I answered my first few LinkedIn invitations with an RDF geek response: “look, I’ll point to your FOAF file if you want to point one at mine.” When I started gathering information for a job search, Ken North suggested that I reconsider my attitude about LinkedIn, so I joined up. The first surprise was how many RDF geeks I saw there. The XML community in general is pretty well-represented.

The next surprise was how much fun it could be to troll around, checking out my contacts’ contacts and finding out how many friends who I didn’t think knew each other actually do. For example, Priscilla Walmsley, Micah Dubinko, Dale Waldt, Betty Harvey, Eve Maler, Tim Bray, and Zarella Rendon all know Bill Trippe. I only met Bill briefly once after a talk he gave in New York, but my brother once worked for him at INSO, my LexisNexis boss Chet Ensign interviewed him when he did an SGML book for Prentice-Hall, I own a copy of Bill’s book on DRM technology… apparently there were far more connections than I realized. So I sent him a LinkedIn invitation to him, and he mentioned it in his blog.

If there’s someone you want to meet, LinkedIn tells you someone you know (in my case, usually former RIA co-worker Dale Waldt) who is one degree closer to the person you want to meet and offers to route a request for an introduction. You write a cover message to the person you want to meet and another to the person you know (“Yo, Dale, please forward this”).

The “product” LinkedIn offers is interesting: a collection of data, stepped access to it, and the ability to run one of the most classic computer science algorithms against it. (Well, actually, one from a family of algorithms—if you came up with a good new member for the family, you would be considered quite the computer scientist.) This is all free, so far; they make their money from advertising and from offering memberships with greater access to data and features.

FOAF files are fun, and interesting demo apps have been built around them, but instead of listing our actual friends they tend to only list friends of ours with FOAF files. I’ve joked that it might be more accurate to call them FOED files, for “Friends Of Edd Dumbill,” because Edd shows up in so many of the ones I’ve seen. (I added the URL for this picture to my FOAF link to Edd to give my FOAF file a little co-depiction juice. I don’t know what the strange bowl of liquid near the sugar packets in the picture is, but I suppose if I assign it a URL then Edd and I will be linked to it.)

It would be unfair to the FOAF project to draw too many comparisons to LinkedIn. I’ll just say that as a dot com with some real money behind them, LinkedIn has built a useful application, and if they opened it up with an API, the results would be fascinating. And thanks, Ken, for pushing me to join!

2 Comments

By Pierre on February 17, 2006 4:49 PM

Hi, the problem with FOAF files is that it seems to me that the only people using it are computer geeks. Of course most people don’t know anything about RDF, how to install a FOAF on a server, what to do with it, how to modify it etc.. whereas creating a profile on LinkedIn is straightforward and it is really easy to search in your network (the problem is to convince non-computer friends to accept my invitation and to extend their network :-) ….)-. Another social networks, less known generate FOAF from your profile:

• O’Reilly (http://connection.oreilly.com/)
• http://videntity.org/
• ?…

On my side, I’ve been trying to generate and display FOAF files for Biologist: see SciFOAF and MyFOAFExplorer

By Libby on February 17, 2006 6:41 PM

Pierre, also if you’ve not seen it: http://rdfweb.org/topic/DataSources has a bunch of exporters of foaf. Your tools are very nice. Care to add something to http://esw.w3.org/topic/SemanticWebDOAPBulletinBoard for us?

Bob, I can’t work out how to ‘friend’ you in linkedin - since I know you already it seems a bit weird to persuade one of my friends to introduce us. Guess I need to upgrade my account…