On 11/22/05, Danny Ayers <[EMAIL PROTECTED]> wrote: > By getting people to upload their data, Google Base bypasses one of > the harder problems of distributed data: finding the stuff. I'd be > interested in hearing of any ideas anyone has around here that might > be helpful for finding the stuff. > > Imagine there are lots and lots of microformat documents out there. I > want to get the data from a subset of these, e.g. restaurant reviews. > How do I do it? > > The strategies I've seen for data discovery to date have all left a > lot to be desired, e.g. in-the-wild links - require crawling; more > controlled link lists (e.g. on a Wiki or at a directory service) - > heavily centralised; UDDI etc - service oriented, hard work; > distributed RDF queries - potentially workable, but still a way off. > Others..? Or are there any obvious ways I've missed of reducing the > drawbacks of these techniques? >
Distruted query stuff is still a ways off, I think. Any centralized-directory-stuff should be the result of the crawl, IMO. We just need more meaningful links, better trust and identity systems, and good syndication technology. Speaking of meaningful links... I am sure this has come up before, so sorry in advance, but could we use rel values to say "there are x and y uFs over there". In other words, should there be a microformats discovery microformat? uF-links or some such? - Luke _______________________________________________ microformats-discuss mailing list [email protected] http://microformats.org/mailman/listinfo/microformats-discuss
