W3C home > Mailing lists > Public > public-html@w3.org > May 2013

Re: [SPAM] RE: Question on "phrasing content" definition

From: Karl Fritsche <karl.fritsche@cocomore.com>
Date: Fri, 17 May 2013 10:03:29 +0200
Message-ID: <5195E451.5000200@cocomore.com>
To: Yves Savourel <ysavourel@enlaso.com>
CC: <public-multilingualweb-lt@w3.org>, 'Silvia Pfeiffer' <silviapfeiffer1@gmail.com>, 'Pēteris Ņikiforovs' <peteris.nikiforovs@tilde.lv>, <public-html@w3.org>
Hi,

On 17.05.2013 00:15, Yves Savourel wrote:
>> I'll try to go through the elements today and come with a possible
>> list a ambiguous cases and proposed defaults.
> Here it goes:
>
> Those three elements should probably be within-text='nested': <script>, <iframe>, <noscript>
+1

> I would tend to say <textarea> may also be within-text='nested'.
+1

> There are a few elements that are clearly within-text='yes' but may contain non-phrasing content. For example <select> can have <option>. So we end up with a non-within-text element inside a within-text element. I see two way to address those:
> a) Do nothing. Just report each element as it is.
> or b) Say that elements not within text that are inside of element within text must default as 'nested'.
> I would tend to go for a) (not say anything).
I also would tend to a as a default for the standard. But this is maybe 
something we can address further in a best practice document?

> For all others I would default on within-text='yes', knowing that there are some occurrences where 'nested' would be better. Three elements especially: <math>, <svg> and <canvas> have good arguments to be within-text='nested', but I think we still would be better with defaulting to within-text='yes'.
+1

Cheers
Karl

>
> The bottom line is that for many elements there is no way 'right' choice. But overall it usually hurts less to have sub-flow text inside another flow than to have non-sub-flow text cut out from its context, so erring on within-text='yes' is probably better.
>
> -ys
>
>
>
>
Received on Friday, 17 May 2013 08:04:10 UTC

This archive was generated by hypermail 2.4.0 : Saturday, 9 October 2021 18:46:02 UTC