in two weeks.
) have popped up since the closure announcement, and that's OK. But there's one highly valuable aspect of Reader nobody can replicate and that will soon be lost forever: Reader's vast archive of cached feed data. Here's how to save it all.
Every feed ever submitted to Reader has its metadata, plaintext, and other embedded media stored in Reader's servers -- even for sites that died years ago. For many sites, it is the last remaining record of their content -- an ordered, organized, comprehensive, and searchable list of thousands of items where Archive.org coverage is spotty or robots.txt files block the way.
This huge trove of data, which made Reader's personalized search of all your subscriptions so powerful, is being destroyed at the end of the month. But while Google Takeout
lets you easily export several kinds of data
, including an OPML subscription list and starred items, this far more complete archive is hidden away in Google's server racks.
Luckily, the unofficial Reader API gives users (unintentional?) access to this database, and can provide a complete list of any feed's contents for backup. Labnol.org offers a page with a plain textbox that will generate this page for you.
Just drop in an RSS feed URL and hit the download link to get a raw JSON formatted list of up to 9,999 items. You can also get these results yourself using this URL:
Note that the feed URL needs to be encoded; you can do that using this tool
. Also, you can tweak the number at the end if you want less data, but I'm not sure if 10,000 is the upper limit.
You can find the feed URL for any site you're subscribed to in Reader by going to its feed, clicking the "Feed Settings..." button up top, choosing "View details and statistics", then copying the blue Feed URL shown in the window.
This is an invaluable tool for any site you'd like to save -- personal blogs, long-dead websites, ones with rejiggered content schemes, paywalls, or contents that are difficult to
Bing.* Two that may be of interest to Mefites: a rolling, daily list of the most popular posts
across the site, going back years.
Lastly, if you'd like to aid Mefi's Own jscott
's Archive Team
in their ongoing effort to discover and back up as many Reader feed contents as possible, you can join them on their project page!
* FROM HELL'S HEART I STAB AT THEE