I recently gave a presentation on social software to an internal conference at IBM. In it, I identified what I see as a number of key webapps out there on the interweb today. Along with blogs, I also talked about del.icio.us, Dopplr, Twitter and Facebook (MySpace being so last year, especially since Facebook’s API went live a few months ago)
One of the accusations commonly levelled at Facebook is that it is a walled garden for data. They appear consume the data that people put in, but it is allegedly remarkably difficult to get it out again.
This morning, Dennis twittered at me to point me at an excellent post highlighting how data can be shared out of Facebook. SAP’s Craig Cmehil has demonstrated how this could be done – interestingly, the scenario he suggests is for companies to pull in information about new hires from their Facebook profiles.
Facebook as a trusted third-party / clearing house for personal data? Will enterprises go along with this? It’s all very interesting. I’ve been saying for some time that entropy will happen as increasingly, new webapps pop up demanding that we create profiles, duplicate data, etc.. Could Facebook be onto a winner here?
(I’m off to a meeting, but may post more on this if my brain kicks into gear later!)
Garrh! I’m glad that facebook are slowly opening up a little, and it is necessary, but the “correct” answer to this problem is much more along the lines of openid.
There’s one problem with this thought; FaceBook doesn’t ‘own’ a lot of this data that people are dumping into FaceBook; they just own the ‘core’ data. If someone were to write a ‘resume’ app, the person writing the app ‘owns’ the data. You need to extend the trust model to the app builders. Which I’m not sure will work. Personally, I’d prefer to own all the data myself, and then give apps permission to use certain pieces of it.
We’ll see calls for this sort of thing as we start to see data we’ve been feeding FaceBook apps bite us in the arse.
It’s actually a call out for a new meta-api for FaceBook – 3rd party data management.
I’m confused by Dopplr. They seem to have a number of basic failings, most obviously that they don’t email you when something serendipitous happens and you’re scheduled to be in the same place as someone else. Since that seems to me to be the main value of the tool, that seems quite an oversight.
It’d also be nice if it synced with my calendar: I can no longer be bothered to keep it up-to-date.
delicious, on the other hand, is incredibly useful. A wonderful tool, which I use all the time (1,600 bookmarks and counting).
Andrew,
I appreciate your comments about Dopplr. When we say Beta, we really mean it – we’re still working hard on getting the site complete and ready for primetime. Email alerts is something we’re working on, and we’re taking time to get it just right. Calendar syncing is also on our Todo list. I hope when we hit 1.0 you’ll find that your problems have been fixed.
Cheers,
Matt.
Yay for Twitter and for mattb – thanks for the input!
@kyb, @Patrick – right. That’s why I used the term “trusted third-party”. Whilst I found Craig’s demo intellectually interesting, and scary like Dennis did… I’m really not sure that an enterprise would want to entrust Facebook with that role. And I’m not sure that lots of people would. And should it even be an enterprise thing, or should something like OpenID work better in this space. I’m a little busy today so my thoughts on the subject are struggling to cohere to one another!
The model for how this needs to work is how Flickr works. If you want to use some 3rd party app to do something with your account, like using a Flickr uploading tool, or Moo (perhaps), then you need to tell Flickr that it’s ok for that service to use your data, and how it’s allowed to use your data.
Flickr provides the data, and the protocol with which to validate who’s allow to do what on your behalf, and can presumably do this without requiring you to hand your Flickr password over the the 3rd party (ala something like OpenID).
OpenID is just part of the picture. I think there’s going to be a market for a data warehouse that allows you dump structured data in (RDF, perhaps), and of course generic ‘media’, and then has a protocol to allow 3rd parties to access it with your approval.
i agree 100% with you Pat. this is why i’m really resisting signing up for facebook. i’m a bit wary so far. popularity aside.
There are many ways of looking at this. Craig did look at the TOS model and shows how permissions might work. I think it’s important to remember we’re still very early in this game so there will be a lot of experimenting.
[…] Facebook – Facebook and the enterprise […]
Matt, OK, cool. Maybe I shouldn’t have laid into you guys quite so much. I’m just jealous really because I had the same idea a few years back and never followed through on it 🙂 Look forward to seeing how it develops – I will come back from time to time.