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

Re: Bug 8404 -- taking it to the lists

From: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
Date: Wed, 2 Dec 2009 19:59:31 +0100
To: Jeroen van der Gun <noreplytopreventspam@blijbol.nl>
Cc: public-html@w3.org
Message-ID: <20091202195931531361.264f64d5@xn--mlform-iua.no>
There were many references to my previous contribution here, so I 
comment:

Jeroen van der Gun, Wed, 2 Dec 2009 16:33:42 +0100:
> Whether something is being labeled as a figure in literature, does not
> matter. The figure element is for things that have the structure of a
> figure.

"[T]hings that have the structure of a figure"  is not the definition 
given in the HTML 5 draft. 

The draft says that one of the criteria for a calling something a 
figure is that it  "could, without affecting the flow of the document, 
be moved away from that primary content". So if you refer to the 
"figure" with a colon, then it isn't a figure anymore:

<p>See this figure:</p>
<figure>...</figure>

The above use of <figure> is incorrect, as it is impossible to move 
that "figure" away from "the flow of the document", without affecting 
"the flow of the document".

Thats the trick: In order that it can be possible to move the figure 
from the flow of the content, then both "the flow" and "the figure" 
must be prepared. All figures, including tables that are presented as 
figures, have a unique, human readable/visible reference, so that it 
one, in the text, can refer to them from "the flow of the document" 
with phrases such as "See Figure number X.Y.Z".

> The element could also be named the figurestructure element,
> but that would be long and annoying.

Personally, I feel that <object> would be perfect.

> Making the caption required seems fair enough to me. Agreed.

+1

> Linking to a figure is easy, just use the id attribute. You don't need
> an a element to define an anchor. 

The issue I took up when I mentioned 

	<caption><a id="figure_1-2-3">Figure 1.2.3</a> This figure shows ... 
</caption> 

was the need for an element that identifies the readable reference of 
the "figure" - the "Figure 1.2.3" text. As I can't see how a an 
academic "figure" (even when the "figure" in reality is a table) can be 
a figure unless it has such a reference. (Because, you know, the 
caption itself, may contain more than the that very reference.) If we 
had such an element, then that element could also be used inside the 
table <caption>.

It might be that <a id="Figure-id"> is not the best element - perhaps 
something new is needed. But it was the best of the existing elements I 
could find, as an anchor with an @id is not valid, unless the @id 
actually contains an idref.

The point for me is: Unless there is something inside the <figure> that 
demands that a unique, textual reference is placed there, then <figure> 
doesn't contain anything that reasonably well ensures that it actually 
can be called a figure. And: A "figure" relies more on the presence of 
such a reference, than on a caption as such. 
-- 
leif halvard silli
Received on Wednesday, 2 December 2009 19:00:12 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:15:54 UTC