W3C home > Mailing lists > Public > public-html@w3.org > February 2009

Re: summary="" in HTML5 ISSUE-32

From: Maciej Stachowiak <mjs@apple.com>
Date: Thu, 26 Feb 2009 11:24:50 -0800
Cc: joshue.oconnor@cfit.ie, HTMLWG <public-html@w3.org>, wai-xtech@w3.org, wai-liaison@w3.org, W3C WAI Protocols & Formats <w3c-wai-pf@w3.org>
Message-id: <F61642CA-5A9A-4CF6-8A74-9649867A3470@apple.com>
To: Philip TAYLOR <P.Taylor@Rhul.Ac.Uk>

On Feb 26, 2009, at 11:16 AM, Philip TAYLOR wrote:

> [The CC list was ridiculous; I've left in the groups,
> but removed all individuals apart from Maciej and Joshue]
>
> Maciej Stachowiak wrote:
>
>> So if such information is put in summary, it would not be  
>> equivalent. It would be providing information to non-visual users  
>> that cannot be learned from seeing the table. It seems that summary  
>> is used at least sometimes to convey such information. Would you  
>> agree that summary providing additional information (not  
>> information about table structure, or a summary of the table's  
>> conclusions, but brand new info that is not in the table at all)  
>> violates equivalence?
>
> But is anyone suggesting that "additional information (not information
> about table structure, or a summary of the table's conclusions, but
> brand new info that is not in the table at all" should be put into
> the contents of summary attribute ?

I believe that Joshue specifically suggested this, since he said such  
information could be put in the summary attribute. But it may be that  
I did not communicate the definition of Category C clearly to him, or  
that I misunderstood his response.

It also seems that below you're also arguing that putting such  
information in summary is OK:

>  My reading of Joshue's message
> is that he most certainly is not : he is proposing that a /summary/
> of the table be put into the contents of the summary attribute,
> which seems eminently reasonable to me.  Even if an author /were/ to
> put "additional information <etc>" into the contents of the summary
> attribute, would you then want to classify the document as non-
> conforming ?

I think this would be a question of good practice, not conformance,  
because I am dubious about conformance criteria that are subjective  
and not machine-checkable. However, the kinds of information that  
people put in there or might put in there, should inform our design of  
HTML5.

>  I hope not, in exactly the same way that I hope you
> would not want to class as non-conforming a document that put
> "additional information <etc>" into the contents of an ALT attribute
> that could not be found in the visual realisation of the image itself.

I think that would also be a poor practice; such additional  
information should go in title, a figure caption, or the content  
around the image. I believe putting information that is additional  
rather than equivalent in alt would be nonconforming per the current  
draft regarding alt text.

Regards,
Maciej
Received on Thursday, 26 February 2009 19:25:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:32 GMT