- From: Peter F Brown (Pensive SA) <peter@pensive.eu>
- Date: Fri, 10 Jul 2009 21:32:11 +0200
- To: "Dan Brickley" <danbri@danbri.org>, "Juan Sequeda" <juanfederico@gmail.com>
- Cc: "Steve Harris" <steve.harris@garlik.com>, "Richard Light" <richard@light.demon.co.uk>, "semantic-web at W3C" <semantic-web@w3c.org>, <public-lod@w3.org>
> -----Original Message----- > From: semantic-web-request@w3.org [mailto:semantic-web-request@w3.org] > On Behalf Of Dan Brickley > Sent: vendredi 10 juillet 2009 12:33 > To: Juan Sequeda > Cc: Steve Harris; Richard Light; semantic-web at W3C; public-lod@w3.org > Subject: Re: Dons flame resistant (3 hours) interface about Linked Data > URIs > > On 10/7/09 12:23, Juan Sequeda wrote: > > Steve is right. > > > > If I am not wrong, when TBL gave his talk at CERN for the 20th > > aniversary of the web, he said that he was amazed that people were > > hacking HTML by hand. He never expected it. > > > > Now... we are the geeks doing RDF, conneg, linked data by hand... In > a > > couple of years we will create tools for the non-geeks > > > > We have to learn from our history and not get ahead of ourselves. > > RDF has been a W3C Recommendation since February, 1999. The RDF work > went public in Oct 1997. > > A lot has happened since then... [PFB] Well, it hasn't, that's exactly the problem. Most people - geeks and non-geeks alike - "got" HTML and HTTP very quickly as most posters have already stated. Bluntly put: if RDF is the answer, why ten years down the line is the take up still so poor compared with html? There *has* to be a serious debate about subject identity and resources being distinct or this sort of debate will continue for another ten years, all the while more and more people are lulled into a false sense of security by the semweb bulldozer and try to build stable infrastructure upon fatally flawed architectural design principles... Remember: Tantalus also thought that going after low-hanging fruit was a neat idea. > > Definitely we've done a lot of hacker-grade stuff in the meantime. But > tools for going mainstream are getting overdue! Even tools for > developers: eg. regular Redland builds on Windows; a solid packaged > Ruby > library, etc. > > Re tools for publishing, given the fiddlyness of doing RDF right, my > vote is for everything that allows tools on one site to post RDF into > another. I've suggested before that AtomPub + OAuth would be a > plausible > starting point, but I'm open to suggestions. > > Re non-geeks, http://www.youtube.com/watch?v=o4MwTvtyrUQ is a must- > watch... > > cheers, > > Dan
Received on Friday, 10 July 2009 19:33:01 UTC