28, 2009


I advocate changes in the site without any input whatsoever.
posted by attackthetaxi at 12:35 PM on November 28, 2009


put in place by fiat
posted by Abiezer at 12:37 PM on November 28, 2009 [1 favorite]


Robots don't give a shit about you.

I think this needed to be said.
posted by found missing at 12:39 PM on November 28, 2009


I gave it a try, but gave up and changed it in preferences after a couple of weeks.

It essentially negated favorites ("has favorites" just wasn't useful), and I missed this, particularly in contentious threads on the blue, where favoriting often seems to work as a little nod of solidarity on an issue.

I didn't miss it as much in Ask, where favorites are often used kinda the same way, but as a more personal judgment (we're discussing individual MeFites rather than a link.)
posted by desuetude at 12:44 PM on November 28, 2009


Fix It Again Tony.
posted by fixedgear at 12:45 PM on November 28, 2009


I'm fine with "has favorites"; given the option to keep it that way, I probably would. I prefer the non-pile-on appearance of a less visible favorite count. I tend to read entire threads and not skim, so favorites were never useful to me in that way. I've given out favorites to the posts and comments I like, and that hasn't changed.

I'm not sure I could say I saw any difference in the way other people behaved, except in the MeTa threads about favorites where a lot of people I normally think of as even tempered and thoughtful seemed to get extremely ruffled about a temporary change. That was a bummer.
posted by oneirodynia at 12:46 PM on November 28, 2009


I switched back to the regular-you-can-see-the-number view pretty quick. When I browse MeFi from my iphone, though, it shows "has favorites." Even though I'm logged in and I have the old view enabled in my preferences, but whatever. Anyway, when I'm on my phone, I can't stop wondering, how many favorites? How many favorites?

Just my data point.
posted by Rinku at 12:59 PM on November 28, 2009 [1 favorite]


The infodump is updated.

I count 164 self-favorites this month, not counting 40 in the thread about self-favoriting.

Skimming, mostly it's one-offs, except eccnineten's 11.
posted by Pronoiac at 1:11 PM on November 28, 2009


I had my favourites off all month; I admit I do favourite things occasionally, but overall I'd be happier if Metafilter were favourite-free. I confess I sometimes clicked through to see how many favourites a particular comment had, and my instinct for heavily-favourited comments was usually right on. Which is kind of depressing actually.

I hope there will be an option to opt out of favourites after this is over. If not, I'll probably be staying away from Metafilter as much as I can manage to. The whole culture around favourites really drives me up the wall, and much of the pro-favourite content in these Meta threads have only added to my irritation.

I applaud the mods for giving this a go.
posted by stinkycheese at 1:12 PM on November 28, 2009 [3 favorites]


Jessamyn, I was one of those in the other thread who didn't see much that was demonstrable in an experiment with vague guidelines and I said so. I apologize to you and Cortex if it sounded like I was calling you stupid. That was not my intent and it isn't how I think.

Anyway, I stayed with the experiment for the month and I haven't really noticed much of a change in the tone of the site nor how I use it. But, to be fair I haven't spent a lot of time in the big-controversy threads. So, I suppose my data point would be with the 'favorites could stay or go' crowd. Solon and Thanks expressed it better here.
posted by kuujjuarapik at 1:16 PM on November 28, 2009 [1 favorite]


In more detail, people who favorited themselves more than once this month:
      2 Faze
      2 np312
      2 lahersedor
      2 jchaw
      2 vbfg
      2 Pineapplicious
      2 Greynaab
      2 hagelslaag
      3 eyeballkid
      4 ebesan
     11 eccnineten
posted by Pronoiac at 1:17 PM on November 28, 2009


interesting -- eccnineten favorites all his own contributions.
posted by empath at 1:20 PM on November 28, 2009


Metafilter: I'm probably just having some sort of fun-holiday hangover at this point
posted by mr_crash_davis mark II: Jazz Odyssey at 1:22 PM on November 28, 2009


enn - oh damn, that's embarrassing. I'm a little consumed with school right now, so it's not surprising that my memory of MeFi is so jacked up that I would place a thread in wrong month. Whoops.

Well, in that case, I withdraw that particular example, but not my main point. The general tenor of conversation here has not changed under invisifavorites. Furthermore, I was never of the mind that the general tenor of discussion here needed to change. The next time anyone feels MeFi's getting too surly, I would encourage them to examine the comment threads on their local papers or Youtube or pretty much anywhere else.

Metafilter was not broken before November. There's nothing to fix.
posted by EatTheWeek at 1:25 PM on November 28, 2009 [3 favorites]


I don't use favorites at all. I stayed with the November experiment, because, well, I don't use favorites. I do find the "has favorites" vaguely anxiety-producing (I'd rather see the number, maybe on mouseover), but not, really, enough to care. "Has favorites" is way, way better than "faved", though.
posted by leahwrenn at 1:33 PM on November 28, 2009


I opted out pretty quickly because of the visual annoyance, and also, for me the "has favorites" designator is a lot harder to tune out than the number. Putting "has favorites but I'm not going to tell you how many so don't ask" after nearly every comment draws way too much attention to the topic, like saying "WHATEVER YOU DO, PLEASE DON'T THINK ABOUT PTERODACTYLS."

Either display favorite counts or completely eliminate favorites and use a See Your Own Bookmarks Only bookmarking system instead as Malor suggested.
posted by FelliniBlank at 1:36 PM on November 28, 2009 [3 favorites]


A bit more info: that's 164 self-favorites out of 137187 137147 for the month, so 0.1195%.

And eccnineten was the catalyst for the self-favoriting thread.
posted by Pronoiac at 1:42 PM on November 28, 2009


closer to 0.1196%
posted by found missing at 1:44 PM on November 28, 2009 [1 favorite]


We were basically accused of practicing science without a license, and for a dumb reason and in a dumb way. By some people. When we changed some things to make it easier for people who were having such strong reactions to change it back, we were told that this made it even LESS scientific and even MORE stupid and even more laughable and even less worth paying attention to.

and of course that just means that people were being assholes and didn't have any kind of point at all when they made criticisms
posted by pyramid termite at 1:47 PM on November 28, 2009


Tried it. Didn't like it.
posted by ob at 1:50 PM on November 28, 2009 [1 favorite]


Whoops, yeah, found missing. (I redid the calculation after removing the self-favoriting thread self-favorites from the total, which rounds down despite having the same leading digits.)
posted by Pronoiac at 1:51 PM on November 28, 2009


and of course that just means that people were being assholes and didn't have any kind of point at all when they made criticisms

Didn't say that. Didn't think that. Who are you even talking to?
posted by jessamyn (staff) at 1:53 PM on November 28, 2009


Didn't say that. Didn't think that.

then why the unfair characterization of those who thought this was a bad experiment? why tell them "fuck it"? why not admit they had a point instead of being so defensive about it?
posted by pyramid termite at 2:07 PM on November 28, 2009 [1 favorite]


Nothing was broke before, after, or during November. There was never an implementation of a fix. The only people who are saying this are the people who are asserting it.
posted by P.o.B. at 2:17 PM on November 28, 2009


why not admit they had a point instead of being so defensive about it?

We've maintained, from the beginning, that this wasn't science. People's insistence that yes it is science and it's bad science is tiresome.

We have been, since the get-go, totally acknowledging that yeah people will be unhappy and we'd like them to bear with us for a few weeks. There's a lot of space between admitting people have a point -- which we've done, over and over and over -- and totally ceding what we were trying to do in the first place and bringing back favorite counts for everyone ASAP because some people were vocally unhappy about it.

Some people act that because we didn't totally roll over and say "okay we're putting it back the way it was, the experiment has failed!" that we're in some way not listening or not responsive to people and that's what's getting me. It's clear to me, clearer than before, that there is no consensus point on favorites. People feel very strongly in both (or all) directions. As someone whose job is trying to keep as many people satisfied with how the site runs as possible, that creates a quandary.

We'd really like to get to a point where we're not saying "well let's stick with what we've got because at least people are used to that" or "making changes is too hard because some people are assholes about it" but in the interests of transparency, some of the meta-issues of this month include the fact that people got so angry about this that they made the site a less great place to come to work in the morning. That's neither here nor there as far as this favorites discussion goes, but I think it will probably inform future "let's think about changing something" discussions. And maybe that's fine with people, but it doesn't totally sit right with me.
posted by jessamyn (staff) at 2:20 PM on November 28, 2009 [7 favorites]


I turned it off as soon as I had the option. The threads these days are just too big to read without some means of prioritizing, ranking or threading.

It was supposed to help with a problem of people posting short, glib, favorite-chasing comments. So if this is bugging people, maybe we should think of other ways of solving that problem.

Now one of the perpetual themes in Ask Metafilter answers seems to be: "If you have a problem with people's behaviour, your first step should be to ask them politely to stop it, rather than try to modify their behaviour with an indirect scheme."

So first, maybe the mods should start by posting a Metatalk thread saying, "hey, this is the kind of comment we're talking about, can you cut down on that please."

I think just defining it would help. I think I have an idea what the problem was, but it still seems a bit vague.

Second, add the description to the FAQ.

Third: it seems to me that there are only a few people, maybe half a dozen, posting this kind of comment. Some of them post in quite a lot of threads. So maybe the mods could just Mefimail them in private and ask them to cut it out or take it easy. (Of course, I might be misunderstanding the precise problem, in which case there could be too many people doing it for this to work.)
posted by TheophileEscargot at 2:22 PM on November 28, 2009 [1 favorite]


We've maintained, from the beginning, that this wasn't science. People's insistence that yes it is science and it's bad science is tiresome.

i don't really see the point of doing it at all unless there was some kind of way of proving something in some way

We'd really like to get to a point where we're not saying "well let's stick with what we've got because at least people are used to that" or "making changes is too hard because some people are assholes about it" but in the interests of transparency, some of the meta-issues of this month include the fact that people got so angry about this that they made the site a less great place to come to work in the morning. That's neither here nor there as far as this favorites discussion goes, but I think it will probably inform future "let's think about changing something" discussions. And maybe that's fine with people, but it doesn't totally sit right with me.

people are going to have opinions about what's done here - and i think the real lesson here is that it's probably better to get those opinions through open discussion here before things are changed, at least if the change is something that's being taken away or going to seriously impact the way use and perceive this place

but as far as i'm concerned, if people want the option of totally not seeing favorites on this site, fine, let them have it as long as those who want to see them can

i'm hoping that people can just decide to live and let live on this
posted by pyramid termite at 2:36 PM on November 28, 2009


We'd really like to get to a point where we're not saying "well let's stick with what we've got because at least people are used to that" or "making changes is too hard because some people are assholes about it"

One thing to bear in mind is that those who are disgruntled have the greatest motivation to post frequently, and at great length, about their disgruntlement. It seems unnatural and brown-nosing to post about how grateful we are for the site and for the work the mods put into making it a valuable part of our lives, but I hope you know (and I think the thread about Matt's recent tribulations serves as a powerful demonstration of that fact) that virtually everyone who cares enough about Metafilter to bother commenting in a MetaTalk thread feels that way.
posted by Horace Rumpole at 2:47 PM on November 28, 2009 [4 favorites]


There definitely is shades of grey as far as opinion goes on this, but at first glance the biggest argumentative stance on this seems to be from people who didn't even try it. I don't mean "hey made it so hard for me to use Metafilter that I opted out", but the whole "this whole thing f*cking blows because it obviously didn't change anything and I didn't notice any difference and there wasn't a problem and did I say it blows because it did and I wan't to complain about it because I didn't even give it a chance." That type of argument seems to be lacking whole informed opinion part.
posted by P.o.B. at 2:53 PM on November 28, 2009 [1 favorite]


Horace Rumpole: One thing to bear in mind is that those who are disgruntled have the greatest motivation to post frequently, and at great length, about their disgruntlement. It seems unnatural and brown-nosing to post about how grateful we are for the site and for the work the mods put into making it a valuable part of our lives, but I hope you know (and I think the thread about Matt's recent tribulations serves as a powerful demonstration of that fact) that virtually everyone who cares enough about Metafilter to bother commenting in a MetaTalk thread feels that way.

Also, to me at least, when the message is "we're going to try this to see how people like it," the implied question at the end is "how did you like it?" For me, the answer to that is "I didn't like it, and I avoided having anything to do with it." That doesn't mean nothing can ever change because I will throw a hissy fit, and it doesn't mean that I am filled with a deep longing to complain. It means you kind of asked, so here it is.

Personally, if the option was preserved, to hide the count of show it, I would be 100% fine with that. If the actual number display was, like the white background, only available to people with accounts, that would be fine with me, too. If people who are vehemently apposed to favorites had the ability to completely disinvest themselves, see no "favorite" info, and not even have their own posts favoritable, that would seem fair to me, if people want it. But if the (implied) question is "what did you think of this?" the honest answer is "I didn't like it and I never saw any point to it."
posted by paisley henosis at 2:58 PM on November 28, 2009 [1 favorite]


I found it really distracting. As my eye grazed over the posted by line on the way to the next post the 'has favourites' would register and I'd be thinking 'really? how many?' and feeling a little jar of frustration that I didn't know. This never went away. Whereas normally my eyes would slide right past the favourite count and nothing would sink in. I might notice the number or I might not but either way it was cool, no jar. So I don't like it on that alone.

But then I don't use the favourite number to change how I read. If I'm skimming I'm more likely to leave out all the long answers (or all the short ones) or very occasionally stop and read something based on a username or a phrase that jumps out. I never used the favourite count as a quality indicator. I also don't use it when giving a favourite myself, I'm not more or less likely to click the plus if there are favourites there or not there already. And I don't fish for favourites in my comments so there was nothing in my posting pattern to change.

So I guess I wasn't the target audience anyway. I had it off for half the month then on and the only difference was that small jar as it coyly told me there were some favourites but not how many. I don't like mystery, I'm bugged by ambiguity, and I'm really looking forward to things being straight forward again.
posted by shelleycat at 3:10 PM on November 28, 2009 [1 favorite]


I joined in November and I have been looking forward to favorites getting back to normal. It's not like a highly favorited comment is higher on the page than others (like reddit) so I don't see the problem with using them to indicate agreement or approval. When I was ghosting I would only notice favorites after I had already read the comment and made my own decision about it. If I saw that a comment I didn't like had lots of favorites I just thought "wow, a lot of people agree with that. I wonder why?" It helped me understand the depth of certain feelings and opinions in the community.
posted by irisclara at 3:10 PM on November 28, 2009


i don't really see the point of doing it at all unless there was some kind of way of proving something in some way

It's possible that not every site decision must be framed as a confrontation requiring a winner?
posted by desuetude at 3:13 PM on November 28, 2009


, but at first glance the biggest argumentative stance on this seems to be from people who didn't even try it

Huh, that makes no sense, since the change was the default behavior at first, so everyone had to try it.
posted by Brandon Blatcher at 3:13 PM on November 28, 2009


Sounds like a lot of people turned it off as soon as they could. I left it alone all month, hoping I would grow accustomed to it and stop noticing it - or perhaps even start to like it - but as it turns out I hate it just as much now as I did at the beginning of the month. I found it extremely distracting (I often stopped to check just how many favorites something had), plus I suspect I was less generous giving out favorites. I gave it my best effort, but I still thought it sucked.
It seems perfectly reasonable to leave it as a preference for people who like it that way, though.
posted by naoko at 3:24 PM on November 28, 2009


i don't really see the point of doing it at all unless there was some kind of way of proving something in some way"

OMG, I FEEL THIS WAY ALL THE TIME
posted by iamkimiam at 3:25 PM on November 28, 2009


Huh, that makes no sense, since the change was the default behavior at first, so everyone had to try it.

