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

Re: On Schemas, Namespaces and Syntax vs. Semantics (longbut w

• To: Dare Obasanjo <dareo@m...>
• Subject: Re: On Schemas, Namespaces and Syntax vs. Semantics (longbut worth it) :)
• From: Amelia A Lewis <amyzing@t...>
• Date: 27 Jul 2002 23:21:48 -0400
• Cc: xml-dev@l...
• In-reply-to: <8BD7226E07DDFF49AF5EF4030ACE0B7E06621E4C@r...>
• References: <8BD7226E07DDFF49AF5EF4030ACE0B7E06621E4C@r...>

I largely agree.  One note, however, worth making.

On Sat, 2002-07-27 at 22:21, Dare Obasanjo wrote:
> a.) PI solution: If some executive fiat makes it mandatory that only raw customer data must sent which conforms to the shared schema then processing instructions can be embedded in the SOAP message which then indicate to the final receiver of the message what to do with each customer. 

Unreliable, depending upon the version of SOAP in use.  Although
permitted at last in 1.2 (unless this changes again), PIs have generally
been either forbidden or strongly deprecated in earlier versions (1.1
being the current widespread implementation).  Even in 1.2, there's a
good deal of antipathy to this usage.

The question of whether PIs should *ever* be used seems to be in process
of resolution on the side of a resounding "no."  My initial reaction to
this was to bristle, because, for example, if PIs were supported in HTML
we wouldn't have so much severely crufty executable-comment embedded
languages about.  Instead we'd have severely crufty PI embedded
languages.  Still, that's at least arguably an improvement.  But for
XML, because of its inherently extensible design (which indeed makes XML
without some sort of extension no language at all), doesn't really need
PIs, it just needs proper extension language (or "vocabulary" or
"schema" or "dialect") design.

I agree, on that basis, that your solution c is appropriate--define the
customer block as an external schema, and import into money-losing and
money-grubbing schemas for appropriate departments.

Amy!
-- 
Amelia A. Lewis       amyzing@t...      alicorn@m...
    Merchant, street girl, beggar, yeoman,
    king or common, man or woman,
    only two things make us human--
    sorrow and love, sorrow and love ....
                -- The Last Song of Sirit Byar

This is a digitally signed message part

• References:
  • On Schemas, Namespaces and Syntax vs. Semantics (long but worth it) :)
    • From: "Dare Obasanjo" <dareo@m...>

• Prev by Date: On Schemas, Namespaces and Syntax vs. Semantics (long but worth it) :)
• Next by Date: RE: URIs harmful
• Previous by thread: On Schemas, Namespaces and Syntax vs. Semantics (long but worth it) :)
• Next by thread: Re: On Schemas, Namespaces and Syntax vs. Semantics (longbut worth it) :)
• Index(es):
  • Date
  • Thread