- From: Geoffrey Sneddon <gsneddon@opera.com>
- Date: Fri, 28 Aug 2009 10:49:46 +0200
- To: James Graham <jgraham@opera.com>
- CC: Simon Pieters <simonp@opera.com>, Dan Connolly <connolly@w3.org>, Ian Hickson <ian@hixie.ch>, public-html@w3.org
James Graham wrote:
> Simon Pieters wrote:
>> On Thu, 27 Aug 2009 16:54:45 +0200, Dan Connolly <connolly@w3.org> wrote:
>>
>>>> Is there any chance we can keep the numbers
>>>> in sync between the two copies?
>>>
>>> The easiest thing I can think of is: for each section that
>>> gets filtered out, keep a stub that's just a heading
>>> and some sort of note a la "for details, see the full spec".
>>
>> Maybe it's possible to first generate the section numbers and the ToC,
>> then cut out the .impl parts (in the ToC too), then generate xrefs.
>> This would get rid of broken links, too.
>>
>
> This is possible but it would be really confusing to users to have
> non-sequential section numbers. Although it is worse for us, I think I
> prefer the situation where the section numbers in the author view and
> the full spec don't match. This makes pinpointing feedback harder (but
> not too hard since a reporter can always give the section title and some
> context), but makes the document read more consistently. Failing that I
> think having stub sections that just say {section relevant only to
> implementors} would be alright.
I'd rather we kept the section numbers the same, and we had sections
with something like: "This section is intentionally left blank, as it
consists solely of implementation information."
--
Geoffrey Sneddon — Opera Software
<http://gsnedders.com/>
<http://www.opera.com/>
Received on Friday, 28 August 2009 08:50:33 UTC