Dec 18

David Recordon

David Recordon

Battling Social Network Fatigue ... By Going Open

Back in February, plenty of us started to think about and discuss social network fatigue: the idea that people are getting tired of joining new services and having to reconnect with everyone they already know. Some have argued that this isn't a real problem outside of the Valley or that people are happy starting over as they move across networks. Social network fatigue is not, however, a made-up problem. If anything has changed, it is that we're now all much more tired of new social networks and the political battles to be "open" then we ever were earlier this year.

Before going to Paris for LeWeb, Blaine Cook (Twitter) and I stopped in Amsterdam to attend an event on federating social networks put on by Mediamatic Labs. Mediamatic is a partly government-funded startup in the Netherlands that builds social networking sites, mainly for non-profits with many focusing on the art community. Mediamatic has found that many of its clients and users have shared interests, causing the walls between social networks to become blurred. During the day, roughly 25 European developers and designers came together to look at how we can work together to help solve these issues. While the event was only a day long (and jetlag was certainly setting in by the end of it) one of the key takeaways for me was that we must be on the right track as the conversations follow the same direction in Amsterdam as they do in San Francisco. Mediamatic Labs will be hosting another event in February (see the company's blog) and in March there will be a WebCamp on social network portability in Cork.

During Web 2.0 Expo Berlin last month, I was introduced to two smaller projects, NoseRub and Lifestrea.ms, both of which originated in Germany. NoseRub (an open source project) and Lifestrea.ms are both social aggregators much like Facebook's News Feed. The key difference is that unlike Facebook, which only sucks content in, they're both built on open standards that let people republish their activities and content around the Web. A few weeks ago, Google and BBC Backstage hosted a BarCamp in London where a major theme was using open technologies to enable distributed social networks. Last week TechCrunch wrote about Spokeo, another social aggregator, created by Stanford students; social network portability was a hot topic at LeWeb3. And don't forget Plaxo Pulse, FriendFeed, socialthing!, and dozens of smaller efforts to integrate your various social services like the profiles on Threadless (past Radar coverage) or MetaFilter. This conversation is being fueled by energy from entrepreneurs and developers all around the world.

2007 was a year of solidifying open technologies like OpenID, OAuth, Microformats, XMPP, and others. The challenge now for 2008 is working together to build upon these technologies to create true data portability -- while keeping people in charge. Social networks may be different around the world, but the one common theme of all these conversations is that we're all getting sick of them!

tags:   | comments: 18   | Sphere It

Previous  |  Next

0 TrackBacks

TrackBack URL for this entry: http://blogs.oreilly.com/cgi-bin/mt/mt-t.cgi/6148

Comments: 18

  Josh Spaulding [12.18.07 06:37 AM]

I'll admit, some of that was a bit over my head. I'm often surprised as to how much is written on social networks these days. I suppose the reason is because, although there are so many it's enough to make your head spin, they continue to be utilized like crazy.

  Nick Vidal [12.18.07 07:45 AM]

The ISS (Instant Syndicating Standards) open standard is something worth looking at. It makes a bridge between the Web and XMPP and helps describe how people are connected semantically.

  David Recordon [12.18.07 07:52 AM]

ISS looks pretty interesting Nick. One of the things I've been looking at is how could we bridge OpenID and XMPP (basically URL based identifiers and JIDs). Seems like you've been thinking in that same area.

  Keith [12.18.07 08:07 AM]

There's also 8hands, a desktop tool that function as a social hub by aggregating your friends and content from all of your profiles.

  Christian Scholz [12.18.07 09:20 AM]

I guess you are also aware of DISO which also sounds very interesting to me.

My dream is still that I don't need to identify all my friends (and myself) on each network but just go there, log in with my OpenID (or similar means) and get to choose which group of contacts is allowed to interact with me in which way on that site.

Adding new friends there should automatically update my global contact list when adding that new contact to a group.

I think it's a somewhat long road to that scenario but I think with all those standards already around it should be possible. The main problem is of course getting social networks to open up.

In the end we might be back with only one social network - the internet - and the rest might simply be social applications (as I think you put it in your talk in Berlin).

(and too bad I missed Amsterdam but these days at the end of the year are simply too packed. It would be cool to be able to attend something in e.g. Second Life or do it on IRC).

  Georg [12.18.07 12:17 PM]

very nice app screencast:

  Matt Galligan [12.18.07 12:20 PM]

@Christian - David mentioned my company, socialthing! in the article...

In regards to what you're asking about...a way for you to just log into a site and it already have your friends...that's part of problem we're solving. We're going to consume the friends lists of the sites that you work with and then also syndicate that out for your own personal use through our API and our own custom solution built on top of OpenID. Anyone that interfaces with us will be able to suck in your existing friend relationships/groups/etc.

