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

Re: what is dt? focusing on Figure

From: Shelley Powers <shelleyp@burningbird.net>
Date: Thu, 17 Sep 2009 08:12:55 -0500
Message-ID: <4AB235D7.8020008@burningbird.net>
To: Maciej Stachowiak <mjs@apple.com>
CC: Jeremy Keith <jeremy@adactio.com>, Smylers@stripey.com, public-html@w3.org
Maciej Stachowiak wrote:
> On Sep 17, 2009, at 1:42 AM, Jeremy Keith wrote:
>> Shelley wrote:
>>> Yes, I think that, among our other concerns, we can also include a 
>>> concern about the mnemonics of dt/dd in Figure.
>> I agree that it feels counterintuitive to use <dt>/<dd> inside 
>> <figure> but, IMHO, it does make sense inside <details> simply 
>> because the word "details" begins with a letter D  just as dt/dd 
>> made sense inside <dialog> when that element existed.
> Indeed, it seems a little more natural because of the "d". But that 
> generalization of <dt>/<dd> semantics makes it a little more natural 
> for <figure> than it would be otherwise.
>> So if we're going to discuss either:
>> a) finding a better element to recycle for captioning <figure>s or
>> b) creating a new element for captioning <figure>s,
> Do you have any ideas for other elements to recycle? I posted the 
> problems with most of the existing ones that sound like a good fit 
> based on existing semantics. It's possible to use elements with weak 
> or null semantics, like <div>, <span> or <b>, but that doesn't seem 
> like good design taste. <header>  could work on a technical level, but 
> the name doesn't seem like a great fit, and I think it could be more 
> confusing than applying the <dt>/<dd> pattern.

I believe that Legend has been found to be an acceptable option, if we 
must recycle existing elements. Personally, I think if we're defining a 
new outer element, then it makes sense, as has happened with table and 
lists, that we also provide a fresh new element for the use within the 
element, ensuring a higher level of success when using the new outer 

This version of HTML is supposedly focused on semantics in markup. If 
so, we can most likely trim the spec because of size in other areas. We 
most definitely should not skimp in the area that supposedly sets this 
spec off from previous versions of HTML.

>> I think it would be good to keep <details> out of the discussion.
> Sounds ok to me, though I'm not sure Shelley would agree. She seems 
> concerned about any alternate use for <dt>.
Actually I have stated in the past that I plan on beginning an argument 
about dropping Details.

We can remove Details, when it comes to considering what we can use to 
replace dt/dd. We can't, though, if we're looking at the overall use of 
dt/dd, not Figure.

We need to keep our topics clearly defined, and this topic thread was 
focused on dt/dd and their redefinition.

Now, since we seem to be in agreement that dt/dd is not appropriate for 
Figure, we can focus on finding a replacement for dt/dd in Figure for 
the rest of this thread. I would have no problem, as long as my concerns 
about redefining dt/dd are not lost somewhere along the way.

Received on Thursday, 17 September 2009 13:13:40 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:15:51 UTC