Let's not do that and really try to be honest here. How long did you try it out for? A couple minutes? Hours? Days? Where did you draw the line at giving it a fair shot? It makes no sense to shit all over the idea if you really did not give it a try.
posted by P.o.B. at 3:26 PM on November 28, 2009


interesting -- eccnineten favorites all his own contributions.

That makes it seem to me like he doesn't understand the system in some way not that he's trying to make some kind of point.
posted by shelleycat at 3:29 PM on November 28, 2009


some of the meta-issues of this month include the fact that people got so angry about this that they made the site a less great place to come to work in the morning. That's neither here nor there as far as this favorites discussion goes, but I think it will probably inform future "let's think about changing something" discussions. And maybe that's fine with people, but it doesn't totally sit right with me.

Well, this wasn't a '"let's think about changing something" discussion', it was more, "Oh we're changing something in a few days, so just bear with us for a couple of weeks".

I think you guys just didn't realize that to a lot of people there wasn't a problem and rather than giving them a choice, you instituted a top down change with a general attitude of "hey, it's just a few weeks, can you just deal, please, while we try it out." The problem was that it only took a few minutes to for some to decide to the change not only sucked, but actively loathed it and ya'll are asking them to grin and bear for a month and frankly it seemed pretty pointless.

If someone told you to try X food and you didn't like it and they said "Oh, just give a month of regularly eating it", you'd probably be annoyed with by them insisting that you just try it for a while.
posted by Brandon Blatcher at 3:30 PM on November 28, 2009 [2 favorites]


I think the one thing I'm really sure about the experiment is that, if you're gonna remove them, don't do it half-assed. Halfway doesn't work.

I don't truly care either way. If I had to put numbers on it, I'm probably 60/40 weighted toward keeping them. But I'm 100:0 against the partial removal.
posted by Malor at 3:35 PM on November 28, 2009 [2 favorites]


Let's not do that and really try to be honest here. How long did you try it out for? A couple minutes?

FYI, you don't get to decide how long I should try something.

I think I tried it for a half hour to at most and I. COULD. NOT. STAND. IT. To me, "faved" and then "has favorites" are useless bits of information, repeated ad nauseam on the page, that replaced actual helpful and concrete pieces of information. Either show favorites or not, but don't dangle the info out there half way and force me to do a another click to find out the information.
posted by Brandon Blatcher at 3:36 PM on November 28, 2009 [8 favorites]


FYI, you don't get to decide how long I should try something.

FYI, you don't need to be defensive.

I'm not sure why people need to be overly aggressive on putting the stink on the idea. And really people don't seem to be able to grasp the idea of what the point of it was, or they're rather fine with the actual and real persistent problems of unwanted site behavior.
The point was (if I've grasped this correctly) never about getting rid of favorites. It was about trying to "trim the fat" by tweaking the availability of the reinforcer. I know people like to be jokey and hand wave away idea with some snark like "haha i can use faves any way I like, cuz sometimes i run out of crumpets and then i just break out the favorites at tea time cuz they are delicious and Alice and The Mad Hatter get really hungry" but if you could've hung in there and actually stopped to ask yourself how you really doled out favorites then good on you. If you couldn't stand it, then make it clear why.
posted by P.o.B. at 3:45 PM on November 28, 2009


Brandon Blatcher writes "There's a big difference between adding a feature and taking away. "

That's a philosophy difference that we'll have to agree to disagree on. It's the reason Office specifically and Windows generally is such a bloated piece of crap. Once MS floats practically any feature they can't stop supporting it in perpetuity. I'd like to think Matt can change his mind and revert on new features even after significant periods of time have elapsed. What if the new feature was avatars or signatures or down voting or threading I'd hope it was easier to revert than to implement in the first place.

mr_crash_davis mark II: Jazz Odyssey writes "This is fun! When do I get my pony?"

I can't figure out if you were rebutting my point or not but, Ya. Exactly. Things just sort of happen here. There isn't any huge lead up or six rounds of sending recommendations to assorted sub committees for discussion before tabling a debate and taking votes. Sometimes Matt polls the users a bit and sometimes he responds to pony requests and sometimes he responds to outrage. I wish people weren't so quick to hit the outrage button for what was after all a temporary change. The Suicide Girls thing was handled much better.
posted by Mitheral at 3:45 PM on November 28, 2009


Here is why it bothers me. First, because it makes me feel out of place. I have no great love for quotes from the Simpsons or Family Man or other pop culture quotes, I think they are lazy and cheap and stand in for original thought. Second, and far more important, we have had people say they study what gets favorited so they know how to comment. What message does a highly favorited Simpsons quote give? That we need more people to pull Simpsons quotes out of their ass? In fact over and over we have heard variations of this: "The problem that I had was that there was no simple way to scroll through a thread, avoid the noise and get to the gold pretty quickly." So quotes from TV shows are "the gold" and original thoughts are the noise. Great.

I dunno, Gravy. I'm okay with people, even many people, having dissenting tastes and opinions from mine. It seems to me that the real way to build a community that doesn't solely reward pop culture quotes is to encourage people who post things other than that. A good way to do that is to show your appreciation through a favorite--that seems more effective, to me, than silencing the pop-culture quoters. Also, why assume the worst of the majority voice? I'm honestly not a huge fan of ill-timed Famly Guy quotes, either, but people are, and that's okay. Just because I don't appreciate it doesn't mean it's noise--and I'd hope that other people reading would feel the same way about more thoughtful, and less favorited, comments.

I mean, even when I've used favorites to skim, I don't assume that the rest of the comments are worthless.
posted by PhoBWanKenobi at 3:50 PM on November 28, 2009


Oh yeah. Some days when I'm reading the letter section in the paper (hard copy) and there's one I actually agree with my mouse hand twitches involuntarily as I look for the little plus down below. So I clearly use them as some kind of agreeing/voting system. (and yes I spend too much time online)

It would have been interesting to see how things went with no favourite count at all. You can assign them, you can see them aggregated in the usual places, but you can't see them in the thread. I would have stuck with that for the full month and it wouldn't have been jarring. I think an aboslute change would have been more likely to have an effect on other's behaviour also, the current experiment seemed too wishywashy (none of which is at all scientific but just my vague feelings). But I think the well has been poisoned now and the mods will never get away with trying that experiment.
posted by shelleycat at 3:51 PM on November 28, 2009


And really people don't seem to be able to grasp the idea of what the point of it was, or they're rather fine with the actual and real persistent problems of unwanted site behavior.

Some people grasp the point but disagree with the premise.
posted by PhoBWanKenobi at 3:51 PM on November 28, 2009 [3 favorites]


So you think threadshitting is fine and good?
posted by P.o.B. at 3:55 PM on November 28, 2009


I don't think there's a runaway problem with threadshitting on metafilter, and i don't think that the threadshitting that exists is particularly tied to favorites, either.
posted by PhoBWanKenobi at 3:56 PM on November 28, 2009 [4 favorites]


It's been stated over and over that people view favorites as a reinforcer. If the "wrong" things are reinforced then the problem has a tendancy to balloon. I don't think there's hard numbers one way or the other on if it is a runaway problem but I believe jessamyn has stated a few times that there is a tendancy for it to ruin threads if inserted early enough.
posted by P.o.B. at 4:01 PM on November 28, 2009 [1 favorite]


Yes, I understand that, but I disagree that it's a problem that the existing methods of moderation aren't controlling sufficiently. Which is to say, I think the overall tone of metafilter is fine and I don't see a need to attempt to institute measures of changing it.
posted by PhoBWanKenobi at 4:05 PM on November 28, 2009


FYI, you don't need to be defensive.

No, it was offensive in the face of someone who presumed to know what I was thinking and wanted to decide how long I should give the favorites experiment a try. To repeat: You don't get to decide how long anyone should try the favorite change.

I'm not sure why people need to be overly aggressive on putting the stink on the idea.

Because they didn't find the interface change useful or warranted.

The point was (if I've grasped this correctly) never about getting rid of favorites. It was about trying to "trim the fat" by tweaking the availability of the reinforcer.

That was not my understanding of the experiment. Instead it seemed like a way to turn favorites "off" for a month to see what kind of effect it would have. I get the impression that the mods didn't have any present notions or were trying to prove any particular view point. Rather it was more along the lines of "Some people say favorites contribute to crappy behavior. We don't think so, but fine, we'll turn them off in a manner of speaking and see what happens". That's about the extent of it.

On preview:
It's been stated over and over that people view favorites as a reinforcer.

I really think you're giving a one sided view and blowing it out of proportion.

It's been stated over and over that there in no single use of favorites, people use them in different ways. Your insistence pushing the "people view them as a reinforcer" seems at odds with that the community has said about favorites.
posted by Brandon Blatcher at 4:16 PM on November 28, 2009


Genuine question to others: did you feel that the experiment reduced threadshitting and snarkiness?

I don't think it had any effect. But because of my timezone and (I guess) the way I do things I'm often somewhat late to threads, by which time they've been cleaned up. So I think I miss a fair bit of it anyway. I waver between following avidly the contentious threads and avoiding them all together but, either way, I'm terrible at remembering names or noticing who's posting so keeping track of specific arguments between users or taking sides isn't something I do a great deal anyway. So that specific behavour (a small group fighting while others egg them on with favourites) might have changed and I didn't notice. It will be interesting to see if the mods found a positive change from their end since they see all the crap that I then get to miss, something I really appreciate by the way.

I also try to avoid posting shitty throw away comments myself (whether I'm successful or not may be another thing of course) and never post specifically to gain favourites, so I didn't really feel like there was anything in my own style to change based on this experiment specifically.
posted by shelleycat at 4:18 PM on November 28, 2009


Yes, I understand that, but I disagree that it's a problem that the existing methods of moderation aren't controlling sufficiently. Which is to say, I think the overall tone of metafilter is fine and I don't see a need to attempt to institute measures of changing it.

I guess the mods felt differently, and (at the expense of making an argument from authority) maybe they would know better. Or at least they felt they could make an honest effort in finding out.
I really don't think they set out to scoff at the collateral damage of ruining anyones experience intentionally but that seems to be the imposition that's being place upon them.

No, it was offensive in the face of someone who presumed to know what I was thinking and wanted to decide how long I should give the favorites experiment a try. To repeat: You don't get to decide how long anyone should try the favorite change.

My point exactly, I wasn't being offensive and yet you seem to be defensive about something you could answer easily. You think it doesn't matter how long someone tried something before they have an informed opinion about something and I would disagree. And for the record I never stated how long you should try it, but obviously I do think it adds weight to a viewpoint in the frame of experience on using it.

It's been stated over and over that there in no single use of favorites, people use them in different ways. Your insistence pushing the "people view them as a reinforcer" seems at odds with that the community has said about favorites.

*shrug* I'm not really up to going around in circles on this one. If you disagree fine.

And perhaps this is devolving into something it should be.
posted by P.o.B. at 4:29 PM on November 28, 2009


*should not be*
posted by P.o.B. at 4:30 PM on November 28, 2009


Brandon Blatcher wrote: I think you guys just didn't realize that to a lot of people there wasn't a problem and rather than giving them a choice, you instituted a top down change with a general attitude of "hey, it's just a few weeks, can you just deal, please, while we try it out."

I'm pretty sure that the Mods knew that a lot of people liked things just fine the way they are. However, they also know that there were a lot* of people to whom there was a problem. By having this time-boxed experiential trial, it may seem or be unfair to fans of the status quo, but isn't it possible that by never trying this, they would be unfair to that group of people for whom there was this concern? As one of the few vocal folks who were ambivilent about favorites (I bookmark mostly), I thought it was interesting, and I didn't feel put upon by the request.

I know that I've learned a lot about my own usage from this. I even spent some time wondering if I should be favoriting as an enforcer more for all of those people who don't have time to fully read the threads when people were saying something important and who should especially be heard even when they had few favorites or I disagreed with their POV. It made me sad a couple of times when comments that I thought were really good didn't have any favorites or 1-2 favorites might be missed by a lot of people. Ultimately, though - and I don't know why - I'm generally not comfortable favoriting-to-agree/say "good job"/ etc.

* I'm using that as a general container, not trying to quantify the number in any way, especially in regards to your usage of lot.
posted by julen at 4:37 PM on November 28, 2009


I guess the mods felt differently, and (at the expense of making an argument from authority) maybe they would know better.

I wasn't saying anything about the mods or their instituting of the experiment. I respect that they wanted to try this, particularly since it was at the insistence of some members, even if I disagree with the choice and the implementation. But I'm trying to say that we get it. We just still don't agree.

In fact, I was under the impression that the mods didn't think favorites were particularly the problem, either. Looking back, in the epic favorites thread, jessamyn said: "Just to be clear, we're not trying this out as a 'hey we're planning on implementing this' but that a lot of people have, for a long time, been saying they think the favorites system encourages crappy behavior. We've been saying 'We don't think so without any real way to test." My suspicion is that there's just as much snark and threadshitting as there was before the experiment, because the general tone of the site feels the same to me. But then, again, I was okay with the overall tone of the site to begin with.
posted by PhoBWanKenobi at 4:38 PM on November 28, 2009


I can't help but wonder, too, if the real "problem" with threads having altered (fighty, snarky, whatnot) tone is really just because early comments tend to establish the overall tone for a thread, regardless of favoriting behavior.
posted by PhoBWanKenobi at 4:40 PM on November 28, 2009 [1 favorite]


Shit. I live in a cave and don't have a television and my relationship with pop culture is distant, at best. Are there a lot of comments that are from the Simpsons or Family Guy? Are there a lot of you who I, in my ignorance, think are terribly witty but you're just quoting cartoons? I am being completely serious, except the part about living in a cave.
posted by little e at 5:02 PM on November 28, 2009 [4 favorites]


It's possible that not every site decision must be framed as a confrontation requiring a winner?

there's no confrontation, losers or winners implied in the idea of proving something; simply a quest for meaningful knowledge

sorry to disappoint you
posted by pyramid termite at 5:06 PM on November 28, 2009


In fact, I was under the impression that the mods didn't think favorites were particularly the problem, either. Looking back, in the epic favorites thread, jessamyn said:

My short definiition of the situation probably put too fine of a tip on the brush when the strokes were much broader and painted a bit of a different picture. Of course I would defer to what any of the mods had stated the intent of the experiment was.
posted by P.o.B. at 5:19 PM on November 28, 2009


Here is why it bothers me. First, because it makes me feel out of place. I have no great love for quotes from the Simpsons or Family Man or other pop culture quotes, I think they are lazy and cheap and stand in for original thought. Second, and far more important, we have had people say they study what gets favorited so they know how to comment. What message does a highly favorited Simpsons quote give? That we need more people to pull Simpsons quotes out of their ass? In fact over and over we have heard variations of this: "The problem that I had was that there was no simple way to scroll through a thread, avoid the noise and get to the gold pretty quickly." So quotes from TV shows are "the gold" and original thoughts are the noise. Great.

Geez, it's like some people never went to middle school. Popularity and merit do not always have a 100% correlation. Such are student council elections, favorites, and pretty much everything else in the world.

