XML in the Wild

I had originally planned for the next entry in this blog to be about mousetraps – who’s got the best mousetrap, why is so-and-so’s mousetrap better than your mousetrap, why is my mousetrap better than yours, etc.

That’ll have to wait for another time, because frankly, I’m still turning the subject over in my mind to the point that I’m more likely to rant freely, instead of writing rationally, on the subject... Oh... and I simultaneously endlessly confused and bored my co-worker on the whole mousetrap thing on our way home from XML 2006... Sorry!

I’m glad I waited on the rant – I can always rant another day. In the meantime, I found a much more interesting thing to write about: Jon Bosak’s closing keynote address to the XML 2006 conference, which has just been published over at the XML 2006 Conference Proceedings web site.

I knew I was going to write about this as soon as I read the text. I also know at least one other person has already blogged about specific parts of Jon’s address. Specifically, other bloggers have chosen to address the merits and values of XML-based technologies, and whether or not vendors are hindering the creative spirit that led to XML in the first place. This point, combined with my own recent research, really hit home for me. I had one of those “YES!” moments when I finished reading.

I’ve been doing a lot of reading on the Semantic Web, Information Architecture, and other library science-ish topics lately. Specific titles include Explorer’s Guide the Semantic Web by Thomas Passin and Ambient Findability, by Peter Morville. I had already finished the first book and was most of the way through the second when Jon’s keynote was posted. The keynote started out with a reminiscence of SGML past. Quite naturally, my thoughts started tripping down a sort of Dickensian broth of past, present, and future. But not so much in context of specific technologies or standards. Both Passin’s and Moorville’s books captured my imagination with visions of what computers can do with semantic information in the future – both with and without human input; which of course, triggers the catastrophist side of me to hope that we won’t create the next HAL 9000..., or at least that we learn from our speculative fictional mistakes as well as our real ones...

But I digress. The point is, none of the speculative problem-solving solutions will ever happen if people don’t think past current technology, or totally forget to think creatively about solutions even with the restriction that we are boxed in by what vendors are willing to offer.

Which leads to my big gripe about the XML 2006 conference: There was simply too much focus on technology, and not enough creative business problem analysis. The fact is, when it comes to standards and recommendations, there are already too many technologies and solutions looking for problems to solve. Oh, and the whole idea that each of these technologies and solutions be XML-based in syntax? That may be getting out of hand, too, as other people have started to compare new technologies written in XML syntax to Kudzu. It’s never a good thing when something gets compared to an alien invasive species of vegetation.

Anyone can fight over which technology reigns supreme, and they are welcome to it. But fights generally don’t solve business problems, and one thing I learned as soon as I hopped over the fence dividing the vendor/consultant pool to the customer pool, is that us customers have problems to solve. Ideally, technology should be something that gets applied to solving the problem, after you know what the problem is.

Well, actually, technology is irrelevant and second choice to problem identification and solution until you run into a technology barrier that prevents you from doing things the way you really want. In the real world, creativity does get constricted by vendor-imposed limits. For example, Microsoft Word supports valid custom XML if you attach a W3C Schema to your document. What if you don’t want to use Schema? What if you want to use the other XML features of Word AND still get valid XML without W3C schemas? Yeah, I get that I could write my schema in RELAX NG and transform it to W3C Schema, but the downside of this is that it adds an extra step to my workflow AND doesn’t solve the problem that I needed RELAX NG to solve in the first place, since you can’t create a valid W3C XSD file from a RELAX NG file that effectively breaks the W3C Schema rules. So much for getting around the famous UPA rule (Unique Particle Attribution). I can’t apply thinking about which standards could possibly solve my problem here, because even though I supposedly have a choice in tools, the fact of the matter is that my workflow requires the use of Microsoft Word 2003.

But going back to the “climber up to the guru on the mount” perspective, I totally get what Jon Bosak’s saying. The creative part of what you can do with XML – that is, thinking speculatively about problem-solving (think big picture – semantic web stuff of the future, other ways to manipulate textual content... the list goes on...) with XML-based technology, should not be stifled by what vendors pick and choose to support... It’s up to vendors to look at business problems, AND potential solutions, AND ALL technologies that might contribute to an appropriate solution. It’s up to customers to remember to hold vendors to this task. Just because you’re a customer doesn’t mean you should stop thinking like an early adopter. Finally, it’s up to leaders like Jon Bosak to make sure we don’t forget the past, while working in the present, and dreaming of the future. Good thing he’s there to remind us.

Labels: XML2006

// posted by Jean K. @ 4:41 PM