Git along, li’l kitty
Mark Pilgrim sensibly repudiates the idea that the entire web will ever live up to the ideal of the Semantic Web proposed by the W3C (an idea others mistakenly thought Pilgrim was an advocate of). He rightly claims that creating a semantic web is difficult even with a limited problem domain. In a previous job, we were attempting to create a semantically-rich SGML/XML-based representation of a documentation set for a complex product, to be rendered in semantically-correct HTML.The group I was working with had been working on this for more than two years when I was laid off last December, and based on what I’ve heard talking to a friend of mine who still works there, they’re still working on it a year later. It’s a tough problem. On top of the fact that you’ve got to define standards, then implement the standards, you’ve also got to get the people creating the documents to do so in a way that feeds into your system properly. It was difficult enough to do that with a few dozen writers who you could order to work in that way; you’re paying them to do so, after all. Doing so for the entire web is simply impossible. It just won’t happen. A problem like that makes herding cats seem like child’s play. So I agree that it’s great to create semantically rich sites for ourselves (I work in much the same way Mark does on that count, particularly on this site), but to expect the rest of the world to fall in line just isn’t realistic.
Posted at 10:36 AM