April 20, 2003  ·  Lessig

Jon Udell’s got a nice piece about the emergingly Semantic Blog. One part he missed was the emergence of CC licenses as part of the semantic blog space. Movabletype and Userland now both incorporate CC license options. The technique we’ve used with html has been questioned, but we are pushing hard to get RDF out there.

What’s needed is a killer app, and here’s where the powerful need to be responsive. Google in particular has not been receptive to becoming RDF-aware, making it harder to build the amazing Google hack that would enable a search such as “give me all the pictures of the Empire State Building available under a non-commercial license.” Perhaps here’s where competition between AllTheWeb, or Yahoo and Google may do some good.

Some say the reasons Google remains Semantic-ly-unaware are dark — that in a world of articulate pages, it is harder to be the dominant search engine. I’m not willing to believe this yet. I suspect instead that it is up to us to spread the semantic word, and then others will find more reason to pay attention to it.

  • http://www.fsconsult.com/ Francois Schiettecatte

    Semantic blogs are a great idea, and I think that incorporating the license terms into the RDF/RSS file is a very good one. While I dont think that this is in the RDF spec ‘per se’ (I checked), you could add it through a module of some sort, say dublin core. The ideal way to do this of course is to define an element in the RDF/RSS file spec (this is the easy part), and provide some sort of coding to specify the license terms (this is the hard part).

    I run an rss search engine and it would be easy to add a search restriction based on a license terms field, allowing to perform just the search you mention.

  • d chalmers

    I wonder if Google’s reluctance to begin RSS/RDF-based searches has anything to do with search engines’ history of being misled by disingenuous <meta> tags (the Semantic Web markup of yesteryear). Certainly RDF markup (especially Creative Commons licensing) is less prone to abuse—after all, who wants to pretend that their work is available for free?—but I can imagine it factoring into Google’s thought on the matter.

    Of course, there is also the matter of slanted searching. Google already has a reputation (from the Register, for example) for being weblog-biased. Considering that the ‘blogspace is really the only area which has adopted CC licensing and RDF/RSS, and considering that an RDF-based search could only include websites with RDF information, results would be heavily pushed towards that side of the spectrum. And while weblogs are great for many things, when you’re looking for pictures of the Empire State Building or public-domain texts, you really want the World Book or the Gutenberg Project, not, say, Dave Winer. Google may be waiting for more “reuptable” sources to use RDF, who are waiting until it is to their advantage to invest in a project of that scope… Chicken-and-egg I suppose.

  • Jesse

    Chandler has nice blog reading capabilities built in already, maybe this sort of idea can be incorporated into Chandler in the future. (I read this entry from ZaoBao in Chandler — not ready for general people yet, but for hackers it is wonderful!)

    Combining this idea with the peer-to-peer nature of chandler and the ideas of swarm intellegence seems like an interesting idea!