Digg increases alexa research

July 1, 2006 5:32 PM

Digg did it.
posted by evariste to MetaFilter-Related at 5:32 PM (44 comments total)

Is the pronoun digg or Digg? I don't care, I hate digg. I just thought the people who were speculating about the doubled-MeFi-reach-on-Alexa thing might be curious to see some interesting reasoning about this. It figures that some population possessing a combination of stupidity and techno-savvy had to have done it, since it was mostly techy sites that gained "reach". Basically, the duggalos* freeped Alexa, and MeFi's extra reach was the result.

*Allusion intentional.
posted by evariste at 5:37 PM on July 1, 2006

was the a result.
posted by evariste at 5:37 PM on July 1, 2006

But it wasn't anything coordinated it sounds like, it was just a lucky fluke that people reading a story about alexa installed the toolbar?
posted by mathowie (staff) at 5:53 PM on July 1, 2006 [1 favorite]

Yeah, I don't think there was some conspiracy by digg readers to install it, they were probably just exuberant that day because they'd finally passed Slashdot's reach on Alexa and a lot of them decided to seal the deal by installing Alexa and "participating" in the "victory".
posted by evariste at 5:58 PM on July 1, 2006

Slashdot, is that thing still around?
posted by keswick at 6:05 PM on July 1, 2006

I love slashdot -- I get mod points pratically every week and I mod down my enemies and the people I disagree with. :<
posted by yeoz at 6:21 PM on July 1, 2006

The fact that MeFi, Slashdot, BoingBoing,Wired and others got almost as big a boost in Alexa stats as Digg did means that those who went to Alexa for the news and got the toolbar are a relatively homogenous chunk* of the internet population... so let's not alienate them by calling them a bunch of brainless sheep (even if they are).

*I always wanted to use the words "homogenous" and "chunk" together... Oxymoron Hall of Fame, here I come!
posted by wendell at 7:27 PM on July 1, 2006 [1 favorite]

Fiendishly cunning in it's simplicity. I had been wondering, since the spike appears somewhat ubiquitous among tech-related sites. If I interpret those graphs correctly, they are showing 'reach', which is the percentage of users with the toolbar who visited the site. The typical result of the spike for techy/web2.0 sites is (roughly) a doubling of reach, which suggests alexa's user pool is fairly small, and that digg users now make up a very large proportion of alexa traffic data. So if you want good alexa stats, spamsubmit your site to digg.

These graphs appear to confirm the spike is largely confined to the digg-metafilter-boingboing set.
posted by MetaMonkey at 7:42 PM on July 1, 2006

Heh. I was just coming here to post this.
posted by youarenothere at 7:43 PM on July 1, 2006

I can digg it.
posted by Faint of Butt at 7:52 PM on July 1, 2006

those who went to Alexa for the news and got the toolbar are a relatively homogenous chunk* of the internet population

This bears repeating. The "geeky internet junkies who like technology and weird links" population isn't as segmented as you might think.
posted by danb at 7:58 PM on July 1, 2006

... that's cuz the workday is eight hours long with a sizable chunk for lunch.
posted by mischief at 8:16 PM on July 1, 2006

Digg has this unholy obsession with the Alexa ratings, like the comely AV nerd in high school who pines away in his journal for some girl he's ever talked to once. History is littered with dozens upon dozens of stories dugg to the front page there about how digg's alexa ranking has inched forward a fraction of a percent. It is truly pathetic.

And this bit of insight, if true, really is damning to the (already nonexistant) credibility of alexa's methods. I wish the whole internet could collectively say, "this site sucks and obsessing about it something that is wholly inaccurate to begin with is stupid" but I don't think that day will ever come.
posted by Rhomboid at 8:35 PM on July 1, 2006

Is the pronoun digg or Digg?

