[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: RDF vs. SOAP serialization (oh yeah, and XMI and XTM)

• From: Michael Fitzgerald <mike@w...>
• To: David Megginson <david@m...>, xml-dev <xml-dev@l...>
• Date: Fri, 22 Dec 2000 09:21:23 -0800

What do you exactly mean by "the SOAP spec mixes protocol and format"?

-Mike

-----Original Message-----
From: David Megginson [mailto:david@m...]
Sent: Friday, December 22, 2000 9:08 AM
To: xml-dev
Subject: RDF vs. SOAP serialization (oh yeah, and XMI and XTM)


Dave Winer writes:

 [on static data representation]

 > If that's what RDF is going to be used for I'd strongly recommend
 > using the serialization format in XML-RPC or SOAP. They both work
 > incredibly well, and are supported in lots of environments and are
 > understandable, very simple stuff that allows data structures to be
 > exchanged between apps on all platforms. Something to think about? 
 > Dave

I don't think that XML-RPC is really in the running -- I'd suggest
that the main candidates for a general-purpose XML data format right
now are RDF and SOAP (as Dave mentions), with XMI and XTM running a
very, very distant third.

Both SOAP and RDF are fairly easy to use (50 points each), both have
good free cross-language tools and libraries (+20 points for RDF,
which has a lot more, and +10 for SOAP), and both have enthusiastic
and active developer communities (+20 for RDF, +10 more for SOAP,
which has a smaller community because it's newer).  Both specs take
advantage of XML Namespaces rather than using less common partitioning
mechanisms like XMI and XTM do (+10 to each for interoperability).
RDF has virtually no high-profile corporate backing, while SOAP has a
lot of public support from IBM and Microsoft (+20 for SOAP, but then
-5 because IBM also backed RDF and then XMI before changing their
minds again, so their support isn't worth much; 0 for RDF).

On the other hand, both specs suffer from painfully muddled design --
the RDF spec dumps reification and knowledge-representation junk on
top of what would have been a simple foundation, while the SOAP spec
mixes protocol and format (-20 each).

According to my very arbitrary scoring, then, we have RDF finishing at
80 points and SOAP at 75 points, which is a virtual tie -- SOAP could
end up winning on electoral college votes, depending on how many
people implement it in Florida and what the Supreme Court thinks.

If SOAP gathers some momentum, it could pass RDF quickly, but right
now, I see at least five RDF announcements for every one SOAP
announcement.  I'm not willing to call a winner yet.


All the best,


David

-- 
David Megginson                 david@m...
           http://www.megginson.com/

• Follow-Ups:
  • RE: RDF vs. SOAP serialization (oh yeah, and XMI and XTM)
    • From: David Megginson <david@m...>

• References:
  • RDF vs. SOAP serialization (oh yeah, and XMI and XTM)
    • From: David Megginson <david@m...>

• Prev by Date: Re: RDF vs. SOAP serialization (oh yeah, and XMI and XTM)
• Next by Date: Namespace Definitions: Cracks!
• Previous by thread: Re: RDF vs. SOAP serialization (oh yeah, and XMI and XTM)
• Next by thread: RE: RDF vs. SOAP serialization (oh yeah, and XMI and XTM)
• Index(es):
  • Date
  • Thread