Re: [Bug 7695] Conformance

Henry Thompson wrote:

> were always intended simply to make it easy for processors to 
> document what they _expose_, not to let them off the
> hook as regards implementation.

I think that most of the confusion here is that I might not have chosen my 
words as carefully as I should;  the main point was intended to be that we 
need to clarify the relationship between the statements quoted from 2.4 
and C.1, and it seems that everyone who has spoken has at least some 
sympathy with that comment.

As to what you say above, I almost agree, but not quite.  They way I would 
say it is:  each conforming processor is written to take certain inputs 
and to produce certain outputs.  The exact form of those outputs is beyond 
the scope of the XSD Rec (e.g. is it a Java API, a text file with results, 
etc.), but we do encourage you to use the terminology in C.1 to document 
what your processor exposes.  I expect we agree on that much, or I hope we 
do.

As to what you implement internally, my answer would be along the lines 
of: wrong question.  It is often the case that the only way to correctly 
compute what you're exposing is to build up information that is isomorphic 
to what's in the components, but there may sometimes be other ways to do 
it (or more likely, there may be ways to do it that involve computing only 
some of the information required by the pertinent components), and I think 
that's up to the implementor.  Expect in cases where processors choose to 
expose something isomorphic to the conforming components, then any means 
you can find to compute the correct bits of the PSVI that you expose is 
OK.  Of course, our documentation of how to do that is all in terms of the 
components.  Do you disagree with that way of looking at it?

Noah

--------------------------------------
Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------








ht@inf.ed.ac.uk (Henry S. Thompson)
Sent by: www-xml-schema-comments-request@w3.org
10/08/2009 04:55 AM
 
        To:     "C. M. Sperberg-McQueen" <cmsmcq@blackmesatech.com>
        cc:     www-xml-schema-comments@w3.org, (bcc: Noah 
Mendelsohn/Cambridge/IBM)
        Subject:        Re: [Bug 7695] Conformance


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

C. M. Sperberg-McQueen writes:

> . . .

> I do not remember anyone ever suggesting that minimally conformant
> processors are or should be required to expose the entire PSVI, or
> assuming that position in building other arguments.

I absolutely agree with Michael here.  The distinction between
"generate' or 'implement' on the one hand, and 'expose' on the other,
has always been fundamental.  The changes we made in 1.1, as reflected
in appendix C, were always intended simply to make it easy for
processors to document what they _expose_, not to let them off the
hook as regards implementation.

Accordingly a lot of Noah's message feels to me like a
misunderstanding.  Which is not to say that the various references to
'minimally conforming' and 'conforming' shouldn't be clarified, just
that full PSVI _implementation_ is not at issue for _any_ level of
conformance.

ht
- -- 
       Henry S. Thompson, School of Informatics, University of Edinburgh
                         Half-time member of W3C Team
      10 Crichton Street, Edinburgh EH8 9AB, SCOTLAND -- (44) 131 650-4440
                Fax: (44) 131 651-1426, e-mail: ht@inf.ed.ac.uk
                       URL: http://www.ltg.ed.ac.uk/~ht/
[mail really from me _always_ has this .sig -- mail without it is forged 
spam]
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)

iD8DBQFKzakVkjnJixAXWBoRAiObAJ9tCgpR3lufWxR/PhgTkU+N9d5JiwCePj3L
hBYFu6vSuvvPlAy4ULiEM6U=
=QvA5
-----END PGP SIGNATURE-----

Received on Thursday, 8 October 2009 14:45:11 UTC