Also the show is "Family Guy," "Family Man" was a pretty mediocre "it's a wonderful life" rip-off starring Nice Cage. But I have a sneaking suspicion you just might have already know that.
posted by drjimmy11 at 5:45 PM on November 28, 2009 [2 favorites]


we have had people say they study what gets favorited so they know how to comment

"Tell me where you got that Sideshow Bob quote!!"

"I learned it from watching YOU, dad!!"
posted by drjimmy11 at 5:48 PM on November 28, 2009 [1 favorite]


I tried it for a day or two and then switched it back. I like being nosy and seeing who favourited what, it's led me to noticing and paying more attention to some very interesting users.
posted by Flitcraft at 6:05 PM on November 28, 2009


One thing is now abundantly clear.

In past threads about favorites, it was in question whether or how favorites have changed people's on-site behavior.

This thread offers incontrovertible proof that favorites do, indeed, alter behavior:

I prefer seeing the favorite count. I logged in on a different computer where my preferences weren't changed. I clicked on a "has favorites" to just see the amount of favorites . . . and the list was a mile long. For some reason I really wanted to see a number - but dang I was too lazy to count them myself.

On my own comments, I like to see how many favorites it gets - not for a popularity contest, but maybe because I'm a little insecure and I like to see that the things I wrote meant something to someone, that I said the right thing, and that my ideas aren't completely crazy.


*

I felt like my totals were inflated, and felt like I (ironically) gave less. Maybe I actually didn't, but the balance of favorite to favoriting was closer going in than coming out.

I like the visual feedback of seeing how much something was liked.


*

I was a little annoyed at how a self-favouriter could get the same amount of eyes to their post as someone with lots of favourites.

*

I think it's made it a little easier to read threads with disagreements and arguments, since there's less of a sense of one side winning or losing. Makes the atmosphere less tense, I guess.

*

I opted out from day one, but I think being aware of the experiment altered my behavior slightly.

For one, I started making a bigger effort to say some form of "Yes, I agree" or "cool comment" in-thread as opposed to just favoriting.

I also assumed (perhaps wrongly) that the experiment would result in lower favoriting activity, so in an effort to add an extra filter, I made a bigger point to add contacts whose contributions I like. Because of that, the last month has actually felt a bit more cliqueish.


*

score upon score of users came there to complain that they suddenly had no way to communicate which comments they thought were right – and, more tellingly, at least half a dozen people told me personally there that they were disappointed that they were suddenly unable to tell me I was wrong by favoriting someone else.

*

Let's say if there were two borderline favoritable comments (meaning they're the same quality/ say the same thing). If comment A already had favorites, I probably wouldn't favorite it, but if comment B didn't have favorites I'd favorite it over comment A. So I guess I used it to even out the favorites on a thread.


*

So my history as a contributor has always involved favorites and the attendant 'yardstick of quality' sense you get from them, especially favorite counts on my own comments and posts. The attention I pay to favorites is frankly embarrassing to admit.

*

I still obsess over how much I get favourited, and occasionally I will click through to see who favourited another comment/post, at about the same rate I would when the number was displayed.

I have noticed in the past that if a comment has a bunch of favourites I am reluctant to favourite it myself, because I hate being part of a crowd.


*

When I'm reading a thread just for the sake of reading or one I may relate to, I use the number of favorites each answer has to filter out all the other 23 answers that have none.


*

Favorite Counts are my laugh track, my Greek chorus, they let me know how people are feeling about things and I didn't realize how much I enjoyed that part of the community until it was gone.


*

I liked it, as it may me more judicious with dispensing favorites and less obscenely fascinated with my own fave counts.


*

The problem that I had was that there was no simple way to scroll through a thread, avoid the noise and get to the gold pretty quickly. If faved (or bananas, for that matter) was reinstated, I wouldn't mind, as long as there was still some way to see what was worth looking at, and at a deeper level, what was more/less worth looking at within the umbrella of "worth looking at".


*

If there's one lesson I've drawn from this, it's that favorites are indeed used by an awful lot of people, including me, to 'keep score'. If you give people something that's called "favorites" and accumulates in their profiles, it will be used as a score, period, no matter how frothy some people get about scorekeeping.

posted by Joseph Gurl at 6:07 PM on November 28, 2009 [5 favorites]


> You think it doesn't matter how long someone tried something before they have an informed opinion about something

Favorites on or off, I am still reading the damn site, so I am well informed on whether the experiment improved the quality of comments. Look, you (and a few others) repeatedly assert that favorites damage the site as if this were a fact we had already agreed on. So, they were mostly gone for most users for a whole month. If you are correct, then surely the quality of comments has gotten better, pop culture references have decreased, and no one makes snarky thread-shitting comments anymore. Do you feel this to be true? Has it had any effect at all?

To be clear I don't even think comment quality is a problem in the first place, but this is a matter of opinion. Whether it's caused by favorites is not. Either there's a correlation between people's behavior and visible favorite counts or there isn't. I'm not sure how opting out changes this or invalidates my take on it. I didn't say how much I hated the experiment, because -- believe it or not -- I don't. I turned it off/on a couple times throughout the month, and it's still off on my iPhone. I prefer the old way, so shoot me.

What does irk me is that people keep asserting something which hasn't been proven. This experiment was supposed to see if it was even true, but no one wants to talk about this, just go back to bitching about what bad behavior they think favorites cause.

I'd really, really like to know what % of active users have opted out. It's possible, I guess, that everyone who is part of this perceived problem opted out. It seems unlikely to me, but that's a total guess.
posted by cj_ at 6:11 PM on November 28, 2009 [1 favorite]


Favorites on or off, I am still reading the damn site, so I am well informed on whether the experiment improved the quality of comments.

Quality? That's an interesting frame, but I haven't looked at it through that lens. But I'll still maintain to have a viable opinion of how the experience affected you, you would've had to have gone through the experience. That's obviously just my opinion, but it certainly does appear that the people who did keep them on for an extended amount of time have a bit of depth to their insights rather than just "I HATED IT!"

Look, you (and a few others) repeatedly assert that favorites damage the site as if this were a fact we had already agreed on.

I have never said that, and as a matter of fact I like favorites as an utilitarian tool. My opinion is even more nuanced than that but I'm not going into that because I haven't argued against favorites.

So, they were mostly gone for most users for a whole month. If you are correct, then surely the quality of comments has gotten better, pop culture references have decreased, and no one makes snarky thread-shitting comments anymore. Do you feel this to be true? Has it had any effect at all?

The answers I give regardless would be pretty subjective, right? Like I said, I don't even take umbrage with most of those things.

What does irk me is that people keep asserting something which hasn't been proven. This experiment was supposed to see if it was even true, but no one wants to talk about this, just go back to bitching about what bad behavior they think favorites cause.

I would also, and the only things I've said I've tried to maintain a simplicity about. If there is something specific I said that irks you let me know.
posted by P.o.B. at 6:36 PM on November 28, 2009


I hate change.
posted by furiousxgeorge at 6:51 PM on November 28, 2009 [3 favorites]


Just another data point -- I left the "has favorites" setting on all month, didn't really notice any difference in the overall Quality Of My Metafilter Experience, if given the choice would probably leave things set this way, but wouldn't pitch a fit if it went back to showing counts.

I have to admit I've never really understood how people use favorites to filter the 'Filter -- to me, it's more time-consuming to look for "favorites" indicators (whether old- or new-style) than to just skim the freakin' comments -- but this experiment has taught me that others read the site rather differently than I do, which is fine. I'll admit that from time to time -- maybe three or four times a week -- I would click into the "has favorites" link on comments to which I had a particularly strong reaction, not so much to get a headcount, but to see which particular users had favorited that comment. I don't care about vote totals, but I do value the way that looking at others users' patterns of favoriting can give me a better understanding of them -- which of course falsely assumes that I know their intentions in clicking the "+" sign, which I don't.

The experiment I would really like to run, had I god-like powers, would be to go back in time and replace "favorites" with "bookmarked" or some such. (Just as I wish I could retroactively set up Livejournal's system to replace "Friends List" with "Reading List.")

Anyway, it seems to me that the people who kindly develop Greasemonkey scripts give us the ability to view the site in our preferred fashion, so I'm not clear on the need for the vehemence. It's really cool that people have many and diverse ways of using the "Favorites" function, but not so cool when we use those differences to bludgeon each other, or the mods.
posted by Kat Allison at 7:28 PM on November 28, 2009


P.o.B., as I understand it, you and others feel that favorites serve as a reinforcement for threadshitty/otherwise bad comments.

Yes. A reinforcer is a reinforcer, it promotes both good and bad behavior in the same way. That's why I thought the trial was a great idea by way of hiding the numbers and not just chucking favorites altogether.

Do you feel that those comments have been reduced in number during this experiment?

From what I've noticed there does seem to be a reduction in thread$h!#!ng. There were a couple of times I was surprised by what I thought was blatant threadcrapping and there wasn't the "usual" favoriting activity associated with it. That's just what I've noticed and perhaps I was just noticing at the *ahem* "right" times.
posted by P.o.B. at 7:29 PM on November 28, 2009


The experiment I would really like to run, had I god-like powers, would be to go back in time and replace "favorites" with "bookmarked" or some such.

Perhaps this should be an option. A free-form text field, under preferences. People who like "favorites" can choose to see "favorites", others can see "bookmarked", or "pumpkin pies", or "nipples". As a bonus, this renders the situation nearly impossible to coherently discuss.
posted by little e at 7:53 PM on November 28, 2009 [1 favorite]


I learned that it actually is cool for a boy to do ballet, even if some creep at school named Dalton calls you a "sissy" and a "pansy."
posted by mazola at 8:11 PM on November 28, 2009


Perhaps this should be an option. A free-form text field, under preferences. People who like "favorites" can choose to see "favorites", others can see "bookmarked", or "pumpkin pies", or "nipples". As a bonus, this renders the situation nearly impossible to coherently discuss.

Now THIS is an idea I can get behind. I would probably go with "XP"
posted by EatTheWeek at 8:31 PM on November 28, 2009 [2 favorites]


MetaFilter: As a bonus, this renders the situation nearly impossible to coherently discuss.
posted by mazola at 8:33 PM on November 28, 2009


OK, I'll say it. Favemongers hate when favorites are removed. The most vocal advocates are people with lots of favorites, who live for favorites. I'm sympathetic to anyone seeking approval, from whatever source, I do that, I respect that, nothing wrong. I've experimented with comments tailoring them to attract favorites. Embarrassingly, I like the favorites I get when what anyone else thinks shouldn't matter. I like to think that my opinions are my own and unaltered by confirmation. I am of course, wrong in that assumption. Worse, some opinions are relentlessly expressed by a vocal (cryptic?) subset and if you're not in the club then they "Watch as the point flies over your head". She's a witch, burn her. I hate that they cause me to comment with respect to some perceived popularity .
Favorites have benefits - less "me too" comments, there are drawbacks - the horde.
Are we better for less noise or are we better for having an echo chamber? Try and find a conservative or a religious person here that isn't shouted or favorited down. There is your answer.
Constructively, I'd like the option for all members to choose for themselves. Cynically I'd like for the favorites to be visible to everyone but the favorited.

Turkey soup.
posted by vapidave at 8:36 PM on November 28, 2009 [2 favorites]


The most vocal advocates are people with lots of favorites, who live for favorites

Perhaps you should't presume to know why people like favorites, you could just ask.
posted by Brandon Blatcher at 8:44 PM on November 28, 2009 [5 favorites]


I liked it and hope that the option to hide the favorites count persists after November. The flow of a thread's reading is much smoother this way.
posted by ignignokt at 9:23 PM on November 28, 2009


Perhaps you should't presume to know why people like favorites, you could just ask.
Why people like favorites has been abundantly voiced.
posted by vapidave at 10:12 PM on November 28, 2009


I like favorites. I feel validated by the community when I get them. I would like for things to go back to normal.
posted by phrontist at 10:15 PM on November 28, 2009 [2 favorites]


Good Christ, I don't know where you all get the strength to keep arguing about this.
posted by Space Kitty at 10:26 PM on November 28, 2009 [3 favorites]


Maybe from the same reserve in which you found the strength to comment on it.
posted by Joseph Gurl at 10:32 PM on November 28, 2009 [1 favorite]


I like not seeing favorites on Metafilter. I wouldn't miss the "has favorites" button if it went away, either. I don't need to see how many favorites a comment has (even if it's just one), and I don't even need to know who favorited it.

It makes the site feel cleaner for me and gives it a less of a "me too" vibe. With favorites, metafilter always felt a little like a place where many people who felt they were above popularity contests were still very popularity-conscious. (Whereas people in places like Digg or Reddit at least flat out admit to karma whoring.) Just a perception of mine, correct me if I'm wrong.

If there's an option post-November to turn off seeing how many people favorited a comment, or the fact it was favorited, I would definitely use it. The only favorites I need to know about are the ones I marked myself, so I can revisit them.
posted by thisperon at 12:03 AM on November 29, 2009


Not bad, but how about -

Metafilter: nearly impossible to coherently discuss.
posted by Pronoiac at 1:01 AM on November 29, 2009


If someone told you to try X food and you didn't like it and they said "Oh, just give a month of regularly eating it", you'd probably be annoyed with by them insisting that you just try it for a while.

If there is one thing that does annoy me it is when I cook something a little unusual and people decide they don't like it on the first mouthful, that really is no way to judge food. Odd enough the way to acclimate yourself to unusual tastes is to persist with them.
I really didn't like the extra hoppy beers when I moved out to Oregon from the UK as they were way outside my usual tastes. I persisted however and the range of what I can appreciate has grown wider for it.
posted by tallus at 1:39 AM on November 29, 2009 [2 favorites]


I lasted a couple of weeks before switching back. I found that on the few occasions when I actually was interested in how many favourites something got, finding out on a separate page broke my reading flow. I'd be all right with a hidden, mouseover type thing, though.

Robots don't give a shit about you.

Dammit, Jessamyn. Now Artoo and Wall-E are hiding under the bedclothes again.
posted by Sparx at 2:07 AM on November 29, 2009


When the counts went away, I stopped feeling the urge to +1 increment the favorite count of the posts I liked. I used to mostly favorite posts that already had favorites, moreso when they had more favorites. Now I just click + when it was valuable to me. Maybe I laughed, maybe it had deep meaning. I prefer it this way, I think.
posted by crysflame at 2:41 AM on November 29, 2009


I'm surprised to find that anyone actually likes the 'has favourites' mode of display. It would seem to me that to people who have no interest in favourites 'has favourites' is just visual noise, while omitting the number withholds useful information from people who do.
To me, seeing or not seeing 'has favourites' was more attention-grabbing than seeing the actual number of favourites - I started to find myself subconsciously reading 'has favourites' as 'has been noticed and approved of by the community' which made for a rather weird reading experience. For this reason, I opted out of the experiment as soon as the option was available.

I guess that if the experiment has taught us one thing, it is that different people have different preferences regarding favourites and that some people feel very strongly about their preferences. Creating some user-controllable options for displaying / not displaying favourites may be the way to go here.

Personally, I would be interested in an option that allows me to see the number of favourites on everyone else's posts and comments, but hides all favourites on my own comments from view. It's not that I don't appreciate the favourites, it's just that I'm not always comfortable with the idea of favourites influencing my commenting behaviour.
posted by rjs at 5:34 AM on November 29, 2009


Data point: I went and grabbed the greasemonkey script to keep favorites off. I haven't monkeyed around with it to change the default mask, so all I see for "has favorites" is a star. I think I like this view best of all.

