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] loosely and tightly coupled systems and type annota tion

[ Lists Home | Date Index | Thread Index ]

It is not well understood because it is being 
misrepresented in academic settings, articles, 
conference papers, and on the lists.  XML is 
tossed around as a term where the system 
frameworks, the application languages and the 
XML 1.0 specification are treated as if they 
were one thing, borne of the W3C and to be 
followed because of the hegemony of the W3C.
This is seriously nuts.

<btw>I know you know these things.  We're 
on the same soapbox.</btw>

Organizations do have ways to enforce these 
things.  Trademark ownership, interlocking 
specifications within and among organizations, 
power of the press, power of the procurement 
agencies, power of local management, power of 
individual application requirements, all of 
these and more are forces.  The trick in this 
is to understand when the use of force is 
necessary and when it is based on superstitions. 

The first and most vital understanding is to 
know that the only choice removed from the 
table by the greatest number of agreeing 
parties is that XML 1.0 is core.  Everything 
after that is a separate negotiation of 
requirements to be applied to individual 
works.  Layers are options chosen to get a 
certain job done.  Without a clear requirement 
for the job, the layering will be ad hoc 
and ad hoc layering is fertile for superstitious 
nonsense, gold plated deliverables and the 
rest of the complexifying, resource draining, 
brain numbing overkill of XML system specs.

We need layers, and we have them.  We seem 
hellbent on making more.  We should be sure 
when and for what we need them.  We should 
understand why XML Query wants strong types 
and know how to apply these to databases. 
We should understand that not all applications 
of XSLT and certainly XML require strongly 
typed databases.   This isn't hard stuff. 
As far as I am concerned, XML Query should 
have them.  XSLT should not as long as it 
is also understood that XSLT will not be a 
query language for these.

Otherwise, have at.


From: Eric van der Vlist [mailto:vdv@dyomedea.com]

But don't consider that any organization can "own" ideas either and have
any way to force people in a direction they don't want to follow :-)



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

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