Follow

@epilanthanomai More linkies...the two essays from Clay Shirky on Semantic Web that I mentioned. I haven't really revisited them in a while, nor met anyone who had worked in Semantic Web and found good success in it, so I can't say I "stand behind" them.
shirky.com/writings/herecomese
shirky.com/writings/herecomese

@roadriverrail Thanks for the links! Yeah, I basically read Shirky's complaints boiling down to "This ontology and arbitrarily complex automated syllogism that TimBL is excited about isn't trivially implementable and useful." Which, to be fair, is true. I'd counter that TimBL's early semweb docs weren't intended as engineering guides: They were a picture of where he thought things could eventually go, along with stabs at first steps to get us there. And as that I think they're pretty good.

@roadriverrail Instead of going there, corporations dumped jillions of man-hours of labor into walled gardens, which is the opposite of that. And they made some nifty, flashy, and even useful stuff. Semweb stuff didn't get that labor, so it doesn't have the flashy.

@roadriverrail So more recently TimBL has backed off on preaching the far-off visionary work of arbitrary logic validation and focused on a more immediate problem: Teasing out bulk data storage and access from the vocabularies used to represent it, and allowing the decentralization of both. Based, ultimately, in another part of semweb that Shirky doesn't appear to comment on at all: representing data as graphs rooted in url space. Which is still pretty nifty imo.

Sign in to participate in the conversation
Signs & Codes

Signs & Codes is a private Mastodon instance built around a community of friends.