It's a noun, and it's "digg"
posted by signal at 8:37 PM on July 1, 2006

wendell-it's not just that they're brainless sheep; that I can live with. There are plenty of websites full of those. I avoid them, don't think about them, and am happy. It's that digg people are polluting the whole damn internet with digg spam. In the last six months, every single interesting article on a blog that I see has "If you like this article, please digg it!" in garish colors at the top and bottom, which makes me feel like a sucker because I ended up at yet another digg-whore's site by accident. I'm not talking about those little del.icio.us | digg | reddit | [etc] blocks, which I think are tasteful and OK. And then to top it off, even when the site owner isn't a digg whore, the digg people can't shut up about the magical fact that they've linked to the guy's website. I'll be reading a perfectly interesting comment thread, and suddenly, droves of diggbots saying "this article has been featured on digg.com!" and "dugg!" and "if you didn't know, you've been dugg ;-)". Jesus, shut up already with the content-free spam! You don't think people who own websites know to have a peek at the referer logs if their traffic seems higher than usual? Every other website manages not to go over and trumpet "We linked to you! We linked to you!" just to celebrate this mundane fact that...they linked to someone.

Basically, I view digg as an advanced form of spam, and I look down on both people whoring to be dugg, and digg members who pollute countless conversations with pointless "you've been dugg!" announcements.
posted by evariste at 9:16 PM on July 1, 2006

It's a noun

Of course it is. :slaps forehead:
posted by evariste at 9:16 PM on July 1, 2006

"We linked to you! We linked to you!"

... [via] MetaFilter [!]
posted by mischief at 9:26 PM on July 1, 2006

evariste : "Every other website manages not to go over and trumpet 'We linked to you! We linked to you!' just to celebrate this mundane fact that...they linked to someone."

Slashdot users have a tendency to do that, though.
posted by Bugbread at 11:36 PM on July 1, 2006

Basically, the duggalos* freeped Alexa, and MeFi's extra reach was the result.

Doesn't this mean we are the duggalos?
posted by scarabic at 11:44 PM on July 1, 2006

scarabic : "Doesn't this mean we are the duggalos?"

Not necessarily (though incredibly likely). It just means people who view MeFi are the duggalos. They could be largely unregistered readers.
posted by Bugbread at 11:55 PM on July 1, 2006

We're not the duggalos or those would be a bit better correlated than they are.
posted by evariste at 1:10 AM on July 2, 2006

I don't get it. What does the Alexa toolbar do other than spy on its user?

Oh, and I just checked. It requires Windows and Internet Explorer. That should tell you something about the people who install it.
posted by sveskemus at 1:23 AM on July 2, 2006

Exactly! But some people think those numbers are the definite authority and keep citing them numbers about my site to me - where only 1.7% of my users drive a windows based machine. AAARGH!
posted by dabitch at 2:32 AM on July 2, 2006

and just for that I posted some digg-alternatives to the front page. There must be hundreds more.
posted by dabitch at 2:32 AM on July 2, 2006

Actually, there are some extensions for Firefox that include Alexa, which probably aren't shown on Alexa's page because they aren't actually made by Alexa. I use SearchStatus, which anonymizes requests (though not by default, unfortunately) to show the Alexa ranking and Google pageranking of everywhere I visit on the status bar. Not amazingly useful, but since my data is anonymized, and Alexa just finds out that some guy, somewhere (at some bogus IP address), is looking at Metafilter, there's no real "spying" either, and I enjoy the trivia.
posted by Bugbread at 2:58 AM on July 2, 2006

Yeah, it seems quite likely that your average digg user would install the Alexa toolbar. The whole site is built up for people who wish they were techies, but really aren't -- it's all "a beginner's guide to PHP" and "how to set up Linux" stuff. Tragic, really. Wannabe nerds.
posted by reklaw at 2:59 AM on July 2, 2006

Funny how single-link posts are frowned upon here

Funny how that myth just won't die.
posted by mediareport at 8:49 AM on July 2, 2006

I didn't say it was policy. I said it was frowned upon. And it is. Lots.
posted by scarabic at 11:01 AM on July 2, 2006

Doesn't this mean we are the duggalos?

