Questions about XML?

Hey! It aint that hard!
Post Reply
Posts: 5284
Joined: 12/06/01, 12:00 am
Location: here and there

Post by SOD » 03/18/03, 9:35 pm

I have been learning to use client side methods with XML. XSL XSLT and xpath are client side or passive methods to parse XML. According to most people XML allows you to organize data the way you wish and for the most part it lives up to that promise. But can XML be considered a catch-all for any data scheme?

An example would be weblogs. As the need to convey more information in weblogs the XML/RSS dtd becomes more complicated(or fragmented?) including CData that is HTML. Parsing such data suggests that you are parsing a string as HTML rather than data that can be addressed by a given dtd. Many script enabled XML APIs fall into this catagory. Attempting to fulfill a complex mechanism with out consideraton for XSL methods. Should data be addressed and parsed by a XML only structure? Does Hyper Text Markup belong in the tag construct of XML?

There is the need or want by some authors to include Hyper Links within the construct of a paragraph. Do links belong within a paragraph or tag entity with respect to XML? Or should links be a part of an array after the fact that has corresponding properties to the text block?

Is XML a means of addressing data and not just a means of simple organization to apply to code to?. Implying that XML is not as arbritary as it seems. With everything from RDF to rss to SAX(and everything in between) XML has been stretched to mean many things depending on a developers perception and rules. The developer makes the rules or API to which a given XML schema can function. Is this XML? Does XML have its own 'identity' and have people strayed form the nature of the beast? Employing code layers and rules to reduce it to fetching a string? Underneath the politics of XML and everybody wanting to come out on top of this ~new technology there is a natural rhythm to XML. Almost all data could be parsed by one schema that respects address, properties of the address and methods of the address. Link schemes could be universal and their properties considered pointers. XSLT and xpath would also be respected. Thus XSL methods and properties could be used as the driver to a given address. Making schema and dtd universal. Enhancing the very nature of XML. By nature XML is not ideological, yet developers have left their fingerprint. Obscuring the ubiquitous nature of a unique tool.

Post Reply