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] Keeping ISO 8879 Alive (was RE: [xml-dev] Markup perspecti

[ Lists Home | Date Index | Thread Index ]

Hello Mike

[mike statement about infosets vs syntax...]

Mike thank you for your message, it summarizes, in my own opinion, what
is happening right now in the SGML/XML world.

When XML is used as syntax it leads to things like RPCs or any other
usages in need of a hierarchical syntax. What we need more is semantics,
not the semantic web which seems to be still in dreamland but more
document types, structures allowing the exchange of information.

When people look at XML as an integration/information exchange tool they
may favor:
a) integration by processes (i.e. function calls)
b) integration by data (i.e. documents)

Since the last 23 years I am in the software business I saw developers
attracted by these two poles.

We have to admit that the XML framework is not well adapted to the
actual developer's programming languages if they want to write a
solution's semantics without too much overhead. We have to agree that
the following expression is a lot more easier to use, comprehend and
read than a full SAX parsing routine( account.balance = account.balance
+ deposit) In the expression between parenthesis we keep the semantics
of the problem to be resolved. Everything the developers has to do in
addition to this expression is simply overhead, with or without XML.

I believe that people developing the XML framework (i.e. W3) missed a
great opportunity to help these people by providing an infoset interface
that kept the semantics. For instance an element
<account><balance>20.00</balance>...other elements...</account> could be
transformed into account.balance without further intervention from the
developer. Instead the XML framework people create an interface to the
syntaxic level of the infoset hierarchy. What the developer is handling
is not an element, since the element reveal nothing about the semantics.
But the keyword account mean something in the problem to be resolved.

What we see is probably the result of sticking to much to the syntax
level and not really trying to fulfill the developers needs. The full
implication is what we see today. 

I still like to think that " a different point of view is worth a
thousand point of IQ". I guess that our lack of intelligence to embrace
the developers' problems is due to the fact that we simply didn't looked
at the world from their perspective and what _they really want to
achieve_

By trying to evacuate the semantics from the XML framework we also
opened the door to schisms.

Cheers
Didier PH Martin






 

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

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