Not Just Semantics

Computerworld - How soon will we start seeing pilot projects implementing Data Center Markup Language, the proposed standard for automating data center management? This year, that's how soon. Last week, the first draft of DCML specs were unveiled. And now Electronic Data Systems, one of DCML's biggest backers, says it will be doing real pilots with end users within months.
That's good news for two reasons. First, with EDS racing to get DCML working for data center management, we won't have to wait
around for years until vendors someday sign up. DCML is being driven by people who actually run data centers. It'll be pragmatic, IT-shop-focused -- and real sooner instead of later.
And second, a fast-tracked DCML will give IT shops a taste of the Semantic Web inside our own glass walls before we have to deal with it in the outside world.
Remember what DCML will do. Right now, if you want to bring a new server online, you have to make configuration changes, set up monitoring and do lots of other adjusting. Your procedures for doing those things may be well defined or ad hoc, but they're almost always performed manually. That makes mistakes easier.
DCML promises to turn those procedures into sets of XML-encoded documents that pass the necessary information around your data center -- information about your procedures, the new equipment and your existing data center environment. DCML will automate data center processes and -- if it works -- make them faster and more reliable.
The hard part is getting it working -- and it's nice to have EDS to do a lot of the proof-of-concept heavy lifting.
Besides DCML making data center management more efficient, IT shops will get an extra benefit. DCML uses some of the same concepts and technologies as the Semantic Web. That's World Wide Web inventor Tim Berners-Lee's vision of the next generation of the Web. And IT shops really need to understand the Semantic Web soon -- before it breaks big into the mainstream.
What is the Semantic Web? In simple terms, the Semantic Web makes it much easier for machines to harvest the information that's currently embedded in the text of Web pages.
For example, today it takes either a human brain or complex machine intelligence to figure out from a toy store's Web site that a Happy Birthday Barbie doll costs $16.99, while a Swan Lake Barbie costs $19.99 and the Swan Lake Barbie Gift Set costs $44.99. The data's all there, but it's set up for people to