W3C home > Mailing lists > Public > xml-dist-app@w3.org > December 2000

Re: [DR 309] Vague?

From: David Ezell <David_E3@Verifone.Com>
Date: Fri, 1 Dec 2000 09:58:52 -0500
Message-ID: <472E220BA79DD11186340060B06B38D9033AD106@tpantmail1.ssr.hp.com>
To: xml-dist-app@w3.org
Tue 11/14/2000 9:36 AM Ray Denenberg wrote:
> Noah Mendelsohn wrote:
>> "In cases where there is prior knowledge of the specific interactions that
>> will arise between given XP implementations, it should be possible to
>> create implementations supporting these interactions using only a minimal
>> amount of XML infrastructure."
>>
>> I cannot tell what this really means to say.
>
>What I read this to say is that as many parameters of the envelope as possible
>should be defined as optional, so that two parties who have a prior agreement
on
>a given parameter's value should not have to explicitly include the value in
the
>envelope just to be compliant with the protocol.

Sorry, but making parameters optional is not at all what this requirement was 
supposed to say.  Unfortunately this item fosters confusion.

The editorial team has tried (unsuccessfully) two or three times to clarify this

item, and we have introduced one more proposed wording into the comments:

>[DR309]
>In cases where the contract between entities is well known, the use of XP as a
protocol 
>to fulfill those application contracts should allow processing without
requiring a 
>complex XML application infrastructure provided the documents exchanged are
well-formed 
>and within the tenets of the XML Infoset.

What we're trying to say is this:  if you only have 256 bytes (not kbytes,
bytes) of
memory, you should be able to *entertain* the idea of using XP as long as you
make
the exchange really simple.  You shouldn't have to have Xerces-J or some other
full
featured parser.   We don't want people to dismiss XP as a possible protocol
just
because they can't "afford" an XML parser. 

If the community doesn't like the proposed wording above, the feeling of some of
the editorial team is to drop this requirement from the requirements document.
Unless, 
of course, someone can propose a clearer wording.

Thanks,
David
Received on Friday, 1 December 2000 12:07:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:57 GMT