Re: Audience Based Validator User Interface (ISSUE-206)

Hi Ben,

I would suggest replacing the current MUST emit conforming HTML5 documents
with 'MUST allow and should encourage authors to produce conforming
documents'.
I think this is a much more realistic and attainable goal for authoring
tools. The current requirement, unless some reference implementations are
available ,should be considered for removal before HTML5 reaches rec.

Note that the requirement you quote is already qualified in various
> ways. If we think additional qualifications are justified, can't we
> add them? For instance, the spec includes this qualification:
> "Authoring tools are exempt from the strict requirements of using
> elements only for their specified purpose, but only to the extent that
> authoring tools are not yet able to determine author intent. However,
> authoring tools must not automatically misuse elements or encourage
> their users to do so." We could expand this to "elements and
> attributes".
>

I am aware of this and am not talking about the cases where author intent
is not known or not discernable.  I agree that as much guidance as is
necessary is provided to authoring tool vendors , but to say "HTML5
authoring tools MUST NOT emit documents that do not conform to HTML5" is
taking theortical purity to its limits. It is just not practical in any
sense to expect any authoring tool to abide by this condition. If I am
incorrect in this assumption I am happy to be disabused.


regards
SteveF


On 5 August 2012 15:48, Benjamin Hawkes-Lewis
<bhawkeslewis@googlemail.com>wrote:

> On Sun, Aug 5, 2012 at 3:26 PM, Steve Faulkner <faulkner.steve@gmail.com>
> wrote:
> >>  "HTML5 authoring tools MUST
> >> NOT emit documents that do not conform to HTML5" to "HTML5 authoring
> >> tools SHOULD NOT emit documents that do not conform to HTML5". The
> >> allowed exceptions would be content author supplied attribute values,
> >> which are engineers control.
> >
> >
> > I think this would be a good change, regardless of issue 206, I do not
> > believe that this must level requirement is realistic or practical.
> > Any editing tool other than in the most controlled of environments can
> emit
> > non conforming documents. I would go so far as to say there is no editing
> > tool which  can guarantee that the output will always be conforming.
>
> I think that when adjusting MUST-level requirements we should try to
> replace them with more specific MUST-level requirements before
> replacing them with SHOULD-level requirements.
>
> Note that the requirement you quote is already qualified in various
> ways. If we think additional qualifications are justified, can't we
> add them? For instance, the spec includes this qualification:
> "Authoring tools are exempt from the strict requirements of using
> elements only for their specified purpose, but only to the extent that
> authoring tools are not yet able to determine author intent. However,
> authoring tools must not automatically misuse elements or encourage
> their users to do so." We could expand this to "elements and
> attributes".
>
> I've suggested elsewhere that's some requirements could be required to
> make a "complete" or "self-consistent" document (e.g. idrefs to unique
> elements that haven't yet been added to a complete document would fail
> that requirement). Possibly this concept would be useful here too.
>
> --
> Benjamin Hawkes-Lewis
>



-- 
with regards

Steve Faulkner
Technical Director - TPG

www.paciellogroup.com | www.HTML5accessibility.com |
www.twitter.com/stevefaulkner
HTML5: Techniques for providing useful text alternatives -
dev.w3.org/html5/alt-techniques/
Web Accessibility Toolbar - www.paciellogroup.com/resources/wat-ie-about.html