If you'd like to chat further about what you think would be cool, go ahead and shoot me an email at matt at socialthing dot com

  Josh Patterson [12.18.07 06:20 PM]

I'm going to pimp http://www.dataportability.org/ a little here. We have a group of over 30 people working towards user-controlled data portability. We are always looking for more people who are interested in this common goal, as we have several projects in development that essentially create a reference "stack" (think the OSI Network model, but for open data interoperability) so an application (web, desktop, mobile) can simply "Get on the Graph" of open data. Even more interested? Join the google group,


which I think is currently private, but may be going public soon. Or you can just email me and we can trade napkin sketches all day.

Josh Patterson

  Dale [12.19.07 12:00 AM]

Nice read, And i like your blog.


  Christian Scholz / Tao Takashi [12.19.07 01:06 AM]

@Matt I will have a look should I get some time during this end year/xmas stress ;-) Thanks! :-)

BTW, similar discussions are now going on in the realm of virtual worlds. Second Life is trying to open up to become an open metaverse and maybe others will follow. And if you see those as simply another social network the same protocols for at least your friendslist etc. could be applied.

  Vero Modas [12.19.07 02:59 AM]

@Matt I guess nobody wants to be stuck again with "a single point of failure" - means:

Why do you build another API for getting my friends out of your system again?

Why don't you use open standards like microformats, XFN and so on?

At least I wouldn't like to see my data in the hands of one service with a propriatary API.

Btw: Who's gonna support your API? Where can I make use of it?

  Brian [12.19.07 12:13 PM]

@Josh Patterson - Has that group thought of working with the Open Social movement? From our what i've been told by Google is they want it to be a community driven, open "source" type standard that exists on its own as Open Social and not as Google's Open Social. I think that the two groups' goals are aligned and if those talks/cross-pollinization hasn't started, we should get on that.

@Vero - we are going to support some of the microformats. We're not making you create yet another connection. We're inheriting relationships you have already established on the Internet. Its not like Plaxo Pulse where i have to be in plaxo. Then i have to be in pulse. Then i have to get all my friends to get in pulse. Then we have to tell pulse we're friends here, here, here, not there, here and here.

We also have plans to not only implement OpenID but add to OpenID to make it even more useful for such a purpose.

  Josh Patterson [12.19.07 01:40 PM]

@Brian - thats a complicated question. I think our model is more centered on putting the control of data firmly in the hands of user and treating the web as a single logical disk / logical db. Open social is more of copying facebook's plugin platform, and making a standard way to plugin widget's into data silos -- we are working to make data queryable across silos, to allow you to say (and this is a far reaching / long term goal) "share this album of wedding photos with only my immediate family" --- and have our model (nicknamed WRFS - Web Relational File/Data System) use a series of abstraction layers (think again here OSI Reference model, but for web data) to pull your distributed social graph, infer the relationship "immedidate family", and give rights to those people (at runtime). Pipe dream? possibly. But those are some things we are looking at, and talking to the OAuth devs about. Email me, we can share notes, we are a very open group just trying to make our way in the startup world. In the end, its time to "Get on the Graph".

  David Recordon [12.19.07 01:50 PM]

@Brian, like Josh said I think at least today the goals of the Data Portability group are different than those of OpenSocial from Google. OpenSocial really seems to be much more about running social applications within social networks whereas others are more interested in how data can be exchanged between social applications and social networks.

  Jonathan Vanasco [12.20.07 11:03 AM]


We (FindMeon) launched IdentityResearch earlier this month. It's a collection of our speeches/presentations (mostly mine) since 2006 - when we first started pushing for Open Social Networks & User Privacy. (actually '05 under another name)


We never really made it on the radar, because every time we released a new product, it was cloned/re-implemented by someone with more cash (millions, now billions)- and our funding round would fall through. Unfortunately, that means we now spend more money on lawyers than development-which is sad for everyone involved.

  Ian Gregory [12.21.07 02:38 AM]

It seems to me that the openest (is that a word?) social network is FOAF. It does not rely on any sort of centralised infrastructure and the network arises organically as people create RDF FOAF files and put them up on their own servers. Then there is the PGP key signing process which also creates a social network in the form of a web of trust. Unfortunately (perhaps because they are not under the control of a single commercial entity) neither seems to be achieving its potential.

  David Recordon [12.21.07 09:44 AM]

How do you see using FOAF to describe different types of relationships? As far as I know it only has "foaf:knows" whereas XFN has a richer set of vocabulary.

  Ian Gregory [12.22.07 02:53 AM]

@David - I have some knowledge of FOAF but until now I had never heard of XFN. My comment still stands in a way because the article mentioned neither (XFN was mentioned by Vero in a comment but I hadn't noticed that before I commented). I just searched for "XFN vs FOAF" and found a useful page which compares and contrasts the two technologies
The final paragraph explains how XFN and FOAF can be combined in a way that "allows the FOAF information to be enriched by the finer-grained relationship information embodied by XFN values". I clearly have some reading to do.

Post A Comment:

 (please be patient, comments may take awhile to post)

Type the characters you see in the picture above.