FriendFeed is shutting down
blog.friendfeed.comWe had a wonderful, thriving life science community [1] on Friendfeed, which we have not been able to replicate since. Some of that crowd even published a paper about it [2]. Most of the original crowd left sometime after the FB acquisition. While many of us hang out on Twitter, it's not the same thing).
----
1. http://friendfeed.com/the-life-scientists
2. http://journals.plos.org/ploscompbiol/article?id=10.1371/jou...
I miss that interface and community. Even if FriendFeed was still around, though, I wonder if it would be as popular. That group contained the early adopters of internet communication among scientists, a movement that has now grown exponentially. My guess is that the signal would be overwhelmed by noise, or broken down into sub-groups by specialty, which would overwhelm the delightful serendipity of it all.
Great question. No idea. The other thing to note is that a lot of the people who were involved then have moved on to other things and probably could not stay as involved (pointing at myself). It remains the best time I've had online.
I'm with you there. I had a lot more free time in grad school than I do now (both personally and professionally). Judging from the updates I see on twitter and blogs, most of that core group has gone on to bigger, better, more time-consuming things.
I wonder if the kids these days are using snapchat to complain about how we're all dinosaurs. :)
FriendFeed has good tools for keeping signal:noise high: Hide, like, best of day
That's where the 'nobody is on [insert social network name other than fb/twitter]' argument fails. There are always some communities that form around these websites for a variety of reasons . For example google plus has got some early success with photograpers, probably because its photo viewing experience was way ahead of the competition at the time.
It is always a pity to see such a space disappear.
FriendFeed had the best social network mechanics of all social networks I ever saw. Part of it made it into Facebook now (but I can't consider really using FB).
Unfortunately, it had no monetization strategy. I suspect that if they had optional paid accounts by now (e.g. $3/mo), many people would pay just to keep the thing afloat and their community intact.
Unfortunately, there seem to be only two ways for an online service to last: to become paid by users commercially, or to become "paid" by users' labor of love when they install and maintain it. The later practically requires the service to open its source code.
Thank you FriendFeed - I'll miss you. At some point I thought Friendfeed was going to take over the world... But as usual I have to be reminded that I am not representative of the mass audience.
+1
best thing to come out of friendfeed, Tornado. I remember asking Bret Taylor on his blog [0] when he gave a Tornado example if he'd ever release it. A bit of time went by, then he did. [1] Thanks Bret.
[0] http://backchannel.org/blog/tornado
[1] https://friendfeed.com/friendfeed-news/9b8fbaed/technology-b...
Goddamn Tornado was/is the best Python HTTP server ever. Asides from the wide utility library of non-blocking IO (years before node got popular) it had Sinatra-style class based routing that was far simpler than django's bizarre route.as_route() stuff.
A friend once said the worst thing about Tornado was how little publicity it got.
Why all the "was/is" in this subthread? Tornado's not going anywhere.
Only reason I wrote 'is/was' because it's been a couple of years since I used Tornado and other Python HTTP servers might have gotten better. I hack node all day now.
What did you think about Twisted?
Twisted was focused on general sockets apps rather than HTTP, and for some reason had code written using a styleguide different from the rest of the Python ecosystem.
Tornado OTOH was very HTTP focused. It also had a bunch of useful examples maintained in the actual repo so they were always up to date.
Agreed, Tornado was/is awesome. Is it still being used in new projects? I somewhat prefer it to node/etc for that style of app, but I would probably still choose JS for real-time web nowadays just because of the mindshare/community around it. I would love it if there was a thriving Tornado community somewhere that I'd missed.
We use it for quite a bit at my job (for everything really). Mostly, I think due to familiarity.
I enjoy working with it (and I've grown to really enjoy python due to my time using it), and it's possible to build a server that uses generators pretty easily. The only problem is that you have to be really watchful for blocking calls mucking the whole thing up.
For a long time, I really would have rather used node, but now, I'm ok with using python. I pushed us into a micro service style architecture, so one day, it's feasible to upgrade different parts of the stack with different technologies.
Will Tornado continue to be maintained? There are still a lot of commits being made: https://github.com/tornadoweb/tornado/commits/master
Tornado has a BDFL (Ben Darnell) who continues to maintain it, and hasn't been a part of facebook/friendfeed for some time now.