I hope the option to hide favorite counts remains.
posted by immlass at 7:39 AM on November 29, 2009


To me, seeing or not seeing 'has favourites' was more attention-grabbing than seeing the actual number of favourites

I'd have much preferred if the link simply said "favorites", even when there were zero. Sparx's mouseover idea is also good.
posted by cillit bang at 8:22 AM on November 29, 2009


I wasn't being offensive

This is what you wrote:
Let's not do that and really try to be honest here. How long did you try it out for? A couple minutes? Hours? Days? Where did you draw the line at giving it a fair shot? It makes no sense to shit all over the idea if you really did not give it a try.
In just a few sentences, you presumed to know what I was doing and that I wasn't honest in some way. Then you proceeded to try and establish some marker as how long a person should try something, then turned around and belittled whatever decision they came to by classifying their voicing of said decision as "shitting all over an idea" and finishing it off by once again trying to establish your personal benchmark as the standard as to how long others should try something before they decide they don't like.

So no, you weren't being offensive, you were being arrogant, rude, condescending, belittling and offensive. If you feel the conversation has gone somewhere it shouldn't, then in the future don't take it there.

I prefer favorties. I loathe "has favorities" with a passionate disdain usually reserved for eggplant, coffee grounds in my coffee, Microsoft Publisher, bad SUV drivers and Sarah Palin. If you don't think the way I do, that's cool, it doesn't make you any better or any worse than me. I am curious on a human level to hear why you and others view things the way you do, it really is fascinating to learn how others view the world, so I'd love to hear what anyone has to say on the matter.

But by the same token, I expect not to be belittled for the way I think and feel about favorites and if we're to have any sort of civil conversation about this, it's imperative that all sides treat each other with respect.
posted by Brandon Blatcher at 8:30 AM on November 29, 2009 [5 favorites]


The flow of a thread's reading is much smoother this way.

Says you. Hidden counts broke my thread reading method, where I skim then read in detail.
posted by EatTheWeek at 8:36 AM on November 29, 2009


It would seem to me that to people who have no interest in favourites 'has favourites' is just visual noise

At first it was- when you're used to seeing a number, it was obvious something wasn't quite right. But it has now been 29 days, and I'm used to it.

Prior to the experiment, I would read the posts and most of the comments of the topics that were interesting to me. If a topic wasn't my cup of tea, I would skim the comments for favorite count and if something had a high count, I would read it. Sometimes, based on that comment, I would end up with an interest in the topic after all. Sometimes I wouldn't. A good example is the wedding-party dancers in that SLYT post. I never read those because I don't have speakers at work, and SLYTs are kinda useless without sound. But the comments in that post were so enthusiastic, I checked it out, and I really enjoyed the clip.

Now that I have the counts turned off, I miss stuff I would have seen before. On the other hand, in the posts I do read, every comment now stands on its own merit. I have no idea how "popular" a comment is unless I click to find out. Pre-experiment, I didn’t think that I favorited comments that already had a bunch of favorites. Post-experiment, it seems I do (well, did). Now I favorite things that are funny/touching/interesting to me, whether or not anyone else thinks so. I’ve also added a few more contacts and actually look at the sidebar now (pre-experiment I had the contact activity thingy closed). Turns out, I’m married to some very interesting and intelligent people.

So.
Has the favorites experiment changed the way I read the site? Yes. I am probably missing some things, but for the most part, I’m learning more too. Reading every single comment in a thread has turned up some gems that I wouldn’t have otherwise seen if I was just skimming for favorites.

Has the favorites experiment changed the way others comment? Hard to tell. I came to MetaFilter to find the best of the web. The comments/discussion/community aspect are just a bonus. Most of the FPPs I read are not really contentious ones and there isn’t too much to snark about. If I need a snark fix, I come to MetaTalk.

Has the favorites experiment changed the way I comment? Not really. I don’t like to talk unless I have something to say, and when I try to talk without having something to say it just sounds weird and awkward like I’m trying to be one of the cool kids and failing miserably.

In conclusion, if it continues to be an option, I think I’ll leave the favorites count off.
posted by dogmom at 8:54 AM on November 29, 2009 [1 favorite]


I'm sorry if November was unpleasant for you.

Well, the tumor wasn't your fault.

Seriously though, I think the people who are getting bent out of shape one way or the other really need to examine their lives if a change in the "scoring" system on a website affected them to the extent that they got excitable about it. I know, I know, "Welcome to the internet."

I found it a bit annoying, stuck it out, but don't really care enough to have an opinion either way. So perhaps I should duck out and let the people who care battle it out, but I am chiming in mostly to register my feedback. There's probably more people like me than people that love/hate it.
posted by cjorgensen at 9:08 AM on November 29, 2009


It would seem to me that to people who have no interest in favourites 'has favourites' is just visual noise...

Yep. I no longer have any interest in seeing favourite counts, yet 'has favorites' irks me. I wrote a Greasemonkey script to eliminate it altogether, which works fine when I'm using my home PC with Firefox installed, less fine when I'm on a lab computer at school. Being able to remove 'has favorites' using the Preferences screen would be awesome.
posted by threetoed at 9:10 AM on November 29, 2009


Seriously though, I think the people who are getting bent out of shape one way or the other really need to examine their lives if a change in the "scoring" system on a website affected them to the extent that they got excitable about it.

As one of those who got bent out of shape about this, I hear what you're saying, but for me I don't see favorites strictly as a scoring system, but as general feedback mechanism for how the community is feeling about a particular comment or post. The site seems more alive to me with favorites, to such as extent that when it was first turned off, I mistakenly thought the front page had always included favorite counts.

The most interesting about favorites, to me, is just that they're incrementing numbers. I remember keeping tabs on how many favorites I had at certain moments, such as when it was to reach my birth year, 111 (personal reasons) 2450 (Duck Dogers!), or 6666, just 'cause it's neat to see. I sent a MefiMail to someone after some mark has been passed, I think 6666, joking about "thanks for making me evil" or some such and they wrote back saying "congratulations on being so popular" and I thought "What? That number has jack shit to with popularity and is essentially meaningless". I still don't get or understand why people would place such significance on the number, it's just a single barometer of a person on a single website. I can't imagine anyone taking favorites seriously, it doesn't get you more sex or money or frequent flyer miles or chocolate cookies. It's just an arbitrary number that's neat for like 5 seconds.

For instance, think of your favorite mefite. Is the first thing that pops into your head their favorite count? Probably not, you just like them for what they've written on the site, favorites be damned.

I'd like the favorite count to stay, because it reinforces, to me, that Metafilter is community of people, with various likes and dislikes. I don't agree with everything the community likes or agrees with it, but I do enjoy "hearing" what the crowd is in to, much like as I enjoy clicking on the "Most Read" or "Most Emailed" tabs on news sites, to see what the crowd is interested in.
posted by Brandon Blatcher at 9:52 AM on November 29, 2009 [2 favorites]


I'm a reasonably new MeFi member (15 months + counting), so can't speak for the pre-favorites era. But I am starting to wonder (inspired perhaps by all the knocking about that's gone on this month) if there's perhaps a generalized lack of appreciation in the community (from all sides) for how very efficiently the established "favorites-enabled" version of the site reconciles so many of the vagaries of online discourse. That is, one small, easy to use feature allows a user to:

a. bookmark
b. register appreciation
c. reward bravery
d. laugh and "be heard"
e. you name it

Seriously, is there any other feature that could allow this much facility, so smoothly, so elegantly? Of course, favorites aren't always used to "say" the same thing, nor are they comprehended in the same way. But they do add depth to the discourse and, as such, invite us to explore the site even more deeply. For instance, there's hundreds of individual users who's profiles I've visited and explored precisely because of some favoriting they've done. Even if I'm doing it because I UTTERLY DISAGREE WITH THEM, it still enhances the community, because community is as much about understanding and recognizing differences as it is about "just getting along".

So anyway, count me in as a supporter of the "old way", even more forcefully now than I was a month ago. And also, count me in as a fan of the "experiment" because it has definitely inspired all manner of thought and reflection as to the nature of the site that would not otherwise have happened.
posted by philip-random at 9:56 AM on November 29, 2009 [3 favorites]


congratulations on being so popular

If anything it should be 'congratulations on posting a lot'.
posted by empath at 10:23 AM on November 29, 2009 [2 favorites]


I like favorites and seeing the favorites count. When I don't have time to read a whole thread, I read the comments with the high favorites count and get value from that. I also like checking how many people have favorited my comments -- sadly, I am more motivated by e-points than I'd prefer to admit, which is one reason I'm much more active here than I am in other online communities that don't have this feature.
posted by Jacqueline at 10:45 AM on November 29, 2009


I did not turn the feature off, even though I strongly disliked it. From looking at my profile, it appears that I used favorites a whole lot more in November than I usually do, which is weird (especially since I came in here to say "I used favorites a lot less than I usually do!" but then thought that I should check that out first). I gave 30 favorites for comments made in November -- in contrast with 15 in October, 12 in September, 7 in August, and an average of 3 per month for every month between July 2009 and September 2007, when I became a member. I don't really clean out my favorites in any systematic way, but it is a safe bet that I may have gone through at some point in the past two years and weaned things out. Still, though, that wouldn't explain the sharp contrast between September and October with November.

So, it looks like I had just started really using favorites in September of 09. I can now understand why I was a little irked in November when things changed after actually using them for only a few months. I can't tell if the fact that I used them more is a function of the change in the way I was using the site in general or a function of the change in favorites display (probably a bit of both) but I'd be happy to have it go back to the way it was. In November, I used them a lot more for "yes I agree with this" than I would like to -- I think of favorites as my personal library of comments I'd like to remember or that made me laugh or made me think. There is a lot of "noise" in my November favorites that I believe was mostly caused by my usage patterns related only to the experiment. I actually think that I used them more like they were votes than I normally did, not less, which seems counterintuitive based on the goals of the project.
posted by k8lin at 11:05 AM on November 29, 2009


I turned off the experiment as soon as I was able. I didn't find anything inherently good or evil in it, merely a mild inconvenience, and a seething dislike of the anti-favorites crowd. They may have a point, but it comes across more like a tantrum. They're painting a very unflattering picture with an awfully broad brush, and got it all over everyone who doesn't have the same perspective.

A side note: removing functionality is a surefire way to peeve users. I think the most useful insight gained from this experiment is that a website is closer to software than it is to a publication. Yes, readers will lob molotov cocktails at the editor's car for pulling "Beetle Bailey" from the comics page, but this change wasn't that extreme. They generally don't go all poison-pen on the letters page for moving the weather blurb from the top of page one to back page of the first section to see what it does for sales. On the other hand, software users flip a lid if the application deliberately removes functionality they found useful on a regular basis. Sort of like removing the table tool from MS Word to see if users would make larger use of OLE and Excel.

So, Metafilter is community expressed through software, in the minds of its users. That's a pretty useful lesson to learn, and one that surprised me.
posted by Slap*Happy at 11:11 AM on November 29, 2009 [1 favorite]


Whoo! I have 1k favorites!
posted by Pronoiac at 11:11 AM on November 29, 2009


I was moving last month, so when I visited the site and started browsing, I had no idea how and why things looked different. I soon clicked over to MeTa, read the banner at the top, and reverted my settings without reading that bigass Oct. 31 thing.

I don't use favourites to navigate comments, but I'm used to seeing either a [1 favorite +] or just a [+] in a comment footer, and the peekaboo [has favorites +] was jarring and distracting, not just aesthetically, but it actually, perversely, made me give favourites more consideration than I normally would or than they deserve.

"hey, it's just a few weeks, can you just deal, please, while we try it out."

And it appears it took less than a day of negative feedback for them to introduce an opt-out. THE DASTARDS.

Good Christ, I don't know where you all get the strength to keep arguing about this.

Seeing a comment favourited gives one a small burst of adrenaline-laced endorphins. It's ascientifically, experimentally proven fact.
posted by Alvy Ampersand at 11:27 AM on November 29, 2009 [1 favorite]


As one of those who got bent out of shape about this, I hear what you're saying, but for me I don't see favorites strictly as a scoring system, but as general feedback mechanism for how the community is feeling about a particular comment or post.

Eh, you kept participating, didn't call anyone names, and didn't close your account. I don't think this can be said about everyone.

I'm not saying people don't have valid ways they use favorites, but the site existed before them, and if they were found to not scale and had to go away entirely for performance issues, I am guessing not a lot of people would leave over there lack, yet some people acted like the TSA was taking their babies.
posted by cjorgensen at 11:54 AM on November 29, 2009


The really useful data here is the revelation of how many people get really worked up and hateful over some goddamn AJAX. Jesus Christ.
posted by middleclasstool at 11:56 AM on November 29, 2009 [1 favorite]


didn't close your account. I don't think this can be said about everyone.

Ooo, I need names! NAMES!
posted by Alvy Ampersand at 11:59 AM on November 29, 2009


Eh, you kept participating, didn't call anyone names, and didn't close your account.

Goddamnit, I can change, I SWEARS!

I'm not saying people don't have valid ways they use favorites, but the site existed before them, and if they were found to not scale and had to go away entirely for performance issues, I am guessing not a lot of people would leave over there lack,

Not getting you here, as there's a difference between dropping a feature because of technical reasons and dropping it for experimental purposes.

yet some people acted like the TSA was taking their babies.

Thank you for equating vocalizing dislike of the change with being a hysterical liar.
posted by Brandon Blatcher at 12:37 PM on November 29, 2009 [1 favorite]


Also in the spirit of chillness and respect for the mods maybe people can post their thoughts once - maybe twice - and keep this thread from becoming a giant back-and-forth argument. Since this is just a reaction, not "convince other people" thread.

Well, it was a nice idea, anyway.
posted by Durn Bronzefist at 12:43 PM on November 29, 2009


Ooo, I need names! NAMES!

He came back, the one I remember.
posted by jessamyn (staff) at 12:54 PM on November 29, 2009


In just a few sentences, you presumed to know what I was doing and that I wasn't honest in some way. Then you proceeded to try and establish some marker as how long a person should try something, then turned around and belittled whatever decision they came to by classifying their voicing of said decision as "shitting all over an idea" and finishing it off by once again trying to establish your personal benchmark as the standard as to how long others should try something before they decide they don't like.

Do you need me to call the Waaahhhmbulance?

That is me trying to offend you.

Did I frame my argument from my perspective? Yeah, just like you did. You can drop the persecution/innocence complex you got going and read what I wrote a little more objectively. Look, I don't have any interest in keeping this up with you. I've corrected you a number of times about this and you seem to want to make this all fighty especially the crap about me saying there was a time limit when I specifically said it is my opinion that it adds weight to somone's opinion about their experience. The people who stuck with it have more insight about their experience rather than "I tried it and didn't like it." And as I recall when you first tried it you thought the idea so abhorrent you wanted to take a timeout from Metafilter and then the option was floated. That's not really giving it a chance is it?

I'm even conflicted about posting this because I'm sure you'll have some retort about "you did this, this, and this" and as much as I would like to revisit elementary school days with you, where the only reality that existed was my own and everybody else was at fault I'd rather not. So if you can't deal with what I said, please, have at it. I'm done with the back and forth.
posted by P.o.B. at 1:58 PM on November 29, 2009


It's ascientifically, experimentally proven fact.

I thought jessamyn asked very nicely could we please stop with the whole "this experiment is ascientific" crap.
posted by UbuRoivas at 1:59 PM on November 29, 2009


That is me trying to offend you.

