W3C home > Mailing lists > Public > public-html@w3.org > July 2007

3.4. Global attributes

From: Ben Boyle <benjamins.boyle@gmail.com>
Date: Mon, 30 Jul 2007 22:31:06 +1000
Message-ID: <5f37426b0707300531k1a028b9fp5bdb55c5c2ca1a7c@mail.gmail.com>
To: "HTML WG" <public-html@w3.org>

3.4.1. id attribute
http://dev.w3.org/html5/spec/Overview.html#id
This section says the id must contain at least one character, but
doesn't specify any limitations on that character. I thought it was
invalid to have things like id="1" and id="$"
I'm used to following the HTML 4.01 rules:
http://www.w3.org/TR/html401/types.html#type-id - is HTML 5 different?
What are valid id values?

3.4.2. title attributes
http://dev.w3.org/html5/spec/Overview.html#title
"on an image ... it could be a description of the image" I believe
this has already been discussed on this list, but there is a
discrepancy between the use of img@alt and img@title. However that is
resolved, it may need to be reflected in the document here. I would
consider removing "it could be a description of the image" to avoid
needless contention.

3.4.3. The lang (HTML only) and xml:lang (XML only) attributes
http://dev.w3.org/html5/spec/Overview.html#lang
Disappointed that this explicitly requires different attributes for
each serialisation (but can't offer a solution).

3.4.5. class attribute
http://dev.w3.org/html5/spec/Overview.html#class
I tracked back through "unordered set of space-separated tokens" to
discover that class names must be "words" but couldn't find the
definition of what was a valid "word". Similar to 3.4.1. (id) I think
there are possibly constraints around tokens? (This information may
very well be in the spec, there just might need to be a link from
these sections so it can be easily found.)

3.4.6. irrelevant attribute

Got a few tangential comments after reading this.

Boolean attributes: in XHTML authors would use irrelevant="irrelevant"
and in the DOM irrelevant = true; (I think some may mistakenly use
irrelevant = 'irrelevant'; but that probably works anyway).

I think the second example (replace a still frame with a video once
loading is confirmed) is so useful it would be nice to have an easier
way to specify it with markup. Could we do something with multiple
source elements perhaps?
<video>
<source src="frame.png">
<source src="video.ogg">
<img src="frame.png">
</video>

In both examples, I don't understand the benefit of @irrelevant when
one could just use the replaceChild() method from the DOM. What's the
benefit of @irrelevant over DOM manipulation? Is there any way to
manipulate the irrelevant attribute other than by scripting (that
could potentially be a benefit).
Received on Monday, 30 July 2007 12:31:28 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:47 UTC