An interesting Quora answer by Bret about the status of FriendFeed some weeks ago.
And FriendFeed would have still be running if not that question. Bret was reminded about FriendFeed and decided to take it down completely.
Bret did not make the decision to close FriendFeed. Bret left Facebook more than two years ago. He had no involvement.
Quite sad to see that move now, especially when seeing that it's still actively used in Iran and Turkey, two countries who block or censor twitter at will.
One of those things...
Friendfeed was technically better medium than twitter in every single way at the time (maybe still) yet it died (don't mean that it failed). Another example of why making social startups is a huge risk.
It was much better for me than Twitter, but once the acquisition happened (FB w/ a rumored $50M), they decided to show the middle finger to its users. Isn't this the same as what happened to the thesixtyone (t61)?
They kept the service running for more than five years after the aqui-hire. I wouldn't call that showing the middle finger.
I suspect he was referring to the FriendFeed founders giving their user community the middle finger by cashing out to Zuck. People are going to do what is in their own self interest so you can’t exactly blame the founders for that.
As I recall FriedFeed’s growth had plateaued so they probably figured it wasn’t going to compete with FB and Twitter. Mobile and apps were also starting to take the world then too.
Did t61 even get acquired? They seemed to just toss their users for fun.
The Archive Team on the case to archive the content as usual #humancentifeed https://twitter.com/textfiles/status/575143304859361280
I think the Archive Team is going to have a hard time preserving Friendfeed content, due to resource limitations of the platform. For example, it only provides a very limited amount of data through the frontend (was 600 posts, now slightly higher), and slightly more through the API (10,000).
I'm currently working on a system, but it requires user input, in order to bypass the limitations of the API.
I personally hate when services let people down. I prefer a much more "future oriented" approach much like the one from Basecamp https://signalvnoise.com/posts/3830-ta-da-list-until-the-end...
or the idea behind https://posthaven.com/
I guess it's all about vision and how to delight your customers. All the people who believed in a product are people who trusted your words, and gave you access to a lot of their information. Yet it's so easy to let them down and shut down a service. It was for springpad, it now is for friend feed. Sometimes it's about the money sometimes it's about the users (I guess it's always about the money though).
But I can't help it, I prefer when you can trust a service,and I'd be willing to pay as long as I know that service will stay up even if I'm the only user.
They continued to run FriendFeed for 5 years after it was clear the core team was moving on. That is above and beyond the call of duty.
37 Signals has a healthy business with good revenue building only "sane" products. To compare the two is apples to oranges.
I know they are not the same kind of business, and I didn't want my view to sound harsh, sorry if it seemed that way.
As you said they did A LOT for the friendfeed platform even after it was somewhat decided they would have not mantained anymore, that is not something I want to ignore.
My point is that many services are born and die, some of them die because they weren't building "sane" products. It's a choice, not complaining about that, but I personally prefer when there's a path ahead that won't let the users down because they weren't relying on funding or other things to stay up, but only on the power of their users.
Maybe I'm asking for something impossible? Dunno, but the PostHaven/Basecamp thing is giving me hope :)
I don't feel let down at all. I was one of the most active users prior to the acquisition, and and still an active member of the site.
It was an inevitability, that the site would shrink, and that FB would either decide it couldn't continue to fund a small community or the handful of employees would decide to depart.
Currently, there are only 3 members of the team working at FB, and I have a feeling a few of them have decided to move on which has precipitated the closure.
Sad to see it officially shutdown. I have been using their API for years on my crusty personal page. It has been rock solid.
How can I export my data? I know there is an API but it will be incredible useful to have a link to download everything.
Google Reader had an option to export your data when they closed their web service.
You could only export your feed URLs from Google Reader. They had content archived, but you couldn't easily export the content. People wrote their own scrapers, it was a big mess and no one knows how much content only existed in Reader that was deleted when it went down.
There is a backup tool on github[1], I use it exported my feed when a saw the news.
It might make sense if they can opensource the code.
+1
I forgot all about friendfeed about the time facebook bought it. I remember it being quite good.