It offended me way the hell over here. Not the previous line. That line... and then the whole patronizing, sneering post. Whatever point you're trying to make is completely lost in the odious way you're trying to make it.
posted by Slap*Happy at 2:09 PM on November 29, 2009 [5 favorites]


Whatever point you're trying to make is completely lost in the odious way you're trying to make it.

I think we've found our "everyone needs a hug" replacement!
posted by Durn Bronzefist at 2:24 PM on November 29, 2009 [8 favorites]


I would like to know if others, like me, find that having the number of favourites received visible is different in AskMe than in Mefi. I've only skimmed this thread but I read the whole giant other one, and it seemed that a key concern for those who dislike counting favourites was that it amplified some comments over others, in a way that brought to mind a popularity poll, or that amplified funny one-liners over more substantive content.

I don't read Mefi as often AskMe but it seems to me that these two concerns (popularity poll, jokes over content) are less a problem on AskMe, because a lot of the time, there are correct answers -- or at least answers that are clearly more helpful than others. I use favourites to indicate that I agree with the answer in the post (which others noted in the old favourites thread). My favourite, added to the favourite of others, helps the asker to see which answers are most useful, according to the hivemind. The polling function is a feature, not a bug.

(For this reason, I hope that future AskMe has visible counts as a default, even if MeFi does not.)
posted by girlpublisher at 2:54 PM on November 29, 2009 [3 favorites]


You can drop the persecution/innocence complex you got going and read what I wrote a little more objectively

It's not a persecution complex if they really are out to belittle your opinion*, hee!

Look, we've been going back and forth and while I obviously feel strongly about the subject, I'm not trying to create lasting animosity either. However, I object and will continue to strongly object to you or anyone else who wants to pull the "you didn't try this long enough" card, because I don't think it's a good point. You're still trying to place a standard as to how long a person should try it, while neatly ignoring that they did try it and didn't like it.


*That's not really giving it a chance is it?

Tell me again about how you're not being condescending, I so love hearing that story over and over!

If it makes you happier, I went back and looked at the original thread and noticed I tried the change when first came out, didn't like it, slept on it, tried it for most of the next day and finally said "aw hell no" and started typing up a note explaining why I was checking out for the length of the experiment before the admins changed it to a preference. Hopefully this hours long trial comes closer that mysterious standard you keep insisting you don't have.
posted by Brandon Blatcher at 3:00 PM on November 29, 2009 [1 favorite]


I like visible favourites because I use them to get the gist of long threads, and get my bearings in threads where I know little about the subject, so I wasn't happy about the experiment but kept it on. Regarding giving favourites, I use them mostly as bookmarks, so no change there.

As for reading: without immediately visible counts, when short of time I skim by scanning for particular usernames that I've come to rely on for thoughtful contributions and productive debate. Visible high favourite counts were a way to bring my attention (not automatic favourable opinion) to comments from unfamiliar usernames that I might otherwise have skipped due to time constraints.

Hidden favourites bothers me more for one issue: I've forwarded the Schrodinger's Rapist thread to people who are self-identified feminists, women who are "not a feminist but," parents with teenage sons and daughters, men who think discussing that stuff is worthwhile, in short a wide swath of people. A lot of them for various reasons don't spend much time on the internet. A lot of others just don't or can't read very fast and since they find skimming difficult, they either read something in its entirety (infrequently) or avoid reading it at all (frequently).

The Schrodinger's Rapist discussion is one hell of a long thread and the complexities of the issue took a while to start getting unpacked in the detail that they needed. Before, I would have said, "You can get the gist of the conversation by skimming for highly favourited comments," and also sent them links to the comments I thought were most worthwhile. This time, I just sent the links. Visible favourites lets me give them another option for skimming the thread that is less subjective than pointing them at my personal favourites, and a much less intimidating project than "Start skimming at the top and keep going for at least a couple of hours."
posted by cybercoitus interruptus at 3:26 PM on November 29, 2009


I think that too many people opted out of the experiment to make it a valuable indicator of anything, to be honest. I still prefer not having favorites at all, but this experiment was too flawed to use as an example of how the site could function now without them.

It's a shame, but favorites seem to be here to stay, and the popularity contest that results from them will likely remain as well. If this experiment is ever run again, there should be no opt-out function whatsoever -- but perhaps the period could be shortened to two weeks or so as compensation. Also, I hold by my suggestion to hide one's own favorite counts.
posted by armage at 6:17 PM on November 29, 2009 [2 favorites]


However, I object and will continue to strongly object to you or anyone else who wants to pull the "you didn't try this long enough" card, because I don't think it's a good point.

With the disclaimer that I don't mean to jump into this exchange on one side or the other so much as just pipe up of my own accord:

For my part, I've been trying avoid doing anything like playing a card on this front, and we tried to be pretty clear in the original thread that we were fine with folks opting out because, at the end of the day (or the end of the month?) we were more interested in people not being tremendously annoyed than we were in having some locked-in testing pool with no choice about the matter. So I think it's fine that people who took an immediate dislike to the change opted out quickly and left it at that. That's your prerogative, no harm done, and knowing that some folks felt immediately and sufficiently strongly about the change that they felt compelled to switch it off post-haste is its own vein of information.

That said, I think there's nothing really unreasonable about saying that opting out quickly reduces the depth of feedback about the experience of the change itself that the person opting out can offer. I feel like that's what P.o.B. was trying to convey, though it didn't go all that well between him and you in this thread.

Dialectical friction or whatever on that front aside, I think it's pretty straightforward to claim that folks who spent a couple weeks or a whole month with the change active can speak to how it affected their view of and behavior on the site and perception of the favorites feature in more depth and with more experiential justification for their perceptions of that than folks who opted out quickly. They folks who kept the change live spent more time with it, and we're hearing interesting things from them on that front (ranging from "I stuck with it a month and was surprised by how my view of it changed in ways x/y/z" all the way to "I stuck with it a month and I still hate it") that we can't really expect to hear from folks who didn't stick it out.

It doesn't make the folks who opted out quickly bad people or wrong in any way to have chosen to do so, nor does it make their opinions about other aspects of this whole thing and favorites in general any less valid or important. While I'm grateful for the experiential reports we're getting from the folks who (with misgivings about the change or not) stuck it out for a significant portion of the month, I don't begrudge the folks who elected for whatever reason not to do so.

But I think it's totally fair to say that there's a difference in folk's ability to speak to experiences in this specific context based on that choice, and that while I don't see much use in drawing some firm line of Exactly How Long Is Long Enough, it's plain that a day or so is different from one or two or four weeks in terms of having a change to let the change sink in and stop be jarring, etc.
posted by cortex (staff) at 6:47 PM on November 29, 2009 [5 favorites]


In terms of comment quality: I can't really say; a month might be too short to tell, but a longer, slower change might be too subtle to notice as it happens.

In terms of user experience: I really liked the change, though I think I would've liked it even more if there were an option to not even display "has favorites". I would like to consider each comment on its own merits. I think that a comment is given subconscious "weight" for me by a high favorite count, which can give it undue influence.
posted by a snickering nuthatch at 6:53 PM on November 29, 2009


That said, I think there's nothing really unreasonable about saying that opting out quickly reduces the depth of feedback about the experience of the change itself that the person opting out can offer.

Something I've been thinking about today is that people who opted out of the experiment are still seeing the same comments as everyone else, and so they're just as "qualified", so to speak, to discuss how the experiment has changed comment quality on the site.

If anything, it seems like that group of people can be sure their perception of comment quality isn't being affected by a change to the user interface, because their UI is still the same as it always was.

Now if there has been a change in favoriting behavior, it's possible that people who still see the favorites count could have their perception of comment quality influenced by that. That's something we could possibly get at statistically, in the sense of has there, or has there not been a measurable change in favoriting statistics of various kinds?

But just changing the UI to say "has favorites" instead of a count could easily influence perception of comment quality, too. So it seems valuable to have two groups of people, one with the original UI, and one with the experimental UI, to see how perceptions of comment quality have changed (if at all).
posted by FishBike at 7:08 PM on November 29, 2009 [1 favorite]


Here is a minor case study to consider, from comments I've posted lately:

- link to actual research, later to be defended [2 favorites]
(although perhaps some overreaching at the end partially explains its relative unpopularity?)
- semi-snarky one liner defending that research, simply making the point "ha ha, your rebuttal is wrong and I am right" [9 favorites]

So, it's probably true what people are saying, that snarky comments are more popular. And not having the number of favorites displayed didn't change either what people rewarded or my own future incentive to post something similar. My primary incentive was to defend my original link, and in checking my "recent favorites," I also realized, "the time I spent editing that comment paid off, I guess."

To change what gets rewarded would require doing away with favorites entirely, or removing the ability to check "recent favorites," both of which would be too great of a loss. Perhaps another way to reduce my incentive to make snarky comebacks would be to balance favorites with something that reflects the fact that while the zinger might've pleased more people, it might've also ticked off more people.

But it's possible that obscuring comment favorites did change the tenor of the conversation. Someone skimming the thread who didn't care about the minor correlation/causation side debate wouldn't receive a clue that they should care about that question, or wouldn't have their attention directed to a snarky comment. Also, perhaps the person I was debating didn't know that nine people agreed with my snarky "ha ha" and therefore might've felt less embattled. My rebuttal didn't receive any responses, if I'm remembering right -- perhaps because favorite counts weren't displayed? It's hard to say.
posted by salvia at 8:02 PM on November 29, 2009 [1 favorite]


I thought jessamyn asked very nicely could we please stop with the whole "this experiment is ascientific" crap.

It was a very nicely put request, definitely, but that doesn't mitigate concerns with the poor use of the word, with respect to the goal of trying to discern something correct and useful from the experience.
posted by Blazecock Pileon at 8:54 PM on November 29, 2009


that doesn't mitigate concerns with the poor use of the word

I just think it falls into the "asked and answered" category, unless you think there's some way in which people could say something about this that hasn't been said already?
posted by jessamyn (staff) at 9:06 PM on November 29, 2009


I just think it falls into the "asked and answered" category

Honestly, I'm really not trying to be clever here, but I'm not sure it was really a question, so much as an observation.
posted by Blazecock Pileon at 9:19 PM on November 29, 2009


Oh, for frig, the "ascientific", was totally a typo, and my comment was meant to tweak the LOLSCIENTIFICRIGORFAIL people, not jessamyn et al. Unclench awready, you semantic sillies.

Apologies if I sound testy, but I am pissed because I missed the Grey Cup and also I cant figure out how to stop the blue characters on my laptop`s keys from appearing when I mean to type the proper ones.

How in the damn hell am I supposed to use my beloved 'small. 'ésmall. tags with this shitÉ!É.
posted by Alvy Ampersand at 9:45 PM on November 29, 2009


I found the experiment really interesting on a personal level, mostly because of what it said about me and my susceptibility to groupthink. Turns out, every time I looked at a comment I really liked, particularly in argumentative threads, I'd check the (now nonexistent) favorite count to see if other people liked it too, and I realized that I was checking to see if my opinions matched the "group consensus", and was asking myself "Did enough other people like this comment that I can feel OK about agreeing with the sentiments?" I'd never noticed that I did this before, and I'm glad I got some self-awareness out of the experiment. (I don't know if anyone else had a similar experience.)

Also, judging from the responses in this and previous threads, you might solve some future UI issues by giving three options which should cover most everybody:

Show favorites with counts     [10 favorites +]
Show favorites without counts  [has favorites +]
Don't show favorites           [+]

(People can still argue about which one gets to be the default.) Also on preview, Alvy can use some of my <small> tags if he likes.
posted by Upton O'Good at 10:02 PM on November 29, 2009 [4 favorites]


In the original spirit of the thread, here's my impressions of the overall impact:

I liked the pre-November favourites system. I didn't opt out of the experiment. I tried not to let it affect the amount of favourites I gave to other people, but haven't checked my stats to see if I succeeded in that or not.

I found it very annoying to do without the feedback that favourite counts give. The best explanation of how I feel is from the previous thread: it's like losing the non-verbal communication in a conversation, the nodding heads, and "haha", and frowns that register how the general flow is going. Some people might not miss or be interested in it, but for me it's what makes Metafilter more engaging and community-oriented than other sites. I feel tone-deaf without this little backchannel of information.

I don't believe that the general tone of conversation has changed by removing the favourite counts. I still saw thread-shitting and am not convinced that it's caused or encouraged by favourites. I'd be interested in stats on whether there've been more/same-number-of/fewer early-thread-shit deletions during November.
posted by harriet vane at 10:30 PM on November 29, 2009


I like the new way, though I think it might be useful to only show [has favorites] for comments that have at least one favorite from someone other than the original commenter.
posted by syzygy at 3:46 AM on November 30, 2009


When I remember that favorites exist, sometimes I favorite a comment I really like to show the commenter that he/she done good, other times I use favorites as a bookmark, and once every blue moon I might notice that a comment or post has a lot of favorites and click on it. This usually happens when I see it on my sidebar. Otherwise, I'm not working up a lather about this experiment. Thank you mods for your unflagging attempts to improve an already wonderful site. We'll get through this and all will be well again, especially if we watch our manners.
posted by Hobgoblin at 5:34 AM on November 30, 2009


I found it very annoying to do without the feedback that favourite counts give. The best explanation of how I feel is from the previous thread: it's like losing the non-verbal communication in a conversation, the nodding heads, and "haha", and frowns that register how the general flow is going. Some people might not miss or be interested in it, but for me it's what makes Metafilter more engaging and community-oriented than other sites. I feel tone-deaf without this little backchannel of information.

This may, perhaps, come down to user personality. I feel more comfortable in smaller groups (think, "dinner conversation"), so the notion of having people going 'yeah!' or 'boo!' or snickering from the sidelines is extremely unappealing. I personally feel that it makes nuanced conversation more difficult, like, 'quit yelling out numbers, I'm trying to keep track of my beans here!' Hah.

It's more difficult to say what happens at a community level, which may be the entire reason for this experimentation. (Whether 'asked' or 'observed', answered or addressed, the scientifical nature of this experiment, suppositional trial, principle test, or whatevs, becomes a lapsed issue around the fortieth or so iteration, yeah?) It's clear that visible favorite counts highlight certain parts of the conversation, and those highlights tend to be at the extremes of opinion (or sides of an argument) in a thread. Some people think this makes the thread a bit more contentious and fighty, and some people think this makes the thread more vibrant-- two sides of the same coin.

The fact is, contentious fightyness makes it more difficult for individuals (or for mods) to predict and influence the tenor of conversation across the community. Moderate, reasonable comments are less likely to get favorites because they're less likely to make someone give it a 'heck yeah!' On the other hand, the illusion of isolation (where no 'me too' lurkers are visible) may also have an influence. People tend to forget that they're in a public space, where anyone can read what they're writing; however, the continuing personal nature of some of the commentary (and, it seems to me, most of the most-favorited commentary) would seem to indicate that faux privacy, or the lack of, doesn't have a huge influence. To take that a step further, the tenor of the community seems to be moderate enough for people to feel comfortable sharing personal things.

My take on the November experiment is that I, personally, feel more relaxed about my MeFi experience without the visible favorites, and that may be the biggest influence on my perception that the tenor of conversation is also more relaxed and smooth. Honestly, I think the only people in the position to determine whether overall problematic (delete-worthy) comments have changed are the mods. It's interesting that this whole snafu seems to have spawned a way to appease those who don't prefer to see favorites by way of appeasing those who do prefer to see favorites-- the on/off switch.

