Re: Edit: Re: [WSG] [Please don't flame :)] HTML, XML what's the difference.

2006-02-10 Thread Patrick H. Lauke

Stephen Stagg wrote:

And how, pray tell, would a screen reader know - based on a series of 
presentational rules - what the meaning of a made-up tag soup is?


The same way that they would with normal HTML, by reading the XML, and 
the stylesheet and guessing,  if an element has the font-weight:bold

element, then it should be emboldened.


Wrong. Screen readers do not look at the CSS and try to guesstimate what 
is a heading, what's a paragraph, what's a list, etc.



Screen-Reader hints are still presentational devices.


Screen readers look at the structure of the document, which is clearly 
defined as it's standardised in the HTML specification.



I believe (tho haven't
checked) that there are a whole load of CSS properties to do with 
controlling assistive-technologies output.


There are aural stylesheets, which only give hints about how to present 
something aurally. They do not define purpose or role of the elements 
they refer to, and THAT is what counts.


--
Patrick H. Lauke
__
re·dux (adj.): brought back; returned. used postpositively
[latin : re-, re- + dux, leader; see duke.]
www.splintered.co.uk | www.photographia.co.uk
http://redux.deviantart.com
__
Web Standards Project (WaSP) Accessibility Task Force
http://webstandards.org/
__
**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: Edit: Re: [WSG] [Please don't flame :)] HTML, XML what's the difference.

2006-02-10 Thread Stephen Stagg


On 10 Feb 2006, at 19:14, Patrick H. Lauke wrote:


Stephen Stagg wrote:

And how, pray tell, would a screen reader know - based on a  
series of presentational rules - what the meaning of a made-up  
tag soup is?
The same way that they would with normal HTML, by reading the XML,  
and the stylesheet and guessing,  if an element has the font- 
weight:bold

element, then it should be emboldened.


Wrong. Screen readers do not look at the CSS and try to guesstimate  
what is a heading, what's a paragraph, what's a list, etc.


Not wrong actually, Good screen-readers DO read the CSS to work out  
various things, incuding to see if someting has a display:hidden.  I  
do acknowledge that this is an area that would have to be developed  
in screen-readers but that does not invalidate the idea.



Screen-Reader hints are still presentational devices.


Screen readers look at the structure of the document, which is  
clearly defined as it's standardised in the HTML specification.


And they PRESENT it to someone with visual impairment, The  
presentational properties should be set in the presentational layer



I believe (tho haven't
checked) that there are a whole load of CSS properties to do with  
controlling assistive-technologies output.


There are aural stylesheets, which only give hints about how to  
present something aurally. They do not define purpose or role of  
the elements they refer to, and THAT is what counts.


As is said, I wasn't sure about the exact nature of the aural  
stylesheets.  Thanks for the info, Perhaps this is something that  
could be developed to improve the designers' control over output to  
screen-readers? no?



--
Patrick H. Lauke
__
re·dux (adj.): brought back; returned. used postpositively
[latin : re-, re- + dux, leader; see duke.]
www.splintered.co.uk | www.photographia.co.uk
http://redux.deviantart.com
__
Web Standards Project (WaSP) Accessibility Task Force
http://webstandards.org/
__
**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: Edit: Re: [WSG] [Please don't flame :)] HTML, XML what's the difference.

2006-02-10 Thread Patrick H. Lauke

Stephen Stagg wrote:

Screen readers look at the structure of the document, which is clearly 
defined as it's standardised in the HTML specification.


And they PRESENT it to someone with visual impairment, The 
presentational properties should be set in the presentational layer


So by your logic we could even have stuck with using font 
size=+3This is a heading/font as screen readers could theoretically 
just have picked that up and magically deduced it's a heading...


--
Patrick H. Lauke
__
re·dux (adj.): brought back; returned. used postpositively
[latin : re-, re- + dux, leader; see duke.]
www.splintered.co.uk | www.photographia.co.uk
http://redux.deviantart.com
__
Web Standards Project (WaSP) Accessibility Task Force
http://webstandards.org/
__
**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: Edit: Re: [WSG] [Please don't flame :)] HTML, XML what's the difference.

