OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   RE: [xml-dev] DTDs, W3C Schemas, RELAX NG, Schematron?

[ Lists Home | Date Index | Thread Index ]

Thanks a lot betty, len, and michael.  The input and
advice is very informative and most definately
appreciated.  i guess in a nutshell, DTDs still carry
a certain significance and an amount of usefulness,
but the Schema+Schematron combination is a 'stronger'
approach to take in XML application development.

len, i will definately check out the resources that
you suggested.


--- "Bullard, Claude L (Len)" <clbullar@ingr.com>
> DTDs are not a "must have" in all contexts. 
> They have a primitive set of datatypes but 
> are a reasonably good means to declare 
> names and organize them into structures.
> DTDs are used to specify application languages 
> such as XML Schema but well-formed processing 
> enables one to start from XML Schema without 
> using DTDs.   I guess the question is, do you 
> need to validate the XML Schema language itself 
> before you use it?   The answer is, not usually 
> but a DTD can be used to design and validate 
> a language like XML Schema which can then be 
> used to validate applications of XML Schema.
> Say you want to design a new parts language. 
> You will want to validate your design.  Once 
> done, someone using it probably doesn't want 
> to do that.  You can start from DTDs, or you 
> can start higher up from XML Schema.  How 
> much validation do you need?  For example, 
> Schemas will do more than DTDs, but not as 
> much as Schemas plus Schematron.  You 
> can do it all with Schematron, but that 
> could be painful.
> Yes, DTDs aren't as strong as XML Schema or 
> RELAX NG for validation.   They are a way 
> to bootstrap something like XML Schema or 
> RELAX NG.   Then you might still need 
> schematron or a means to enforce business 
> rules, and that could just be code.
> BTW:  you  might want to look into the 
> new proposals for pipelining as a means 
> to precisely specify when and in what 
> order all of this occurs.
> Also, ISO DSDL (don't know, haven't looked) 
> may prove to be the best combination of 
> all of these.  XML by serving as a one 
> size fits all systems connector is forcing 
> all of the systems being connected to be 
> be rethought.  That is the price for 
> being On the Web. 
> len
> From: tariq abdur-rahim
> [mailto:ecliptoid330@yahoo.com]
> Now, to continue.  So, is it safe to say that beyond
> the realm of "basic validation" DTDs are weak in
> terms
> of usefulness?  Or is it still a *must-have* in all
> and any XML development; irrespective of whether or
> not it is employed in the development of XML
> applications, XML systems, or applications that uses
> XML applications?


T. A b d u r - R a h i m
W e b  D e v e l o p e r

More sacrifice, 
creates better living.

Do You Yahoo!?
LAUNCH - Your Yahoo! Music Experience


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS