RE: Footnotes in HTML5.1

Michael[tm] Smith wrote:
>
> David MacDonald <david100@sympatico.ca>, 2014-08-07 15:56 -0400:
> ...
> > I think it would be great to have a working <noteref> and <note>
> > elements that opens (and/or jumps) to the footnote and returns the
> > user back to footnote.
>
> You might want to consider taking this topic to the www-style list
> instead.

Personally, I think this is a horrible recommendation.

We're not worried about how it looks Mike, we're concerned about the
semantic representation of the data being marked up (and as an aside Mike,
can you point us to the previous discussions at the W3C?)

Footnotes(1) and asides(2) have distinct and unique definitions, and
further, they generally have different interaction (reading) patterns(3),
which by rights should remain separated.

You are correct, <aside>, as defined by the HTML5 spec, can be "...used for
typographical effects like pull quotes or sidebars, (and/or) for
advertising..", but that is not what a footnote is, nor does: the footnote
has a very specific role to play, and is often a critical requirement for
scholarly works, legal documents, and other "non-app" applications.

One key function, not provided with the <aside> element is a programmatic
binding of the content to a specific reference (although, we *might* also
consider creating rel="footnote" -
http://www.w3.org/TR/html5/links.html#linkTypes - but that is not the
charter of www-style, but rather the charter of HTML5-WG) (4)

Relegating it to simply a visual "styling" completely misses the whole
point.


>
> The reason I'd suggest www-style is that the idea of having specific
> markup for footnotes is not a new suggestion -- we've had a lot of
> discussion over the years about markup for footnotes, and that
> discussion was a large part of what led to the <aside> element being
> added to HTML.
>
> So the fact that at this point we don't have a more specific element
> than <aside> for footnotes is actually not an oversight -- it's by
> design.

<aside>
"The aside element represents a section of a page that consists of content
that is tangentially related to the content around the aside element, and
which could be considered separate from that content. Such sections are
often represented as sidebars in printed typography.

The element can be used for typographical effects like pull quotes or
sidebars, for advertising, for groups of nav elements, and for other content
that is considered separate from the main content of the page."
source: http://www.w3.org/TR/html5/sections.html#the-aside-element

(JF notes that footnotes are not "tangentially" related to the content in
proximity to the <aside>, but rather are specifically related to a specific
piece of content.)
</aside>

>
> That is, the <aside> element is, by design, a presentation-agnostic
> means intended for marking up all types of notes as a general class. As
> far as footnotes specifically, the prevailing agreement that emerged
> from past discussions was that footnotes are mostly just a particular
> presentational way to handle (render) notes as a class (<aside>s) in
> particular presentational environments -- and so therefore the right
> solution for footnotes is for CSS to provide some means to handle them.

That might handle the visual conundrums, but it does not handle use-cases
where styling is moot (i.e. screen reader usage). Once again, a pointer to
this "prevailing agreement" would be most useful here, as, being a longtime
subscriber to the HTML5-WG mailing list, I don't recall that discussion
being documented.


>
> That's why the CSS Generated Content for Paged Media (GCPM) spec has an
> extensive overview of footnote handling, and a solution:
>
>   aside.fn { float: footnote }
>   http://www.w3.org/TR/css-gcpm-3/#creating-footnotes


Again, that deals with the visual output, but does not address the specific
semantic difference between and aside and a footnote. In the past, when
suggestions of the idea of starting to add semantic reasoning to CSS have
been suggested, the howl and cry on how "stupid" that idea might be has been
heard (loud and clear). Thus, using CSS to address the semantic requirement
MUST not be considered here either.


> Anyway, that's all why I'd suggest www-style as the best place to
> restart any new discussion about handling of footnotes -- e.g., to ask
> why browsers haven't already implemented "float: footnote" yet and/or
> to find out if there's some proposed CSS alternative to "float:
> footnote" that's been discussed and that browser implementors might be
> more likely to implement at this point.

I do not think there is a clear understanding of the use-case problem. It's
not about how it looks, it is about how it is perceived - and so taking this
to www-style does not solve the problem at hand, that group simply would
figure out how to make it pretty :)

In fairness, another approach *might* be to use ARIA (role="footnote" +
aria-describedby: feels a bit awkward), which would address the "screen
reader problem", but I think a more universal, native approach would benefit
*more* than just the non-sighted user, which is something that many
non-accessibility-specialists have often argued for in the past (for
example, as an argument *against* @longdesc...)

JF

(1) Footnote: "a note of reference, explanation, or comment usually placed
below the text on a printed page" - source:
http://www.merriam-webster.com/dictionary/footnote
See also: http://libguides.csuchico.edu/content.php?pid=214757&sid=1793692

(2) Aside: "to or toward the side - used with put or set to describe
something that is being kept or saved for a future use; - used to say that
something is not included in a statement that follows" - source:
http://www.merriam-webster.com/dictionary/aside

(3) Footnotes generally are expected to have an interaction pattern that
allows the end user to reference the content, and return to the reference,
with relative ease. This XML definition of footnote is a good example:
http://docs.ubmatrix.com/webhelp/XPE/3_5/ubmf_functions/ubmfi_footnote-links
.htm

(4) Very quickly knocking together an example:

	<p>Gombrich associates Daumier with the political cartoonists
	and not "the French tradition of great art."
	(<a href="#gombrich" rel="footnote">7</a>)</p>

	...blah blah..

	<aside id="gombrich"> 7. E.H. Gomrich, art and Illusion (Princeton:
	Princeton University Press, 1961), 336, 345, 252</aside>

(Sample text taken from
http://libguides.csuchico.edu/content.php?pid=214757&sid=1793692)

{Missing here however is a specific mechanism that would allow the user to
return to the initial reference link, although, as a basic anchor element, I
suppose the "Back" button would do so...}

Received on Sunday, 10 August 2014 18:00:07 UTC