The Artima Developer Community
Sponsored Link

Weblogs Forum
What is XML?

34 replies on 3 pages. Most recent reply: Nov 22, 2005 5:37 PM by Chris Eidhof

Welcome Guest
  Sign In

Go back to the topic listing  Back to Topic List Click to reply to this topic  Reply to this Topic Click to search messages in this forum  Search Forum Click for a threaded view of the topic  Threaded View   
Previous Topic   Next Topic
Flat View: This topic has 34 replies on 3 pages [ « | 1 2 3 ]
Frank Iannarilli

Posts: 1
Nickname: frankeye
Registered: Oct, 2005

Re: What is XML? Posted: Oct 24, 2005 9:29 AM
Reply to this message Reply
Advertisement
One of the points of a markup-based interchange format is that a datafile can be extended (with additional tags) to version N+1 without disrupting version N parsing. Sorta' like sub-classing for data.

Ray Schweray

Posts: 2
Nickname: ngjans
Registered: Oct, 2005

Re: What is XML? Posted: Oct 24, 2005 9:18 PM
Reply to this message Reply
Good point; Other formats can hint at the shape of data equaly as well as xml.
In my experience XML's practice of separating elements and attributes is a little cleaner than most formats; analogous to how it is beneficial in an sql schema to know which tables represent entities and which represent relations. Unfortunately there is nothing but art keeping you from mixing elements and attributes or entity and relations in unholy ways.

Also I think I meant, "requires a resolver” rather than "requires a revolver" and "leafs" rather than "leaves". Hmmm, Maybe xml is taking a bigger toll on me than I thought.

Dan Creswell

Posts: 49
Nickname: dancres
Registered: Apr, 2003

Re: What is XML? Posted: Oct 25, 2005 4:42 AM
Reply to this message Reply
> One of the points of a markup-based interchange format is
> that a datafile can be extended (with additional tags) to
> version N+1 without disrupting version N parsing. Sorta'
> like sub-classing for data.

I'm not sure that's a property of a markup-based format.

I'd have thought it was a property of anything that had records which had a well known start and end token plus some kind of well-defined location for a unique id indicating the kind of record. I could accelerate parsing further by also including a length indicator allowing me to quickly skip the rest of the record. In fact, if I have a length indicator I don't necessarily need the marker tokens but that's another discussion.

If I have those things, I can ignore a record if I don't recognise it. Thus I see this as nothing more than a discipline for file formats or protocols or whatever rather than something inherent in markup languages. I could adopt this approach with a binary format for example.

V.H.Indukumar

Posts: 28
Nickname: vhi
Registered: Apr, 2005

Re: What is XML? Posted: Oct 31, 2005 11:35 AM
Reply to this message Reply
>> which is a lower quality of data than the RM. as one of my profs said: "there is more hierarchical data in programs than in the real world." Codd proved this. insistence is not equal to truth, justice, and the American Way.

Well... I agree that not everything is hierarchical. But /some/ are definitely hierarchical. So to say that each and every format must be relational in nature would be overkill. There are definitely use cases for hierarchical representation of data.

Chris Eidhof

Posts: 53
Nickname: chriseidho
Registered: Nov, 2005

Re: What is XML? Posted: Nov 22, 2005 5:37 PM
Reply to this message Reply
XML is just a format to store data in. In most of the documents, the semantics are not defined in the document itself. That means that for every document the application which is about to process it has to know the semantics of the document.

A computer cannot reason about XML-documents without knowing what it means. And, because there's no explicit meaning in most of the documents, there has to be an interpreter for every type of document: XHTML, SVG, etc.

One of the layers on top of XML, RDF, provides a framework for describing data in an organized way. You can describe objects, properties, subproperties, etc. You can link between objects. Also, with an ontology-language like OWL you can define the meaning of certain relationships and you can place restrictions on, for example, properties.

We're talking about the topic of semantics now, and this is widely studied in, for example, AI. If your domain gets more complex, your semantics will be come really complex too. It's just a very difficult problem, and the W3C's approach with RDF and OWL solves a part of the problem, but doesn't try to be complete.

You should really read: http://www.w3.org/2001/sw/.

Flat View: This topic has 34 replies on 3 pages [ « | 1  2  3 ]
Topic: Upgrading Jini lookup to include more explict control and flexibility Previous Topic   Next Topic Topic: Generics: Unbounded wildcard puzzle


Sponsored Links



Google
  Web Artima.com   

Copyright © 1996-2014 Artima, Inc. All Rights Reserved. - Privacy Policy - Terms of Use - Advertise with Us