Xml Messaging

[Mark Baker]

>should XML messaging be based
>on a "metadata" language, or on a "domain" language?

IMHO, both.

>Method 2: "Domain" language
>-------------------------------------------------------------------
><message>
>  <update target="accounting" id="1">
>    <GlAccount oid="12345" version="1">
>      <ChartOfAccounts oid="47"/>
>      <GlAccount.name>Bank Account</GlAccount.name>
>      <GlAccount.type>Asset</GlAccount.type>
>      <Balance>
>        <Currency>USD</Currency>
>        <Amount>15000.00</Amount>
>      </Balance>
>    </GlAccount>
>  </update>
></message>

That's sort of both, in that "update" is a metastatement about what is
intended of the document (which appears to be comprised of the content
starting at the GlAccount tag).

Your previous method 1 example requires a translation process that appears
unnecessary.  Just dump the document (the account) over the wire with
sufficient context for it to have the desired effect on the other end.

><aside>
>These two methods contrast the "Xml-Rpc" type approach, which I feel
>is not appropriate for all areas of distributed client/server systems
>(in the same way I feel CORBA and COM are sometimes touted
>incorrectly).  This opinion is based on another which says that
>clients use data differently than servers, and so clients should not
>be exposed to a server's fine-grained APIs, nor the object model(s)
>thereof.
></aside>

Right.  Ditto for SOAP.  State transfer, not method invocation.

MB

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)