Re: HTML5 should not 'strongly encourage' authors to use only H1 level headinsg

Hi Robin,

Actually, the outline algorithm is currently up on the "at risk" list. So I
> don't think that we should be removing anything that relies on it without
> having removed the actual source.
>
> I therefore suggest that if we do remove the outlining algo at the end of
> the spec, we should likely remove this advice as well. But not sooner.
>
>
The text in question is non normative advice, whether or not the outline
algorithm goes or not is immaterial. The issue I am raising is that we have
bad advice in the spec, that is urging authors to construct outlines for
non existent consumers to the detriment of all exisiting consumers. we see
commits to the spec making editorial changes and clarifications etec on a
weekly basis, why do we need to wait a year or so to remove bad advice?


regards
Steve


On 26 October 2012 20:50, Robin Berjon <robin@w3.org> wrote:

> Hi Steve,
>
>
> On 26/10/2012 10:46 , Steve Faulkner wrote:
>
>> The HTML5 spec currently states the following advice [1]:
>>
>>
>>     "Sections may contain headings of any rank
>>     <http://dev.w3.org/html5/spec/**the-aside-element.html#rank<http://dev.w3.org/html5/spec/the-aside-element.html#rank>>,
>> but
>>
>>     authors are strongly encouraged to either use only |h1
>>     <http://dev.w3.org/html5/spec/**the-aside-element.html#the-h1,**
>> -h2,-h3,-h4,-h5,-and-h6-**elements<http://dev.w3.org/html5/spec/the-aside-element.html#the-h1,-h2,-h3,-h4,-h5,-and-h6-elements>
>> >|
>>
>>     elements"
>>
>>
>> Given that this advice is predicated on the implementation of the
>> outline algorithm in browsers which  has not occured and by all
>> anecdotal indications this is unlikely to happen soon if at all, I
>> suggest that it is not appropriate to include this advice in the spec.
>>
>
> Actually, the outline algorithm is currently up on the "at risk" list. So
> I don't think that we should be removing anything that relies on it without
> having removed the actual source.
>
> I therefore suggest that if we do remove the outlining algo at the end of
> the spec, we should likely remove this advice as well. But not sooner.
>
> --
> Robin Berjon - http://berjon.com/ - @robinberjon
>