Edd Dumbill sighed when he read my comments on FOAF and social networking:
<danbri>http://jonudell.net/udell/2004-01-04-the-heisenberg-uncertainty-of-social-networks.html
* edd sighs
<danbri> its because foaf is associated in ppls minds w/ the 'social networking' sites
<danbri> and the 'f' in 'foaf' can't help
I wanted to know more about that sigh, so I wrote to Edd, and heard back from both he and Dan Brickley. Edd wrote:
For me at least FOAF's point is as the personal homepage technology of the semantic web. Like we all made homepages back in 1995. In fact the links of significance in FOAF are the rdfs:seeAlso, not the foaf:knows bits: the dumb seeAlso is the parallel to the dumb <a href="">. (Except it turns out we can hang more information on a seeAlso.
I don't agree with your assertion that Google's enough: there are many circumstances in which that isn't true. FOAF and other techs are useful in a lot of scenarios where we can and want to be more precise about making a link about our involvement with projects and people.
And Dan wrote:
I guess my take is that since this 'social software' bandwagon came along, commentators have largely lumped FOAF in with it, on the assumption (which to be fair the spec doesn't do enough to counter) that FOAF is all (and only) about explicitly representing typed relations amongst people. You _can_ do that with FOAF, and foaf:knows is an intentionally gentle starter in that direction, but you can also take things from a softer, less explicit angle too. Hmm I scribbled about this before somewhere [rummages], ah yep in reply to Shirky's piece:[[SW [Semantic Web] technology, specifically RDF, makes it *possible* to goof up in various ways, but it also allows for subtler treatments, which is where (hopefully) FOAF is headed through its focus on describing the photos, events, collaborations etc that are the evidence friendship leaves in the world, rather than crudely taxonomising classes of friend. ]]FOAF is a playground where we can try out different takes on the explicit/articulated vs soft / evidence-based approaches. My own bias is towards the latter, but folks who find value in formally taxonomising their relationships can exchange their data in RDF/XML with FOAF + extensions, and nothing much should break (except perhaps a few hearts; not everything that can be written down should be ;)
What struck me later about this interaction was its miraculous subtlety. I wrote something that made Edd sigh, I overheard his sigh, and we had a discussion about what provoked it. Now let's look at how this happened. My original comments were posted on this weblog. Edd and Dan may or may not subscribe to my blog, but given their central involvement in FOAF it was virtually certain that the item would come to their attention. Their reaction to it, on the FOAF chat channel, was logged on a public page. I became aware of it when somebody followed the link to my item from that page, which created an entry in my referrer log. A truly remarkable chain of events. This kind of thing happens every day, but I continue to find it astonishing.
It seems we all agree on the need to simultaneously explore what Dan calls "explicit/articulated vs. soft / evidence-based approaches." My ability to "overhear" Edd's sigh is an example of the kind of softness that's already inherent in the systems we're evolving. My ability to search this blog for items in which I quote Edd and Dan (experimental server-based version, not guaranteed to stay up, but see also the client-based version) is one way to be more explicit. I'm all for more ways to be explicit, but we've got to weave the stuff into our ordinary and natural activities. And if it's really going to scale, "ordinary and natural" has to mean something very different from what it means to Dan Brickley, Edd Dumbill, and me.
Former URL: http://weblog.infoworld.com/udell/2004/01/06.html#a880