From Archiveteam
Jump to navigation Jump to search
FriendFeed logo
FriendFeed is the easiest way to share online
FriendFeed is the easiest way to share online
Status Offline
Archiving status Partially saved
Archiving type Unknown
Project source friendfeed-discovery, friendfeed-grab
Project tracker friendfeeddisco, friendfeed
IRC channel #archiveteam-bs (on hackint)
(formerly #humancentifeed (on EFnet))
Data[how to use] archiveteam_friendfeed

FriendFeed was a standalone social network, also aggregated RSS data for a user from the latest hip sites and put it all in one place. On 2009-08-10, it was announced that FriendFeed had been acquired by Facebook. On 2015-03-09 it was announced that FriendFeed would be shut down on 2015-04-09. After a long vigil FriendFeed got actually cut off on 2015-04-10 around 21:00 UTC.

Shutdown notice

Posted by Benjamin on Monday, March 09, 2015

Dear FriendFeed community,

We wanted to let you know that FriendFeed will be shutting down soon. We've been maintaining the service since we joined Facebook five years ago, but the number of people using FriendFeed has been steadily declining and the community is now just a fraction of what it once was. Given this, we've decided that it's time to start winding things down.

Beginning today, we will no longer accept new signups. You will be able to view your posts, messages, and photos until April 9th. On April 9th, we'll be shutting down FriendFeed and it will no longer be available.

We want to thank you all for being such a terrific and enthusiastic community. We're proud of what we built so many years ago, and we recognize that it would have never been possible without your support.

- The FriendFeed team[1]



It got thrown into ArchiveBot (job:7v7q0bcx27e43a527mdhd765f). But later a Warrior project also popped up.

ArchiveTeam was able to discover nearly 1.5 million public accounts and groups, but in lack of time (we had about a week) we couldn't download every single post from everyone. Instead, what was saved for everyone is the first twelve pages of their posts. However, in the beginning, and after that upon request, we also downloaded quite a few accounts in its entirety (or, in fact, up to last 10,000 posts, as accessing more was not possible even using the API).

Size of archives (compressed): 2.4 TB. The WARC files can be found in the archiveteam_friendfeed collection of the Internet Archive.

A group of Russian-speaking users self-organised in an effort to avoid a dead end. A non-profit entity MTÜ was registered in Estonia and a number of volunteers joined efforts to enable users to archive and later migrate their accounts to a new service.

User account data archiving was done using Clio backup script by Victor Shepelev. Script was quickly banned by Facebook and a huge effort to bypass the ban using multiple nodes was needed to make data archiving possible. Both private and public accounts and groups were backed up for every user who opted in to back their data. All posts available for the user to access: all data in private groups, all posts and comments including those by users that did not opt-in into the backup - were saved and data archiving was running in real time up to the last minute of lifetime.

An open-source clone Pepyatka was used initially as a near-ready platform for the service. When it's architecture proved to be unfeasible, team organized a crowdfunding campaign on Indiegogo to help volunteers re-develop the system. Campaign was successful, team raised $11,700 and the beta version of it was at [1]. is currently an Open Source (MIT licence) project, democratically run by the community of its active users.

Other efforts

Many users have migrated to the Italian FriendFeed clone project


Although ArchiveTeam was informed only about the Italian clone project, there must be quite a few others out there.

Some users may have got json-exports of FriendFeed accounts. Although they differ from ArchiveTeam's preferred format (that is WARC) and won't be ingestable by the Wayback Machine, they are more than welcome in the Internet Archive. (If it's a private account, though, don't share its content without the user's consent!) Upload them – preferably compressed – to an item (or several items) to the Internet Archive, add metadata (description, tags) so that they can be found by search, and also let us know, maybe on IRC, so that they can be moved to the corresponding collection.