[whatwg] Figure Captions

[Aryeh Gregor]

On Sun, Sep 27, 2009 at 12:13 PM, tjeddo <tjeddo at gmail.com> wrote:
> Would anyone mind summarizing the design rationale for choosing the 'dt'
> element as opposed to a 'caption' element when specifying the caption of a
> figure?
>
> Looking through the current draft of the HTML5 spec it appears that the
> surrounding context of 'dt' element determines how the contents of 'dt' are
> interpreted (e.g., dl, figure, details).
> However, from a readability, and I guess aesthetic perspective, a <caption>
> tag seems like the correct fit here.  A <caption> tag is more explicit in
> terms of the intended semantics associated with a figure; whereas, <dt>
> (defined term?) is more abstract (IMO).

The goals were to 1) use an existing element that 2) existing browsers
wouldn't mangle too horribly when they found it in an unexpected
place.  <legend> and <caption> are both effectively unusable outside
their expected places in several major browsers right now -- they get
eaten, reparented, or any number of other horrible things.