Given the way favorite counts can highlight the extremes (and I do think it's a given), and given that we are probably going to see favorites stay the way the were before the experiment, it may be worth the community's while to consider ways to keep the tenor of conversation such that people feel comfortable, rather than arguing about whose way sucks more. Should we drift toward a harsher community environ, how could the community as a whole (rather than the mods only) help to correct the course?
posted by zennie at 6:44 AM on November 30, 2009 [3 favorites]


I think there's nothing really unreasonable about saying that opting out quickly reduces the depth of feedback about the experience of the change itself that the person opting out can offer

Not sure I'd agree with that. It sounds like a reasonable statement, but it seems like assumptions are being within it. Simply spending more time with something isn't a guarantee of having more insight with it, especially when it comes to be the variety within people. If a non-insightful person spends more time with the change than an insightful person, which opinion are you going to give more weight to?

Hopefully you'll take in both opinions without trying weigh one against the other, which is the gist of the point I was arguing with P.O.B. Rather than viewing the opinions of who opted out as less than those who stuck with for ___ (where's that line drawn?), perhaps see it as just another aspect of the great experiment i.e. are they opting out just because it's change or are there other reasons given and if so how do those reasons factor into the equation?
posted by Brandon Blatcher at 7:42 AM on November 30, 2009


Simply spending more time with something isn't a guarantee of having more insight with it, especially when it comes to be the variety within people. If a non-insightful person spends more time with the change than an insightful person, which opinion are you going to give more weight to?

I won't argue against this for specific hypothetical cases, sure. But I don't think those specific cases make the general statement any less reasonable. The average insight on that specific subject of someone who spent more time with it is almost certainly greater than the average insight of someone who spent less time with it, regardless of whether there's some genius opters-out in the mix and some thickos sticking with it, especially since by the same reasoning there are thickos and geniuses on the out and in sides respectively as well.

(Not that I think anyone in particular is a thicko in this context, but "thicko" is very fun to say. It sounds like a colorful modeling clay.)

At the root of it is this: those who spent a significant time with the change can tell us about their experiences with it. I'm glad for that, and value that. I don't disregard the thoughts of those who opted out quickly and think it's all a valuable part of the discussion, but the folks who opted out quickly by definition did not spend significant time with the change, and so "how did spending time with the change go for you", specifically, is not a question they can really answer in the same way as the folks who did. Insightfulness of any given observation is a side issue.

Hopefully you'll take in both opinions without trying weigh one against the other, which is the gist of the point I was arguing with P.O.B.

Absolutely, and I didn't really think it's a question that even needed to be asked at this point. That may be where some of the friction earlier was coming from; I can't speak for P.o.B., but from my perspective having been mostly quietly observing this thread (and this goes for my reaction to some of the stuff I saw in the original as well), it felt at times like there was some suggestion that we were out to marginalize or otherwise suggest a going and fucking of one's self to anybody who opted out, that we weren't interested in their opinion at all simply because were were interested specifically in the experiential reports of folks who spent a lot of time with the change.

That's an glommed together impression, not an attempt to characterize any particular person, and it reads more strongly than anything I think you've said in this thread, but there is that feeling in part in how I've read some of the reactions. A kind of ungenerous assumption that we were going to be bastards about this in a way that kind of shocked me because, seriously, when have we ever been bastards like that? Why would "you're not just going to ignore us or turn this into a fight" ever be a question that made sense, at least from folks who have been around a while?
posted by cortex (staff) at 8:06 AM on November 30, 2009 [1 favorite]


BB, nobody's trying to discount your opinion. With regard specifically to how this test affected individual users' interactions with the site, it follows that people can only testify to how the change affected them for the duration that they were exposed to the change, and when they participated longer, they may have experienced something different due to longer exposure (and indeed some people have reported just that). Conversely, for the period people didn't participate, people can still voice an opinion about why they stopped participating, and people can still have an opinion about how the test affected MeFi as a whole. Or they can still have an opinion about how simply knowing about the test affected their interaction with the site. In any case, I don't see where you are getting the idea that anyone's creating discrete worthy/unworthy categories or deriving weight functions.

[On preview, what cortex said. But hey, I've already typed this up.]
posted by zennie at 8:22 AM on November 30, 2009


My two cents - I installed the plug-in that got rid of favorites entirely and I love it! I still use favorites to bookmark but now I don't have that cognitive dissonance where I'm all confused about why some really incendiary, mean-spirited comments get any favorites at all!
posted by muddgirl at 8:27 AM on November 30, 2009 [2 favorites]


I won't argue against this for specific hypothetical cases, sure. But I don't think those specific cases make the general statement any less reasonable.

Why deal with this in the hypothetical at all? We've heard from numerous people in this thread and the other that they were initially opposed to the change and found it jarring... and that over the course of the month their perspective changed, contrary to their expectations. This is why we do experiments. If you could accurately predict your own changes in view... you'd probably have those views in the first place.

But why take anyone's word for it? You hated brussel sprouts as a kid and you'd probably hate them now. Don't try them. They suck.
posted by Durn Bronzefist at 9:01 AM on November 30, 2009


Fun fact: I just went over to the brussel sprouts camp last month. The trick is grilling them.
posted by jessamyn (staff) at 9:03 AM on November 30, 2009


Seriously? Hmm. Never tried that, and I do love the grill.

I steam them with butter, or melt aged cheddar over them, and am going to try baking them with brie shortly. To be honest, I think there's probably some biological change at work, because I just recall them being bitter bitter bitter as a kid, and I don't think melted butter would change that much. But about every five years I'd take one bite of them to confirm I hated them and that was that (until maybe 6 or 7 years ago). Them good eats.

Course, my appreciation for bitterness has also changed, itself (after-work stout here I come).
posted by Durn Bronzefist at 9:09 AM on November 30, 2009


but the folks who opted out quickly by definition did not spend significant time with the change, and so "how did spending time with the change go for you", specifically, is not a question they can really answer in the same way as the folks who did.

On this, I cheerfully agree. As for the rest, we'll just have to agree to disagree, but as always, cheerfully.

(Not that I think anyone in particular is a thicko in this context, but "thicko" is very fun to say. It sounds like a colorful modeling clay.)


Bubber is light modeling clay, not sticky, doesn't dry out, good for light, fun stuff.


BB, nobody's trying to discount your opinion

Oh, I'd disagree with that. Those who dislike favorites seem to be trying pretty hard at times to discount other opinions.

You hated brussel sprouts as a kid

I've always liked them, they're little bundles of heaven and go well with chicken. They should never be broiled, just sauteed in olive oil with salt and pepper.
posted by Brandon Blatcher at 9:13 AM on November 30, 2009


I've done the same thing as jessamyn. My wife grilled some up in olive oil and some salt and a couple other spices and I was like, holy shit, this is a salty grilled thing! So me and brussels are on good terms now.

I have shaken a number of my old food dislikes over the years. I still hold a grudge against pumpkin, though, one that I'm stubbornly willing to cling to.
posted by cortex (staff) at 9:14 AM on November 30, 2009


What have you done with it? And how do you feel about the various kinds of squash?
posted by Durn Bronzefist at 9:16 AM on November 30, 2009


thicko

Yeth?
posted by dirtdirt at 9:35 AM on November 30, 2009


Zennie's experience was very similar to mine: I, personally, feel more relaxed about my MeFi experience without the visible favorites

Fun fact: I spent at least ten long seconds trying to figure out how jessamyn had "gone over to the brussel sprouts camp." Actual thoughts that went through my mind: "That's a funny type of camp." "She wasn't a counselor or a camper, she just went over and visited?" "Isn't it the wrong season?" "The brussel sprouts camp? How does everybody know about these things except me?"
posted by salvia at 9:50 AM on November 30, 2009 [4 favorites]


I chop 'em in half, toss them in a bowl with olive oil, sea salt and garlic, and then place on a cookie sheet. Bake for 20 minutes or until the bottoms are nice and caramelized. Brussel sprouts become salty, sweet morsels of love.
posted by Blazecock Pileon at 10:04 AM on November 30, 2009


My experience, as someone who did not disable the change:

It felt like an awkward compromise at first, then rapidly grew on me, then started feeling like an awkward compromise again but for different reasons.

For the record, I was one of the strident anti-favoritists when the feature was first introduced. Many of the predictions made at the time -- that some percentage of people would become more concerned about scorekeeping than about content, that favorites would become a de facto filtering system by which users would ignore parts of the conversation, that people would treat favorites as an indicator of quality, etc -- have in fact come true (as Joseph Gurl's summary above makes abundantly clear). I still think the site would be better off if they had never been introduced. But the ship's clearly sailed on that one, it's a waste of breath to keep arguing about it; as long as they're here, yes, I use them too. And, frankly, I've come to agree that they often are a reasonably good indicator of quality, and I get a charge out of collecting karma points as much as the next guy.

So that's where I'm coming from on this.

With this experiment in place, I found that when I see favorites on other peoples' comments, it doesn't matter to me at all how many of them there are; whatever hint of 'this is more worth reading than that' is just as adequately expressed by the existence of any favorites at all than by an exact count.

By contrast, when I see favorites on my own comments, I every single time have the narcissistic impulse to click through to see how many of them there are and who they came from. And since the experiment's behavior is to show you a list of names rather than a plain number, I've discovered that who favorited something is a lot more revealing and interesting than how many, which has gradually led me to clicking more often to see who favorited other peoples' comments as well. This has caused me to pay more attention to who is speaking, rather than just on what they're saying; some individuals have popped out as people I consistently agree with or generally disagree with, and I end up paying more attention to those individuals as a result.

(Whether this is ultimately a good thing or a bad thing in terms of overall quality of conversation, I could really argue either way.)

So now it feels awkward again, because while I still don't care how many favorites a given comment has, I do now care who's doing the favoriting.

My perfect-world implementation would now be something like a small icon next to any comment that has any favorites, which on hover would show a little popup panel containing a list of names. (It could be all AJAX-y and have a dropshadow. Dropshadows are awesome.) The length of this list could be capped at some reasonably small number: once something has more than a dozen or so favorites it no longer really matters who they came from, it just means a network effect has kicked in.
posted by ook at 10:16 AM on November 30, 2009 [1 favorite]


I've kept the new setting the entire month, and I can say with some confidence that I don't really give a damn either way.
posted by wonderchicken at some time or other, days ago now

I guess you can call me a StavrosTruther. As ever.
posted by Kwine at 10:18 AM on November 30, 2009


I have not turned favorites "back on" for the month. I still see "has favorites". My own behavior re: how I use favorites (bookmarks, reference, clean them out pretty regularly, don't skim threads based on favorites) has not changed a bit. I have noticed, however, that fighty threads don't make me nearly as anxious/mad/bitchy-feeling/fussy as they once did, and I think it's because I don't see a number associated with the favorites given to a snarky or fave-whoring-type comment. Somewhere in my mind I feel like "it could be one, it could be a hundred, who cares, lalala". I would be really interested in whether that kind of commenting was off-putting to new users, or perhaps creates a barrier to participation that isn't really in the spirit of MetaFilter. Or maybe so many people are used to the Reddit-type "eat or be eaten" particiaption (my own perception) that they don't care. I have to assume it's probably some of each, though.

This is all based on my OWN experience and perception - I'm not drawing any conclusions about anyone on any side of the conversation (I don't even want to say debate). I think between this thread and the monster thread that kicked off the experiment, there may be a somewhat uneasy consensus that yes, favorites do influence some stunty or snarky behavior, but that there are a lot of people who use them as a utility to guide their interaction with the site, and that for some people they create a feeling of community consensus that is valuable to them. It seems like there might not be a way to ameliorate the impact that favorites have on the first kind of user without seriously impacting the experience of the other kinds of users. From my own standpoint, if favorites aren't going to be surgically removed (including any workarounds that allow people to retain them) I'd be peachy keen with the option to remove the favorite counts and the "has favorites" indicator as a user preference.
posted by ersatzkat at 10:44 AM on November 30, 2009


An easy recipe is to half-them, heat up a few tablespoons of olive oil, one pressed garlic clove, a tablespoon or so of lemon juice, and a teaspoon or so of honey in a skillet with a tight fitting lid. Toss in the sprouts, which will sautee while steaming in the closed pan, and turn every few minutes. They're done when browned to suit. Use salt if you must.

This is basically what I do, except I use butter instead of olive oil because, y'know, butter.
posted by dersins at 11:06 AM on November 30, 2009


I retry brussels sprouts every few years to see if I've developed a taste for them yet -- because they seem like something I would like, and I like nearly every other veggie. Hasn't worked thus far. And I'm sorry, but you could make a gym shoe or a big hunk of dried earwax tasty by dredging it in olive oil or butter, salt/spices and then grilling or roasting it until caramelized.

Steam five sprouts without any adulteration, eat 'em, and then tell me if you like brussels sprouts.
posted by FelliniBlank at 11:18 AM on November 30, 2009


...and do grilled or roasted brussels sprouts still taste like cooked cabbage? Cause that's what I don't like about them...
posted by ersatzkat at 11:20 AM on November 30, 2009


and do grilled or roasted brussels sprouts still taste like cooked cabbage

I love brussels sprouts but could easily live the rest of my life without eating cabbage. To me, sprouts have a crisper taste and feel when they're grilled or roasted as opposed to soggy cabbage.
posted by Brandon Blatcher at 11:31 AM on November 30, 2009


Do this with most food and see if you still enjoy eating it.

Well, you can do this to most veggies and enjoy it. But just because something takes a bit of preparation doesn't mean that it's no good at all.

Also, on the subject of roasting: I love corn on the cob, and I love grilling things. I've been to countless (ok, maybe 6) corn festivals, and I will never understand the appeal of a grill-blackened cob. Two rights sometimes make a wrong.
posted by Durn Bronzefist at 11:42 AM on November 30, 2009


Speaking of self-favoriters, on profile pages, instead of "Favorites: ... & Favorited by others: ...," it seems "Favorites given & Favorites received" would be more precise.
posted by Pronoiac at 12:43 PM on November 30, 2009


Steam five sprouts without any adulteration, eat 'em, and then tell me if you like brussels sprouts.

Yes. I microwave them. One of my other favourite snacks is a plain microwaved potato.
posted by jacalata at 1:33 PM on November 30, 2009


Fresh brussels sprouts, halved, steamed just so, and lightly buttered and peppered: heaven on a fork.

Instant Sprouts: Glass bowl, a few tablespoons of water at the bottom; add pile 'o split sprouts; loosely cover and microwave on high 2 to 4 minutes. Should be tender but still bright green. Butter, pepper, or salt (or nothing) to taste. Mmm...
posted by zennie at 1:35 PM on November 30, 2009


And I'm sorry, but you could make a gym shoe or a big hunk of dried earwax tasty by dredging it in olive oil or butter, salt/spices and then grilling or roasting it until caramelized.

So, if I served this preparation to you, you would eat it? I'll even throw in letting you pick the gym shoe and/or earwax of your choice.
posted by Blazecock Pileon at 2:41 PM on November 30, 2009


If he eats his gym shoe, I'll eat my hat.
posted by cortex (staff) at 2:54 PM on November 30, 2009


And I'm sorry, but you could make a gym shoe or a big hunk of dried earwax tasty by dredging it in olive oil or butter, salt/spices and then grilling or roasting it until caramelized.

