Re: conformance to "software MUST NOT assume" measurable?

Hello Dan,

This is a personal answer to your comment. I'm copying Dominique,
who has supported your comment, and the I18N IG, for discussion.

At 08:27 04/03/19 +0900, Dan Connolly wrote:

>This is a last call comment from Dan Connolly (connolly@w3.org) on
>the Character Model for the World Wide Web 1.0
>(http://www.w3.org/TR/2002/WD-charmod-20020430/).
>
>Semi-structured version of the comment:
>
>Submitted by: Dan Connolly (connolly@w3.org)
>Submitted on behalf of (maybe empty):
>Comment type: substantive
>Chapter/section the comment applies to: 3.2 Units of aural rendering
>The comment will be visible to: public
>Comment title: conformance to "software MUST NOT assume" measurable?
>Comment:
>Regarding:
>
>C001   [S]  [I]  [C]  Specifications, software and content MUST NOT assume 
>that there is a one-to-one correspondence between characters and the 
>sounds of a language.
>
>How does one test/measure/observe/demonstrate that? Would you please
>point me at a test case?
>
>I think it's fine to write:
>
>   Take care not to assume a one-to-one correspondence between
>   characters and sounds of a language.
>
>followed by the examples you give,
>but I don't see how making this a conformance clause is helpful.
>
>This applies to C002 and C003 as well.

I'm not sure why C003 ([S]  [I]  [C]  Protocols, data formats and
APIs MUST store, interchange or process text data in logical order.)
ended up in here; it's definitely extremely observable, for protocols
for example by sending some Arabic or Hebrew over the protocol and
using a hex dump to check that the characters indeed come in the
right order. Can you explain why you lumped it in together with
the others?

C001 and C002 ([S]  [I]  [C]  Specifications, software and content MUST NOT
assume a one-to-one mapping between characters and units of displayed text.)
are indeed of a very similar nature, and it makes sense to discuss
them together.

With respect to conformance, it is important to remind ourselves that
all these criteria are predicated by the definitions in Section 2,
such as (simplified):
   A specification conforms to this document if it does not violate
     any conformance criteria preceded by [S],
(and similar for implementations and content).

That means that for specifications, software, or content that do not
deal with aural rendering or recognition (C001) or visual rendering
or recognition (C002), the respective criteria are simply not
applicable.

And if you look at technology that indeed deals with aural rendering
(e.g. SSML) or visual rendering (e.g. SVG),..., you will be able to
verify (by reading the spec, and testing whether the implementations
conform to the spec,...) that indeed the don't make such assumptions.
As an example, in SSML, the <phoneme> element works on strings
(words,...), not characters (see
http://www.w3.org/TR/2003/CR-speech-synthesis-20031218/#S3.1.9).
As another example, <altglyh> in SVG allows to define alternative
renderings for groups of characters, not only for single characters.


So these criteria are very definitely observable. Of course, because
of the architectural level of the spec, it's not as easy as just
creating a test case.


Regards,    Martin.

Received on Saturday, 20 March 2004 15:28:12 UTC