[
Lists Home |
Date Index |
Thread Index
]
- From: David Megginson <david@megginson.com>
- To: xml-dev@lists.xml.org
- Date: Tue, 10 Oct 2000 09:15:09 -0400 (EDT)
Joshua Allen writes:
> > XML itself is like IP -- it's a good foundation, but it's too raw to
> > give us interoperability. We need to build the equivalents of TCP,
>
> I disagree; I've used XML to build connections between MVS, UNIX,
> and NT systems that have survived system and software upgrades with
> little interruption.
Congratulations -- during the 80's, similarly talented coders used to
write quite good word processors and spreadsheets in assembler code as
well. Older coders wax nostalgic from time to time, but I don't see
any of them uninstalling their widget libraries.
Of course you can make a data-interchange system work now -- you don't
even need XML to do that -- but making it work and then keeping it
working can be expensive, especially once the XML specialist moves on
or gets hit by the proverbial bus.
[snip: my complaint about SOAP's layering violations]
> Could you elaborate? People find it easy to think about SOAP in
> terms of RPC and HTTP, but the spec is very agnostic about these
> things. I have usually heard people complain that SOAP doesn't
> attempt enough. Maybe there needs to be some better augmenting
> documentation?
Data representation is one application domain, and RPC is another.
While RPC needs a data representation format, data representation does
not necessarily (usually?) need RPC. Putting both in the same spec
introduces unnecessary complexities and interdependencies.
All the best,
David
--
David Megginson david@megginson.com
http://www.megginson.com/
|