See, I thought I hated parsnips until I had them made so they were all carmelized in butter with onions. Mmmmm parsnips. But I'm not going to say that I don't like parsnips just because I don't like them when they're just steamed plain, or whatever it is one does to parsnips. I do like parsnips - just prepared in a particular way. I don't like cooked carrots, mostly, but I like them raw, or roasted with other root vegetable (olive oil, salt, pepper, etc.).

Maybe you don't like brussels sprouts at all, no matter how they're cooked. That's okay. More for the rest of us!
posted by rtha at 2:59 PM on November 30, 2009


Steam five sprouts without any adulteration, eat 'em, and then tell me if you like brussels sprouts.

Steam a chunk of cow without any adulteration, eat it, and then tell me if you like steak.

Steam some dough, tomatoes, and cheese without any adulteration and then tell me if you like pizza.

Steam a... well you get the point.

I hope.
posted by dersins at 3:06 PM on November 30, 2009


Your point seems to be that steam is boring. But have you ever coated steam in some olive oil and sauteed it with some muddled sun-dried tomatoes? Because that will blow your mind.
posted by cortex (staff) at 3:13 PM on November 30, 2009 [3 favorites]


So things-that-are-not-vegetables tend to not taste great on their own, steamed? Intriguing.

So, back to brussels sprouts...
posted by Durn Bronzefist at 3:25 PM on November 30, 2009


Rough-chop some chorizo. Pan sear with halved brussels sprouts. Eat with some good bread to sop up the last of the chorizo grease. Yum.

Steamed sprouts is fine too, as long as you don't overcook them (slimy) or undercook them (bitter). Tastes nothing at all like cabbage if you get the timing right (which admittedly is a bit tricky.)
posted by ook at 3:47 PM on November 30, 2009


On second thought, cabbage may also taste nothing like cabbage if you get the timing right. I'll have to try that one of these days.
posted by ook at 3:52 PM on November 30, 2009


If he eats his gym shoe, I'll eat my hat.

Hey, if it's good enough for Werner Herzog, it's good enough for me. I've voluntarily eaten Slim Jims, which have to be far less nutritious and appetizing than a nicely roasted olive-oily Keds. Anyhow, I tend to eat veggies (raw or cooked) fairly plain except for salad, and that still tastes nice without the dressing. I must just be doing something wrong with the sprouts because pretty much all forms of cabbage and leaves and greens are delightful. Maybe roasting or grilling is the answer, which is a pity since I neither roast nor grill although I do broil.

Now white "chocolate," that's just gross.
posted by FelliniBlank at 4:07 PM on November 30, 2009


My first ever experience with brussels sprouts (at age 25) was just boiling them for a few minutes (not quite steaming, but boiling generally isn't considered to produce tastier results in most cases), and then added a little butter (because everything tastes better cooked with butter).

They were delicious. I decided they were a reasonable dinner green veggie (along with broccoli).
posted by that girl at 7:53 PM on November 30, 2009


My two cents: About half the comments in a randomly selected nontrivial thread will "have faves", which means the indicator tells you whether a comment is in the fiftieth percentile. I checked 8 threads and 7 of them were in the range of 50-60% favorited. A lot of those have only one favorite which is meaningless because it doesn't aggregate information from mulitple users, so it's very noisy.

The problem is that you are telling us a statistic about faves but it's not informative enough to be useful except as a very noisy predictor of comment quality. If it were less noisy then I could use it.

To me, the "has faves" indicator in the thread could be useful for:
- Finding really interesting comments in a thread I'd otherwise skip
- Finding worthwhile comments in a noisy thread
- Finding the best comments in a big thread I get to late

Is knowing that a comment is better than median useful for these tasks? Not for me: the top 50% of comments are not going to be that great, and in a big thread it's still more than I'm willing to read (there've been several 100+ comment threads this week, and I'm either going to read the whole thing or the top 10; why would I read the top 50?).

If you instead showed an indicator for the top X% of comments (where X is 10 or less), conditioned on there being a good number of favorites in the thread (i.e., your confidence interval is small), then it would be useful to me.
posted by jewzilla at 10:27 PM on November 30, 2009


It's now officially December 1st in Pacific Time and we've gone back to the old style of favorites as the default view. We've also added a new preference setting called Comment favorites style. You can choose between three different options: 1.) Show Favorite Counts, 2.) Show "has favorites", and 3.) Hide Favorites. Option 1 is the default, and is exactly the same as before the November exercise. Option 2 is almost exactly like the view during November with the addition of an HTML title that gives the favorites count as you hover over the link in many browsers. And Option 3 is new—you can completely remove counts and any sort of favorites indicator from your view.

As with any of the display preferences, they're based on cookies so you'll need to re-save your profile preference in each browser you use. If you spot any problems with the new system, let us know here. Just keep in mind that it's late here in Pacific Time and I'll fix up any bugs tomorrow. Thanks.
posted by pb (staff) at 12:03 AM on December 1, 2009 [18 favorites]


NaFaFuMo comes to an end!
posted by fleacircus at 12:13 AM on December 1, 2009 [1 favorite]


You guys are awesome, pb. Thank you for taking the time to read our comments and incorporating several options. Way to go, metafilter! :0)
posted by thisperon at 12:23 AM on December 1, 2009 [1 favorite]


And count me in as somebody who really appreciates option three. I feel like I'm reading "pure comments" without the "and what does everyone else think of this?" noise in the back of my head.
posted by thisperon at 12:32 AM on December 1, 2009


I find this really interesting. What I interpret as nodding heads, and "haha", and frowns is seen by other people as 'yeah!' or 'boo!' or snickering from the sidelines (quoting zennie here, but I don't think it's an unusual interpretation).

I really just don't see it as being so emphatic or noisy or adversarial as all that. I like small group conversations too, but body language is an important part of them in person, and I need a substitute for that when I'm online, I guess. It's a relief to me to have the little numbers back.

Thanks, mods, for giving something new a try, and for making 3 options for people from now on. I appreciate the efforts you put in to make the site a fun place for everyone.
posted by harriet vane at 1:17 AM on December 1, 2009


Option 3 is new—you can completely remove counts and any sort of favorites indicator from your view

I'm very pleased to have this - thanks!
posted by tomcooke at 1:40 AM on December 1, 2009


Option 3 is interesting, might try that for a bit.
posted by Brandon Blatcher at 2:12 AM on December 1, 2009


Yay, welcome back favorites counts!
posted by ThePinkSuperhero at 4:25 AM on December 1, 2009 [1 favorite]


How come you guys didn't just skip the "experiment" and implement the 3 option idea from the beginning?
posted by gman at 4:40 AM on December 1, 2009 [1 favorite]


How come you guys didn't just skip the "experiment" and implement the 3 option idea from the beginning?

They wanted to irritate you, gman. Duh.
posted by flapjax at midnite at 5:02 AM on December 1, 2009 [4 favorites]


I think Chairman Mao answered that for us in his seminal (cum again?) Combat Liberalism:
But liberalism rejects ideological struggle and stands for unprincipled peace, thus giving rise to a decadent, Philistine attitude and bringing about political degeneration in certain units and individuals in the Party and the revolutionary organizations...

To indulge in irresponsible criticism in private instead of actively putting forward one's suggestions to the organization. To say nothing to people to their faces but to gossip behind their backs, or to say nothing at a meeting but to gossip afterwards. To show no regard at all for the principles of collective life but to follow one's own inclination...

To let things drift if they do not affect one personally; to say as little as possible while knowing perfectly well what is wrong, to be worldly wise and play safe and seek only to avoid blame...

Not to obey orders but to give pride of place to one's own opinions. To demand special consideration from the organization but to reject its discipline...

To indulge in personal attacks, pick quarrels, vent personal spite or seek revenge instead of entering into an argument and struggling against incorrect views for the sake of unity or progress or getting the work done properly...
Only in the fiery crucible of praxis can we forge the ideology that will let Chairman Maothowie's name take its rightful place alongside his predecessor in the pantheon.
posted by Abiezer at 5:15 AM on December 1, 2009


I want it back I want it back I want it back

So you start reading a comment and your eye can't help but keep on slidding down to the favourite count, making you incredibly aware that the comment you are reading has 6 favourites. "wow that's an interesting point your making here {6 FAVOURITES} although I think you might have misunderstood {6 FAVOURITES} some of the {6 FAVOURITES} facts..."

I agree that having to click on the favouite link to see how faved a comment is does highlight the favourite count even more so, but I know feel like I'm having favourite count shoved in my face whether I like it or not.
posted by litleozy at 6:11 AM on December 1, 2009 [1 favorite]


Just saw pb's comment. Feel sheepish.

Thank you pb that is awesome.
posted by litleozy at 6:11 AM on December 1, 2009 [1 favorite]


Hey cool, new options. Thanks, Mods. I can now enjoy another month of experimentation with out having to break out test tubes and scalpels and small flasks of colorful mysterious goos.

I think there was real benefit in the experience/experiment. I think we all learned a lot - if not about our own individual (non)use of the feature, then about the variety of ways people use and perceive the feature - to approve/agree (and by extent to disapprove/disagree of a comment positing the opposite), to bookmark, to encourage/reward bravery or quality, and to say "ha ha!" or "I appreciate this shared pop-cultural reference" or "I was thinking that."

Plus: many people were reintroduced to contacts and others found folks whose pattern of contributions (in favoriting and in posting) who appealed to them as a result of this experience. It seems to me that this ended up with a net good for the site even if it was a little painful to get there: it highlighted the diversity of site usage; some tighter connections between users have been formed, and people have more ways to filter the site to make it a more enjoyable experience for individuals.
posted by julen at 6:33 AM on December 1, 2009


Yes, can we publicise how to change back to experiment mode in the top bar/sidebar?
posted by cillit bang at 6:35 AM on December 1, 2009


This level of customer service for a one time fee of $5? Please start a cell phone company.
posted by vapidave at 7:02 AM on December 1, 2009 [8 favorites]


Yes, can we publicise how to change back to experiment mode in the top bar/sidebar?

I think we'll do some stuff today, yeah. We're in the process of waking up and facing the day, etc.
posted by cortex (staff) at 7:07 AM on December 1, 2009


Yes, can we publicise how to change back to experiment mode in the top bar/sidebar?

I'll toss it on the sidebar now that I'm up.
posted by jessamyn (staff) at 7:08 AM on December 1, 2009


Wow. No visible favourites -- this is awesome. Thanks for making it an option.
posted by Durn Bronzefist at 7:16 AM on December 1, 2009


I hated (and still hate) "has favorites," because of the combination of visual clutter and flattening of information. But I'm going to experiment for a while with the "no favorites" option -- I don't know if I'll make it a few hours, a few days, or if I will turn out to love it.

So, thanks for giving the choices in this way; it's awesome and it'll be fun to see how this evolves.
posted by Forktine at 7:33 AM on December 1, 2009


Yeah, what the Fork said.
posted by Brandon Blatcher at 7:34 AM on December 1, 2009


Great work, Team Mod! Thanks for bringing back the proper mode for fans of favorites, and keeping the hybrid as well as adding the total removal option for those who just can't stand them.

I look forward to never hearing the anti-favorites complaint and never having this argument ever again. LONG LIVE METAFILTER.
posted by EatTheWeek at 7:42 AM on December 1, 2009


I look forward to never hearing the anti-favorites complaint and never having this argument ever again.

I find your lack of pessimism disturbing.
posted by FishBike at 7:45 AM on December 1, 2009 [1 favorite]


Could we have an Option 4: Extra-spicy szechuan favorites with egg drop soup?

Thanks a lot for the choices -- Option 3 is exceedingly groovy so far, making MeFi even sleeker and more aerodynamic.
posted by FelliniBlank at 7:49 AM on December 1, 2009


Thanks for the no visible favourites option. May I suggest, now that it is something that you have to turn on rather than the default, that we go with something shorter for option 2. Either a single character, I'd suggest $ but * would probably be less controversial; or a single preferably short word.

If the first line could look like "posted by Foo at 12:01 AM on December 1 [* +] [!]" or "posted by Foo at 12:01 AM on December 1 [+] [!]" depending on whether a comment has a favourite or not.

I'd probably go with option 2 instead of 3 if the favourite indicator didn't take up so much room.

EatTheWeak writes "I look forward to never hearing the anti-favorites complaint and never having this argument ever again. LONG LIVE METAFILTER."

I don't see how this option effects the belief that favourite seeking behaviour is bad for metafilter and results in noisy comments being posted to garner favourites. Especially considering visible favourites is the default.
posted by Mitheral at 7:50 AM on December 1, 2009


Begun again, this favorite war has.
posted by Brandon Blatcher at 7:53 AM on December 1, 2009 [2 favorites]


I'm going for option 3 for a while. I might change it back to 1 or 2 at some point, but 3 is the most interesting to me.
posted by Kattullus at 8:00 AM on December 1, 2009


I think I'll shuffle around between 1 and 3 on various machines and totally forget about the settings. This way I'll log in one day in the near future on my laptop or at work or wherever and go huh, slow day on MeFi, no one's favorited anything!

I like confusing myself that way. In moderation.

Not too much moderation, of course.
posted by goodnewsfortheinsane at 8:00 AM on December 1, 2009


Thanks for all the new options, y'all! It was quite jarring this morning to see favorites counts again. I think I might try option 3 for a while to see how it affects the way I use the site.

Thanks for the no visible favourites option. May I suggest, now that it is something that you have to turn on rather than the default, that we go with something shorter for option 2. Either a single character, I'd suggest $ but * would probably be less controversial; or a single preferably short word.

Can we just let this thing be for a little bit? I feel for the mods; it's as if anything they try (which is far above and beyond what I'd have the patience for) isn't ever quite good enough. I sometimes imagine them having to walk away from their computers and scream for a bit.
posted by runningwithscissors at 8:07 AM on December 1, 2009 [2 favorites]


I reserve the right to tell people that I have favourited their comment with extreme prejudice.

And if you dispense with them altogether some day, that I have favourited their comment in my mind.
posted by Durn Bronzefist at 8:11 AM on December 1, 2009 [1 favorite]


[n+]
posted by klangklangston at 8:12 AM on December 1, 2009


FishBike, I think you meant:

I find your lack of hate disturbing.

You know, for euphony.
posted by cgc373 at 8:29 AM on December 1, 2009


yay for my smug sense of superiority for having stuck with a thing I hated for a month!
I hear that from so many of my exes!
posted by Abiezer at 9:02 AM on December 1, 2009 [2 favorites]


This is a sad day indeed. I SHALL NOT REST UNTIL ALL ARE COMPELLED TO SHARE MY EXACT EXPERIENCE OF METAFILTER.




This includes needing to comment informatively on all posts (Front Page, AskMe and Meta) or else a fine shall be imposed (or favorites deducted).

It's only fair.
posted by philip-random at 9:04 AM on December 1, 2009


I am far, far too lazy to change my settings, so I stuck with the experiment all month. My reaction to the "has favorites"option as a pretty much strict AskMe user was as follows:

Eh.

It wasn't horrible, and it came nowhere close to penetrating my Laziness Force Field and impelling me to opt out, but it also wasn't good enough that I'm going to go opt back in.

It did have some interesting effects, though. Looking at favorites I got, it appears that while a not-unusual number of my comments got favorites, each one got way less than average. A quick eyeball of my favorited comments indicates that having only a single favorite is not at all the norm, whereas in November only 2 of 9 comments got more than one. That is interesting, in that it seems to support the idea that favorites have a pile-on effect (lots of favorites gets you even more favorites) and it's not awesome imo, in that I have no idea if any of those comments were particularly good or if they just happened to make one random person happy. (This is anecdata at best - see above, re: lazy.)

