Mr. Buzz,
I will counter, and perhaps I will convince or we will 'agree to disagree'.
Human readable has nothing to do with the value of XML, as it is focused on the software applications understanding each other. The value of XML is not as a standard, in and of itself, but for a simple way to describe various standards.
And the Standards are emerging, driven by the 'big kids', ARBA with cXML, CMRC with CBL, etc. And the committees are working with a superstandard (DTD's, XML Schema, etc) to alleviate some of the object definition issues you raise with <EMPLOYEE>.
The explicit use of a GT, LT signs around a tag, like <DATE> are arbitrary, it could just as easily be ?,# or *. The revolution is not in the describing or the descriptors, but in standardizing the way to set up the descriptors. Instead of your Notepad example of 0010050000106, which is indecipherable, except by predetermined agreement between you and I as describing a payment of $100.50 on January 6, 2000.
XML would tag each individual field. Making it recognizable and interpretable very quickly by a stupid computer, without requiring intervention by a slow, but intelligent human.
Also, I feel your arguments are proprietary information are spurious, as Internet security is already available for these documents through SSL and Digital Certs. And performance is not an issue, any 'pipe' that can carry a picture or a movie or a music file, will have no problems with a measly text document.
I will give you that XML is not nirvana, but it is clearly a building block IMO. |
