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] What does SOAP really add?

[ Lists Home | Date Index | Thread Index ]

Dave Winer wrote:
> 
> OK, let's take this one step at a time.
> 
> 1. If others were shut out I didn't hear from them. Usually people scream in
> an unmistakeable way when that happens, they don't usually go quietly. This
> may be an exception to that.

Well in part that's what the fuss is all about. There's a reason that
the people talking about it are in xml-dev and not "decentralization"
for example. XML people want to apply XML tools to XML data sets. When
those XML data sets are hidden behind programmatic APIs that becomes
much harder and less scalable.

> 2. I think of SOAP as glue for connecting scripting environments. Always
> have. You may think of SOAP as something else. I'm only now getting that
> this has been a source of disagreement, even with people we collaborated
> with. Often we don't patiently explore the assumptions others are making. It
> leads to disconnects.

If SOAP were restricted to being a glue for scripting environments
nobody would complain about it. In fact I prefer XML-RPC in that role
but SOAP is okay. But I don't really see how the Google database (for
one example) or the UDDI repository can be considered "scripting
environments". If Google is a "scripting environment" then what isn't?
Oracle?

> 3. I liked the Google XML interface. I'm sorry it's behind a firewall now. I
> have tried, several times, in private exchanges with Google people, to get
> them to reinstate it, or at least let me try to program against it in test
> mode to see what it's capabilities are. In the middle of the last effort,
> they showed us the SOAP API. We got busy when that happened.

I would have preferred if you had pointed out at that point that
providing the information through URI-addressed XML would have been
useful to more people for more varied projects without excluding
anybody. Plus, more bandwidth and CPU efficient to boot.

I beat on this issue because I want people to stop and think about that
issue when they design these APIs. Today, they seldom do, but I think
the message is getting out.

 Paul Prescod




 

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

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