It didn't seem to affect the number of favorites I gave out all that much, though. I'm not much of a favoriter - I don't use them as bookmarks or anything, mostly because I'm too conscious of the implied approval they give, so I don't seem to give out more than 5 in a month anyway.
posted by restless_nomad at 9:04 AM on December 1, 2009


I don't see how this option effects the belief that favourite seeking behaviour is bad for metafilter and results in noisy comments being posted to garner favourites. Especially considering visible favourites is the default.

You've been here long enough to know that before there was favourite seeking behavior there was plenty attention- status- infamy-seeking behavior. I agree with koeselitz in the bigass 0ct. 31 MeTa that loudmouth lookitmeism was just as bad - if not worse - back in the good ol' days, where egos and mouths ran rampant. Not crediting the introduction of favourites with the decline of that sort of thing (More like those slightly-above-average-sized fish realized their puddle was becoming a pond, then a lake, and then an ocean, and not liking how that scaled one little bit), or claiming there would be a resurgence in it if favourites were eliminated, but ultimately, people are the problem.

Seriously, Option 3 is like a movie theater without cell phones, a hiking trail in a national park without chatty picture takers, a skyline without smog...

See above. Favourites don't cause asshattiness, asshats cause asshattiness. You don't like the system or find it useful, fine. This seems like a great way for everyone to get what they want, but ultimately the fault, dear Burhanistan, is not in our [+] but in ourselves.

If I can't have my childhood fantasy of a post-Rapture empty city all to myself, then this is the next best thing.This would probably be closer to the reality of having an empty city all to yourself. After the novelty wears off, I doubt it's really all that keen.
posted by Alvy Ampersand at 9:10 AM on December 1, 2009 [3 favorites]


That's too dismal, Alvy. I imagine the post-rapture city would be more like this.
posted by Meatbomb at 9:25 AM on December 1, 2009 [2 favorites]


AHH this looks WEIRD
posted by azarbayejani at 9:39 AM on December 1, 2009


This would probably be closer to the reality of having an empty city all to yourself. After the novelty wears off, I doubt it's really all that keen.

Ouch.
posted by Blazecock Pileon at 10:05 AM on December 1, 2009


I should probably clarify that I don't think MetaFilter would be better off without Burhanistan or vice versa, and I certainly wasn't inviting him the Hit The Button. I do think his scenarios and fantasy are completely impractical and in a lot of ways antithetical to the ideal I think MeFi aspires to, essentially a community that has hundreds upon hundreds of active members interacting with one another as individuals. It's certainly messy and aggravating at times, but the alternative view offered seems pretty stagnant and isolating.

The idea that the jerks in the movie theatre or sightseeing goobers would STFU if someone took their toys away is off-base too.
posted by Alvy Ampersand at 10:31 AM on December 1, 2009 [1 favorite]


So if you're going with Option 3 of no favorites you might want to stop reading this now and move to the next comment. It's ok. I'll wait. . . . . Ok, so on the extreme end, those of us who used the CommentFilter script to sort a page by the number of favorites will be happy to know it works again and can be found here.
posted by jeremias at 10:44 AM on December 1, 2009


I just created a Greasemonkey script to display the favorites the way I think would be most useful. "Liked", "loved", and "best".
posted by Plutor at 10:48 AM on December 1, 2009 [1 favorite]


I'm surprised the mods didn't delete that.

It's been a really busy morning and the comment had only one flag on it, which under the best circumstances is kind of a weak draw for Metatalk content. That said, looking at it now: Alvy, that does come off as kind of a cheap shot, even if you didn't intend it that way. Not so much of that, maybe.

THAT said, I think having discussed it now and it not being some kind of deep-search hatchet job, it'd make more sense to leave it than nix it and the followup reactions and create a conspicuous hole here, so I'll leave it be.
posted by cortex (staff) at 10:55 AM on December 1, 2009


Sweet! I'm going to turn off favorites right now. Everything everyone says is my favorite!
posted by fuq at 11:09 AM on December 1, 2009 [1 favorite]


I initially opted out, thought better of it, and went with the experiment for its course.

A fair amount of the time during the experiment, I would click through on the "has favorites" link to see the names (and, in a histogram-sort of way, a visual estimate of the count). But not always, and it didn't feel like a burden.

Today, however, when I saw the favorites numbers again, my mind said "yay!" and I realized how much I had missed them.

Summary: MetaFilter works for me without a favorites count, but even after a month, I prefer it with.

I'm interested in whether our dialogs were different without the favorites count. Were conversations better? Worse?
posted by zippy at 11:10 AM on December 1, 2009


And I voiced my opinion that I like seeing the comments without favorites next to them. Others like it. Hooray for options. But you trawling my posting history like that is kind of weird, and it really doesn't even fit. I'm surprised the mods didn't delete that.

If you can't see how your scenario of the world being a wonderful place if someone waved a magic wand and made all the foos disappear and fantasy of being all alone are more analogous to not participating on MetaFilter (Which, since you came back, obviously wasn't a satisfying experience) than they are to participating, I'm not sure what else I can do to illustrate my point.

I get that you were being flip and your tongue was at least partially in cheek, and I'm sorry if you think I was trying to embarrass or humiliate you; obviously, you are totally entitled to your opinion regarding favourites, and again, I'm happy everyone has an option now - but from where I'm standing, your reasoning and analogies are wrong-headed. Is disagreeing on MeTa the next thing you people are going to try to take away from me? Because that, cadging favourites and making FPPs that use the AlanMoore'sBeard tag are about the only things I have left that make me happy in this mean ol' world.

As for the horrific allegations of trawling:
i)I remembered being dismayed when Burhanistan left and happy when he returned.
ii) That MeTa was really funny.
posted by Alvy Ampersand at 11:15 AM on December 1, 2009


I had no idea there was an experiment under way, just figured the "has favorites" was the new world order. I got excited when I saw the numbers back and came over here to figure out what the dealio was. So, I guess I like the numbers. I don't know if mefi and ask.mefi operate differently but I never felt like they were part of a feeling of "popularity contest." I think they are a good way to get a gist of how people in general feel about a topic and also cut down on a lot of "me too" type posts.

Anyway, good work all around, mefi mods!
posted by amanda at 11:23 AM on December 1, 2009


I'm thoroughly enjoying re-reading November threads and comparing my first reads from new favorite-count-enabled reads. Great fun!
posted by iamkimiam at 11:49 AM on December 1, 2009


Thanks for taking our comments into account and doing the work for a rollover! Rock on.
posted by Solon and Thanks at 12:13 PM on December 1, 2009 [1 favorite]


I like the favourites. It's kind of how I read Mefi. If I don't have much time, I scan for things that are heavily favourited. If I have a bit more time, I read anything that's favourited. If I'm participating actively in the thread, I read everything. Favourites are a simple, effective way to filter for one's own time constraints.
posted by gonna get a dog at 1:34 PM on December 1, 2009


(Also, new thread.)
posted by cortex (staff) at 1:36 PM on December 1, 2009


I just created a Greasemonkey script to display the favorites the way I think would be most useful. "Liked", "loved", and "best".

I edited Plutor's script so it shows *, ** and ♥ It's pretty easy to modify to have whatever words/symbols you want.
posted by jessamyn (staff) at 3:15 PM on December 1, 2009 [1 favorite]


Actually, I changed that to *, <3 and ♥. You can download it here.
posted by jessamyn (staff) at 5:25 PM on December 1, 2009 [1 favorite]


Feelings on favorites:

—I found that favorites were largely noise for me, and that it was easier to ignore a "faved" than a number.

—I still liked getting them just as much, perhaps even more due to the feeling that they were more rare in November.

—I gave many fewer, but that's in part because I found out/remembered that they were searchable.

I will say that how I read is different from what most other folks seem to be saying, in that I go top to bottom, read comments before names or favorites, and that the discussion of links really is secondary for me to the links themselves. If a comment is really important, it will be quoted, and if I skimmed over it, seeing it quoted will remind me to read it in more detail.

But really, November coincided with a couple things, my getting more busy, trying to again dial back the aggro that can well up when I see things that I consider risible, and getting sick of the same dumb arguments from the same folks again and again. The threads that I really enjoyed were the Schroedinger's Rapist and sequel (weird as it is to say that I enjoyed them), but I thought they were just as awesome without favorites. Just like how the TSA TOOK MY DINGOES thread was a shitfarm even with favorites—and the idea that people needed favorites to avoid duplicating comments seemed pretty well given the lie there.

I was kind of surprised at how much I ended up in the anti-favorites camp, in part because I was pro-experiment and being attacked over that made me really think about and articulate the anti-favorite position without feeling like there was a countervailing pro-favorite argument that really held true for me. There were plenty of arguments that I felt, and still feel, are great for other folks, but they don't reflect how I interact with the site. This was the first time that I've ever had to block someone on MeMail, and it's not like I'm a shrinking violet. I wish that distaste for that didn't influence how I felt about favorites, but it did. It felt like favorites were just encouraging that asshole, and every argument he made for favorites just left me thinking, "Of course you wouldn't see a problem with that, asshole."

I still haven't decided whether I'm going to turn them back off again. I wasn't on the site all that long before they were instituted, and I think I got used to them pretty quickly—might have even been excited, but I don't remember. I know rolling over 10000 felt more cool than it should have, and I definitely still check to see what has gotten 'em through the recent activity. I'd like it more if I felt like folks were really going back to them but I ain't gonna lie, a hell yeah high five is pretty cool too. I might give this a month to think about too, y'know? Try some other reading styles.

I will say that I think that the fundamental issue is one that's ancillary to that of favorites, and that's that the community itself has grown bigger than anyone ever imagined. As a $5 n00b, but one who lurked for years, I have mixed feelings about that too. But then—and Orville Sash will know what I mean—I waited in line to get into my high school. The idea of ordeal preceding participation may just be normal for me. What might help, and there's no way to do this through system or structure, is an effort for high-volume regulars to decrease their contributions on hot-button stuff, but I'm just spitballing regarding what annoys me, and I know I'm not a pure priest of light on this one either. (It's not even that they have these opinions or proclaim them forcefully, it's just, like, given any contentious news link, there are probably some 30 members who I could predict their comments down to punctuation—it's the same as having the friend who always wants to talk about the one issue they're moderately informed on and you're like, oh man Paul's talking about fucking veganism again I can't wait until he breaks down drunk and crying with a sack of sliders.) But then there are new folks who have the same dumb ideas and you kinda want the consistency of a regular. I dunno.

Also, I only liked brussels sprouts (that's what my stylebook calls 'em) after having them roasted in a little bit of butter with salt, pepper and lemon. A friend made 'em for me and they were awesome. Still not sold on the artichoke though.
posted by klangklangston at 8:02 PM on December 1, 2009 [2 favorites]


Is disagreeing on MeTa the next thing you people are going to try to take away from me?

>>That's disingenuous. I simply made a silly remark about enjoying reading the comments without seeing favorite counts. Where did I call for ending favorites for you?


That's not so much me being disingenuous as making a silly remark of my own (I'm actually pretty indifferent to getting favourites too, but the thing about making posts about Moore's beard was completely sincere. That thing is awesome.)

Anyhoo, it's pretty obvious I've poisoned this well, sorry about that.
posted by Alvy Ampersand at 10:49 PM on December 1, 2009


jessamyn: "Actually, I changed that to *, <3 and ♥. You can download it here"

I spent some time trying to come up with symbols that would be easy to scan, but wouldn't indicate "better"ness. (I discarded simple zero to three stars because it seemed like too much of a rating.) It's surpising how hard it is to see "liked", "loved", and "BEST". Searching for ♥ would probably be hard, though, if you were interested in finding the most-loved comment in a thread.
posted by Plutor at 4:55 AM on December 2, 2009


Yeah the heart was intentional. I like that you can't really search for the ♥ you can only find the ♥.
posted by jessamyn (staff) at 7:44 AM on December 2, 2009 [7 favorites]


Home is where the ♥ is.
posted by flapjax at midnite at 6:58 PM on December 2, 2009


I like that you can't really search for the ♥ you can only find the ♥.

It's fairly easy to search for ♥ in Firefox, and I was about to explain how but I would feel like a heel if I did.
posted by grouse at 7:12 AM on December 3, 2009


grouse: "It's fairly easy to search for ♥ in Firefox, and I was about to explain how but I would feel like a heel if I did"

Although on my copy of Firefox, both the menu and toolbar clearly say "find" and not "search", so jessamyn is still technically correct.

The best kind of correct.
posted by FishBike at 7:47 AM on December 3, 2009


I thought it was an idiom, like a watched pot spoiling a bunch of broth.
posted by fleacircus at 3:09 PM on December 3, 2009


I thought it was an idiom, like a watched pot spoiling a bunch of broth.

A watched pot never boils.

Too many cooks spoil the broth.

Too many cooks watching a pot of never-boiling spoiled broth, while discussing idiom mashups.
posted by flapjax at midnite at 4:11 PM on December 3, 2009


I probably wouldn't have convulsively hated "has favorites" if it didn't make me feel like a semiliterate kitty cat. So, yeah, all in favor of those shorter options.
posted by eritain at 9:05 PM on December 3, 2009


The problem with "has favorites" for me, is that it kind of reminds me of playing poker, and someone goes out and lays their cards face down on the table. You just naturally want to turn them over and see the hand, even though it has no impact on the game. The "has favorites" is like a face-down hand, and my natural curiosity wants to look at it, and that's extra clicks I could have spent opening some more interesting link.
posted by StickyCarpet at 1:15 PM on December 4, 2009 [3 favorites]


A stitch in time saves nine lives.
posted by UbuRoivas at 1:30 PM on December 4, 2009


Look before swine
posted by found missing at 2:10 PM on December 4, 2009


All things come to he who is always lost.
posted by BrotherCaine at 2:45 AM on December 5, 2009


It's an ill wind that blows no good for the gander.
posted by UbuRoivas at 11:44 AM on December 5, 2009


I march to the beat of a different kettle of fish.

Ironically, I did not make this up. I got it from a friend, years ago.
posted by rtha at 11:52 AM on December 5, 2009 [1 favorite]


Wait. How is that ironic?
posted by Pronoiac at 1:33 PM on December 5, 2009


Ironically, I have eliminated the wrinkles in this shirt.
posted by found missing at 1:43 PM on December 5, 2009 [1 favorite]


MetaFilter: Where a Bucket of Cocks meets a Plate of Beans
posted by Secret Life of Gravy at 1:53 PM on December 5, 2009


I think of it as ironic because it implies, in a mixed metaphor way, that I go my own way and am all original and shit, and yet I can't be arsed to create my own mixed metaphor.
posted by rtha at 4:19 PM on December 5, 2009


Your creativity & originality are better spent elsewhere than blowing your own horn in a new way.
posted by Pronoiac at 5:49 PM on December 5, 2009


Your creativity & originality are better spent elsewhere than blowing your own horn in a new way.

Absolutely. All you people should go blow someone else's horn in an old way.
posted by flapjax at midnite at 6:23 PM on December 5, 2009


Metafilter: more RAR! than some spammy backwater Rapidshare-linking Battlestar Galactica blog.
posted by koeselitz at 7:46 PM on December 5, 2009


« Older Is polling wrong even if helpful?   |   Has favorites for cuteness Newer »

You are not logged in, either login or create an account to post comments