W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2005

Re: Re: Accessibility of "CHM" format resources

From: Orion Adrian <orion.adrian@gmail.com>
Date: Tue, 7 Jun 2005 08:55:59 -0400
Message-ID: <abd6c80105060705554c4442a0@mail.gmail.com>
To: w3c-wai-ig@w3.org

On 6/7/05, John Colby <John.Colby@uce.ac.uk> wrote:
> Many thanks for that information. It seems that Brainbench will not give us 
> what we require. I was looking for gradute and postgraduate level material 
> for designers  rather than electronic testing, which I do with my first year 
> undergraduates.  What is offered is too much down the scale of the Bloom's 
> Taxonomy - it doesn't really test the synthesis, judgement and critical 
> appraisal that I'm looking for, merely some aspects of knowledge.
> And certainly the 'Master' level is far removed from the Masters (MSc) level 
> courses I'm involved with and teach here in the UK, so the nomenclature 
> would not be acceptable in an academic environment.

Of course that wouldn't. I would imagine it more deals with theory and
general design practices.

> I see that others have already responded to the main CSS observation. Had 
> you hoped to open substantive debate then incorrect statements will 
> naturally get hammered down.

I don't mind being corrected. What I do mind is saying, even taking
into perspective the commenters, that calling X, Y it someone
eliminates all my experience. I wasn't by the way just having a bad
day, but people do make typos, mistakes and so on. The rapidity that
people came down on me to challege my knowledge was disturbing.

> I will make but one comment: In even attempting to follow pixel perfection
> you seem to be oncentrating on the display of pages on a PC with not much
> consideration of other devices such as phones or PDAs, and then ignoring 
> the alternative devices such as text and speech readers. Use of CSS in the 
> correct way and targetted at the correct device will style the content 
> effectively for each device. This is an undoubted benefit for those who have 
> to use alternative devices (and I am not here limiting Accessiblility to those 
> who are disabled in some way) and expands the potential user base of any 
> page. Any hacks are, in my experience, mainly to get round the failings of a 
> single browser that is manufactured by a company who subscribes to the 
> Standards body you criticise.

You seem to think I'm on the wrong team. I'm not. I don't advocate
pixel perfection. In fact the opposite. The problem is I like many
designers have trouble getting some pages to work at all without pixel
manipulations. Fixed width and pixel sizes are often applied because
the calcuations needed to make sure everything is at least where it
should be relative to each other can be done before hand since there
is no method of doing it in the CSS itself.

I am not looking for pixel perfection and would be happy to see the
entire system go. And even in the case of images I'm much more of a
fan of inches/cm than I am of pixels.

What I'm looking for is a mechanism that allows me to easily achieve
the layout I want (where things appear relative to each other on the
page) without having to use tools that weren't meant for it (e.g.
float) and having the solution read in such a way that it's
immediately obvious.

It takes awhile to code up a page and far too long to analyze one. CSS
doesn't read well. Now people who have been working with it a long
time I think will come up with a different answer. That, of course, it
reads well.

I fear for the learning curve of some of the specs that have come out
of the W3C recently. XML namespaces, CSS 3, XLink and especially RDF.
I took one look at an example RDF page and knew I would never use it.
It took reading page after page of documentation on it to figure out
even close to what it was doing. It isn't immediately obvious which it
needs to be. I think one key problem it has is that it was force-fed
into an XML formate when a custom data format would have been better.
For instance:

<item> <relationship> <item>[, <item>]*
<item> <relationship> <item>[, <item>]*

and so on.

Just a simple list of items related (and how related) to other items.
XML isn't always the clearest method of doing things even if it can
work. I laughed at XQueryX. I thought, this XML thing is going way too
far. XQuery is the much better syntax. RelaxNG's compact syntax has
mucho benefits over its XML format. XML gave us one major benefit, it
gave us a single format to standardize on that added great power in
our toolbelts. However while the idea is a good one, I don't think
there should be only one. XTree, XList, XContent might all be good
substandards. Who knows.

Orion Adrian
Received on Tuesday, 7 June 2005 13:23:00 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:21 GMT