2006-02-10 Thread Lachlan Hunt

Patrick H. Lauke wrote:
generic XML + CSS would be meaningless without some third technology 
that defines semantics (a DTD, XBL, etc)


Neither a DTD nor XBL define document semantics at all.  A DTD only 
defines the document syntax and structure.  XBL is only a binding 
language for attaching behaviour to an element, it doesn't define 
semantics either.


The closest thing there is for describing semantics is the XML 
namespace, but even then it only loosely associates the elements with 
the semantics defined in the relevant specification (if one exists).


See this post for an interesting discussion of why DTDs don't define 
semantics.

http://groups.google.fi/group/comp.text.sgml/msg/c3e53dee2c152a81?output=gplain

--
Lachlan Hunt
http://lachy.id.au/

**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: Edit: Re: [WSG] [Please don't flame :)] HTML, XML what's the difference.

2006-02-09 Thread Patrick H. Lauke

Stephen Stagg wrote:

I understand that this is already possible in most modern browsers but 
it will never be used or properly implemented unless HTML is dropped as 
a language.  Worried about screen-readers? I don't see why, the 
screen-readers would have to parse the CSS to find clues about how to 
read the content, but then modern ones already do.  :)


And how, pray tell, would a screen reader know - based on a series of 
presentational rules - what the meaning of a made-up tag soup is?


--
Patrick H. Lauke
__
re·dux (adj.): brought back; returned. used postpositively
[latin : re-, re- + dux, leader; see duke.]
www.splintered.co.uk | www.photographia.co.uk
http://redux.deviantart.com
__
Web Standards Project (WaSP) Accessibility Task Force
http://webstandards.org/
__
**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Edit: Re: [WSG] [Please don't flame :)] HTML, XML what's the difference.

2006-02-08 Thread Stephen Stagg
Sorry, it's late in England. I'm gonna go to bed now :)How could you know what style to apply to meaningless content?That's what the style-sheet is for.  We are relying more and more on the display: element of CSS, why not define a well-thought out and extensible set of display types to replace the default behavior of many current tags. Want to include flash on your site? define a CSS rule:flashmovie{ display:flash;} and then your document reads:flashmovie src=""file://a.c.v/me.swf">file://a.c.v/me.swf" /Hell, even I know what that means :))Effective styling depends on document semanticsWrong, I see the point you are trying to make, but Styling is totally autonomous, It takes pre-defined rules and applies them to a list of tags, the CSS processor in modern browsers shouldn't care WHAT the semantic content of its tags is. div class="h"Foo Bar/div.h { font-size: large; font-weight: bold; }Would you agree that that is a bad idea?No (except the h doesn't provide any clue to the content) , but it seems silly to use a DIV element, which REDUCES semantics, having no meaning to anyone.  Rather use, similar to that which you suggest:mydocument	paragraph		headingThis Heading Belongs to this Para/heading		contentblah, blah, /content	/paragraph/mydocumentThis is not meaningless, It is more readable than HTML, to a human.  It may not have semantic meaning, but who needs semantic meaning.A major factor in the development of microformats is that they reuse existing document semantics, where possible.  They aren't just about making up new class names and relationship values. No, they re-use existing Standard formats, where possible, not Semantics.  'Semantics' means 'meaning in the context of a language'.  Take the hCard format, a sample from the specification reads:span class="tel" span class="type"home/span: span class="value"+1.415.555.1212/span/spanHow in any way does a span element have semantic meaning? The micro-format adds semantic meaning to the span elements in the example.  Why not remove it. A sample from my imaginary XML hCard format reads:tel	typehome/type	value+1.415.555.1212/value/telNow THAT also to has real semantic meaning in the context of my (imaginary) proposed hCard format, and is easy to read for a human. Oh and it's lighter on bandwidth also.  "Micro-Namespaces" is a term you just made up, it means nothing.I DID make it up but NO it is not meaningless, If you take the two parts separately, micro means small(ancient greek, µikros = small), namespace is a defined XML feature.  My point is that When we get to the stage of using pure XML, the namespace and the format ideas could merge to allow a hCard namespace to be defined, if the hCard is a micro-format, then the xmlns hCard(or whatever) could also have a micro- stuck before it.  :)I understand that this is already possible in most modern browsers but it will never be used or properly implemented unless HTML is dropped as a language.  Worried about screen-readers? I don't see why, the screen-readers would have to parse the CSS to find clues about how to read the content, but then modern ones already do.  :)Stephen.