I am on my way home from a long trip in the US (writing these lines on the plane, to be posted from home). Few days in Seattle, SemTech 2010 in San Francisco, finally the “RDF Next Steps” workshop in Palo Alto (i.e, Stanford). I do not want to write about the last one now, simply because we hope to have a more extended public report available within 10-15 days. I.e., more about that later.
Seattle consisted of a number of company visits, but it also included a talk at the SemWeb Meetup in Seattle. I gave a presentation on what happened at W3C the last year which, I think, was was well received. (Although one is never sure about these things.) I had a bunch of discussions and chats after the presentation; it was pleasant, relaxing… I and mainly my colleague from W3C, Eric Prud’hommeaux, had also a long discussion with two developers from Microsoft who are involved in the oData work; that was really interesting because we reached the conclusion of possibly outlining together a possible plan whereby we could write down how to “export” oData into RDF, and publish that, e.g., as W3C note (note that there are already systems doing something like that out there, but I am not knowledgeable enough to judge how complete those solutions are). I think it would be good for the community if this happens. It is important for a general Web of Data to include, well, all the data on the Web…
Semtech… it was big. Bigger than last year (I heard and read a figure of a 30% increase in attendance). This industry is lively indeed! The only problem that it was almost too big; it was the conference of eternal frustration:-( Indeed, there were so many things in parallel that one always had the feeling to have missed something because another, parallel session may have been more interesting! I heard presentations from Facebook, from Google, saw stunning visualizations of RDF graphs, or heard about plans on ontology hosting and management. There was a report on the US and UK governmental data work (this stuff still amazes me, though it is not the first time I hear about it), there was a presentation of BestBuy (alas! I missed that one). There was a separate track on the publication world as a separate “vertical” area (and we also had some great discussions with the people from the New York Times with whom we outlined a possible first step in gathering that community). Lots of hallway conversation with companies and institutions and, of course the social life, chatting with David, and Ian, and the other Ian, and Eric, and the other David, and Christine, and Jeremy, and Jim, and Fabien, and Sandro, and Jenni, and… I should stop and not even try to list everybody because it is simply impossible! I also gave an introductory Semantic Web Tutorial (quite a lot of people in the audience, and I think it went well), we had a panel on the W3C RDB2RDF work and another one on SPARQL 1.1. As a nice little touch, I could announce the publication of the W3C RIF Recommendation as a primeur during the tutorial when as I was talking about RIF (the publication itself happened while I was talking…)
There were, as every year, some “buzz” topics. My impression that the linked open governmental data effort was a buzz and was still new information for many. Facebook’s keynote on the Open Graph Protocol crated another buzz. More generally, RDFa was definitely a buzz (big time!). I.e., as I said, this industry is lively and continue to be exciting.
But there are of course challenges. The way I feel it the biggest challenge is not technical. Yes, of course, there are technical issues, but those will be solved, eventually. The issue is outreach, to get to those new communities who may understand the value of a Web of Data in general but have not enough guidance on how to start doing something. How to publish the data, how to link it to other data, how to consume it, use it, mash it up… How to talk to “C-level” people, how to reach out to them. There are books, of course, but not enough; there are tutorials and guides, of course, but not enough; there are experts around but definitely not enough. As one of our discussion partners put it: if I go to any better bookshop, there are rows of books on, say, XML (good or bad, but they are there). But books on RDF, on Linked Data, on SPARQL, on SKOS, on OWL: only a few here and there (comparatively, that is), and some of them are actually quite old. Let alone the problem of trying to hire experts that could do the job. I really feel that this is the biggest challenge our community faces. I say “community” and not only a single organization like W3C or other; the challenge is too great to be solved by one group only. We have been fighting with this issue for a while now, but it is still a challenge… And a challenge for us all who care about that stuff!
It was a good week!