Fw: [XHTML2] 4th WD comments

----- Original Message -----
From: "Christoph Päper" <christoph.paeper@tu-clausthal.de>
To: <www-html@w3.org>
Sent: Monday, February 03, 2003 7:48 AM
Subject: [XHTML2] 4th WD comments


>
> Just a few comments based on the diff marked version:
>
> 6.4. Edit Collection
> | <p>I will do it next <span class="deleted">week</span>
> | <span class="inserted">month</span>.</p>
>
> Err... you just introduced the 'edit' attribute to replace it by 'class'
in
> the example?
> The defintion is fixed to visual media and CSS.
>
> 6.6. Embedding Attribute Collection
>
> The 'table'/PNG example is a good one, the 'p' a very bad one.
>
> 6.7. Image Map Attribute Collection
>
> I don't see the need for the 'map'/'usemap'/'id' combination any more,
given
> that the element that references the map graphic can also contain the
links:
>
>   <nl src="navbar1.png" type="image/png" usemap="#">
>     <label>Navigate the site:</label>
>     <li href="guide.html" shape="rect" coords="0,0,118,28">Access
Guide</li>
>   ...
>
> 7. XHTML Structure Module
>
> Is the minimal content model for 'head' really just "title"?
>
> 8.9. The heading elements
> | Numbered sections and references
> | XHTML does not itself cause section numbers to be generated from
> | headings. Style sheet languages such as CSS however allow authors
> | to control the generation of section numbers.
>
> Please provide at least in a sample stylesheet, if not in an inline
example,
> the code for this.
>
> 8.10. The hr element
>
> If 'hr' stays, it--as an empty element--definitely doesn't belong into the
> *text* module.
>
> 8.11. The kbd element
> | The kbd element indicates text to be entered by the user.
>
> Extend meaning to cover other input devices (<kbd>double-click</kbd>)?
>
> 8.15. The quote element
>
> The 'quote' element is bad, at least without requiring the UA to indicate
it
> /somehow/ as a quote. You can't rely on CSS (or a hypothetical other style
> language) alone and it's hard to use both, manual quotes *and* CSS
generated
> ones. I also think, 5 letters is too long, you could just have redefined
'q'
> if you wanted.
>
> 10. XHTML List Module
> | <dl>
> | <dt><strong>Lower cost</strong></dt>
>
> Drop the <strong/>s from the example. Suggest stylesheets instead.
>
> | <dd>young excitable person who may mature
> |    into a <em>Nerd</em> or <em>Geek</em></dd>
>
> <em/>s should be <dfn/>s, no?
>
> I still don't see a way to continue lists with something like <ol
> continue="#firstpart">. Neither seem the well known transitional
attributes
> 'start' and/or 'value' be preparing to return.
>
> 14.2.1. Referencing object data
> | <object
> |    data="mymovie.mpg"
> |    type="video/mpeg">
> |    A film showing how to open the printer to replace the cartridge.
> | </object>
> ...
> |   This user agent cannot process this movie.
>
> That's giving a *really* bad example to readers. The text should be an
> alternative, not a description. There are even more examples showing this
> discouraged behaviour.
>
> In my comments on the previous WD, I said that *if* 'meta' isn't
necessarily
> an empty element any more, this should also apply to 'link'. I'd like to
> extend this statement ot 'param'. This element has almost the same
function
> for 'object' as 'meta' has for 'html' / 'head' / 'body' and is also just
> 'name' -> 'value'.
>
> 16.1. The style element
> | Example:
> | <head>
> |  <style type="text/css">[...]</style>
> | </head>
>
> This is invalid XHTML2.
>
> 17.2. The col and colgroup elements
> | <style type="text/css">
> | #colgrp1 {width: 300}
> | #col1 {width: 100}
> | #col2 {width: 50}
> | </style>
>
> This is invalid CSS.
>
> It seems as if *all examples* will have to be carefully checked at some
> time, as they often show errors and bad design, and therefor encourage it.
> The main reason for this is probably copy&paste from previous versions of
> HTML specifications.
> I propose using the periodic table of elements instead of a calendar as an
> example for correctly using tables.
>
> I doubt that it's a good choice to order elements alphabetically.
> 'Blockquote' and 'quote' for instance are several screen pages apart.
>
> If you ever seriously come to the question whether to drop 'a', I suppose
> you better drop the even more meaningless 'span', which would save 6 bytes
> each time.
>
> Christoph Päper
>
>

Received on Wednesday, 5 February 2003 07:45:14 UTC