Spend some more time over there. You'll soon realize that we aren't the duggalos. And thank god for that.
posted by puke & cry at 11:28 AM on July 2, 2006

I said it was frowned upon. And it is. Lots.

wrongfully so.
posted by mathowie (staff) at 12:07 PM on July 2, 2006 [1 favorite]

holy shit there has been a digg blacklash for this long?

man my netdar is so broken.
posted by fishfucker at 12:16 PM on July 2, 2006

And it is. Lots.

Only by newbies. Weird. Somehow the meme stuck in its moronic form, and it keeps getting spread (in part by offhand comments like yours, honestly).
posted by mediareport at 1:30 PM on July 2, 2006

[Didn't mean to sound so snippy at you, scarabic. It's just annoying to keep seeing that idea show up.]
posted by mediareport at 1:51 PM on July 2, 2006

Digg and Slashdot audiences are the same? Holy shit!
posted by furtive at 1:57 PM on July 2, 2006

More like the intersection of the Digg, Slashdot, MetaFilter, BoingBoing audiences is very large. Most likely you can throw in CNN and a few other sites without making much of a difference.
posted by mischief at 2:03 PM on July 2, 2006

Single link threads should only be created by advanced thread starters ... People who know what they're doing.

Newbies should stick with the encyclopedic form until they're more experienced.
posted by crunchland at 2:41 PM on July 2, 2006

Link quantity directly proportional to user number...
posted by mischief at 2:51 PM on July 2, 2006

I'm skeptical of this explanation for Alexa's data glitch.
posted by Nelson at 3:39 PM on July 2, 2006

if alexa's installed userbase is not very large, it makes sense to me.


We should do a survey to find out how many people on metafilter installed the toolbar.
posted by empath at 4:18 PM on July 2, 2006

empath-no survey needed, #1 can just mine the logs for the info. Alexa toolbar users inadvertently advertise the fact in their useragent string.

grep "Alexa Toolbar;" /path/to/server/log/file.log | sort | uniq -W 1 | wc -l

would give the number of unique IP addresses visiting MeFi who have "Alexa Toolbar;" in their useragent. Divide by total unique IPs and you have a percentage.
posted by evariste at 4:25 PM on July 2, 2006

My own logs show 4 Alexa toolbar users out of 3469, which is quite miniscule indeed. My readers tend to be less techno-literate than the MeFi level, so I imagine it's even less than one in a thousand here.

discardedlies:~# grep "Alexa Toolbar" /var/log/lighttpd/access.log | sort | uniq -W 1| wc -l
4
discardedlies:~# cat /var/log/lighttpd/access.log | sort | uniq -W 1| wc -l
3469

posted by evariste at 5:06 PM on July 2, 2006

So I think the duggalos-freeped-Alexa theory is quite credible, since it doesn't take many new users of Alexa to skew it dramatically. If the intersection on a Venn diagram of MeFi readers and digg readers is, say, 10,000 people, only 10 of them needed to install Alexa to double the number of Alexa users. If the intersection is 100,000 people, you only need 100 people to double it.

Also, the scale of the numbers means that we are innocent of being the duggalos.
posted by evariste at 5:16 PM on July 2, 2006

the comely AV nerd in high school

Most of the AV nerds I've known have trended a little more closely toward homely.
posted by dhartung at 2:01 PM on July 4, 2006

mediareport, I appreciate the fact that maybe you don't think single-link posts SHOULD get frowned upon, and that if I say they often are, I am in some way reinforcing the frowning.

But dude, I'm not advocating for the frowning, nor am I making up the fact that it goes on. mathowie says "wrongfully so," but that's different than saying it doesn't happen at all. And you gotta wonder why this "myth" keeps coming up if it's just not so.

If you look at what I originally said, it might even stand to support single-link posts:

Funny how single-link posts are frowned upon here but they're the only format available there.

Question being: if a site like Digg can subsist solely on single-link posts, they can't be all bad.

I don't mind if you get a little snippy but I am rather confounded by your insistence that I'm making this up. People frown on a lot of things here. Newsfilter. Pepsi Blue. Etc. Etc. Even mathowie's deletion's are often cited, at least in part, to "single link-ness," as in "single link op-ed."
posted by scarabic at 6:29 PM on July 5, 2006