- From: Egon Willighagen <egon.willighagen@gmail.com>
- Date: Tue, 13 Oct 2009 22:13:33 +0200
- To: Matthias Samwald <samwald@gmx.at>
- Cc: public-semweb-lifesci <public-semweb-lifesci@w3.org>, Mark <markw@illuminae.com>
On Tue, Oct 13, 2009 at 5:14 PM, Matthias Samwald <samwald@gmx.at> wrote: >> Back in 2006 in the "Creeps" paper, Ben and I wrote that the SWHCLS >> community has spent too much time focusing on "Semantic" rather than on >> "Web"... and it really hasn't changed much nearly 4 years later :-( I >> honestly think that, if we all pulled in the same direction, we could make >> the "Web" aspect of the Semantic Web work better than it currently does... > > This is quite dissonant with the impressions that I have. The Linked Data > paradigm is very popular, and a lot of work in this area (as well as query > federation) is going on at the moment. What exactly are you missing in the > work that is currently going on? You wrote: "Besides, even though linked data URIs and federated queries are nice, it is quite practical to have all relevant datasets accessible through a single SPARQL endpoint." Linked Data focuses on crawling the web. At least, that's the impression I have... yet, a single store to query is indeed much more convenient... it's sort of contradicting: I can appreciate Mark's comment, as we have yet to come up with good solutions for when to stop crawling and start analyzing the data... the crawling is iterative, and each new analysis step may trigger new queries (and probably should)... A single federated query is not what I expect to be the final solution; instead, I expect an iterative process, where possible steps may be federated, but iterative nevertheless... Having one single SPARQL end point indicates the crawling is done, where we have only just started linking things together... that said, I also don't think the final SPARQL end point should be remote at all, and it steps over the current data licensing issues we still (unfortunately) have to deal with... And then I just think about remote services calculating things on the fly, which will likely not be part of a SPARQL end point anyway... Mark, or perhaps things like SADI should have a SPARQL interface? :) Anyways... looking forward to meeting some/many of you in Amsterdam! Egon -- Post-doc @ Uppsala University Homepage: http://egonw.github.com/ Blog: http://chem-bla-ics.blogspot.com/ PubList: http://www.citeulike.org/user/egonw/tag/papers
Received on Tuesday, 13 October 2009 20:14:28 UTC