I was just reading Claire Ross’s latest post which is about integrating visitor interpretations as part of the museum experience and how that experience might be gauged for future improvements. In the course of her discussion, she uses the phrase “exhibition labels” which made me immediately think of markup. Though I natter on occasionally about using markup for research, my feet have only just gotten wet; I don’t pretend to know a whole lot about it. However, thinking of markup while reading about her idea for an organic museum experience by way of visitor interpretations caused me to think about issues with semantic markup. That is, is there a such a thing as dynamic markup? I suppose that doing xsl transformations on xml is dynamic in a sense, but from what I understand, it’s still non-dynamic in the sense of using preprogrammed selections that can be run dynamically. What I really am asking, I think, is if there is such a thing as organic markup–markup that can be fed back into the original markup to grow it, rather than making use of pre-interpreted markup–a crowd-sourced, on the fly, sort of markup? The reason I’m wondering is, that I think it could help future interactions of previously marked up texts–as a way to evolve with future interpretations of not only the text, but of the tag set used to mark up that text. I’m guessing that that is what natural language processing folk are dealing with–trying to interpret the text instead of the tags. Of course, I know even less about that group. But I would imagine that such organic markup could aid natural language processing . It just seems to me that something like this might treat the interpretive act of marking up text more as conversation rather than a monologue by one person/project team.I can’t help but think that people have already been working with such an idea. Is Wikipedia really this kind of markup?
In the interest of full disclosure, before reading Ross’ post, I had also just watched this video by Barry Ridge, who is a Ph.D. student at the ViCoS lab, showing how their robot, George, used interactive learning to create knowledge updates (basically, how they started off with a simple knowledge schema and slowly grew it by way of his asking questions of the humans). Very cool stuff. And another thing I think is cool about it is that I think the core of what Ridge is doing is also what Ross is getting at (but in terms of a different discipline). It also shows how my procrastination tends to guide my reading into very cool things… Oh, the positive reinforcement!