[Techniques] Draft General Technique for GL 3.1 L1 SC1

Guideline 3.1 L1 SC1 requires:
<current>
The natural language of the 
document as a whole can be identified by automated tools[i] 
</current>
 
The following wording is proposed as a draft for a General Technique to
satisfy this requirement. Please review and suggest any
changes/corrections that should be made.  Hopefully this material will
be included in the next internal working draft and will move from there
(after appropriate changes) to the next public working draft.
 
<proposed>
Short-name for this technique: 
Natural language of the delivery unit
 
Task:
Markup or other standard methods are used to identify the natural
language of the 
delivery unit as a whole.
 
Description:
The natural language of a delivery unit may be identified in the
following ways:
 
For HTML documents
Use the lang attribute within the html element.
 Set the value of the lang attribute to the appropriate language code
from the list of 
ISO 639 standard Codes for the Representation of Language Names. The
W3C's 
Internationalization Working Group advises use of two-letter language
codes if 
there is a choice between two- and three-letter codes. 
 
For XHTML documents served as text/html
Use both the lang attribute and the xml:lang attribute within the html
element.
Set the value of both the lang and xml"lang attributes to the
appropriate language 
code  from The ISO 639 standard  list of Codes for the Representation of

Language Names. The W3C's Internationalization Working Group advises use
of 
two-letter language codes if there is a choice between two- and
three-letter 
codes.
The lang and xml:lang attributes should have the same value. 
 
For XHTML 1.1 and other XML-based documents served as text/xml
Use the xml:lang attribute.
Set the value of the xml:lang attribute to the appropriate language code
from the 
ISO 639 standard list of Codes for the Representation of Language Names.
The 
W3C's Internationalization Working Group advises use of two-letter
language 
codes if there is a choice between two- and three-letter codes.
 
It is also possible to identify the language of a delivery unit in the
document's http 
header. However, the W3C's Internationalization Working Group
advisesthat this 
approach must be used with care because individual user settings may
have 
unpredictable results.

When the language of the delivery unit is identified in markup, both
assistive 
technologies and conventional user agents can render text more
accurately. 
Screen readers can load the appropriate pronunciation rules, and visual
browsers 
can display characters and glyphs in appropriate ways. As a result,
users with 
disabilities will be better able to understand and perceive the content.
 
Some languages are spoken in several countries, and there are subtle but

important differences in the way the language is spoken in each country.
In such 
cases, it may be necessary to identify the regional variant of the
language.  For 
example, Canadian French is not the same as the French spoken in
Seneghal. 
Brazilian Portuguese is different from the language as it is spoken in
Portugal, and 
so on. Ideographs used in some Asian languages differ slightly from
language to 
language. Users with low vision and users with learning disabilities or
cognitive 
impairments might be confused by text that does not conform to their 
expectations.
 
Some documents contain text in more than one language. The language of
each 
foreign-language passage should also be identified.  See Guideline 3.1
Level 2 
Success Criterion 3.
 
Resources
HTML techniques
 <http://www.w3.org/TR/WCAG20-HTML-TECHS/#lang-att_primary>
http://www.w3.org/TR/WCAG20-HTML-TECHS/#lang-att_primary"> Identifying
the 
 
primary language
 
CSS-techniques
 
Other-resources
 <http://www.w3.org/International/tutorials/tutorial-lang/#declaring>
http://www.w3.org/International/tutorials/tutorial-lang/#declaring W3c
I18N Tutorial: 
 
Why and how to declare language
 <http://www.w3.org/International/tutorials/tutorial-lang/#specifying>
http://www.w3.org/International/tutorials/tutorial-lang/#specifying
Specifying 
 
language attribute values
 <http://www.w3.org/International/tutorials/tutorial-lang/#negotiating>
http://www.w3.org/International/tutorials/tutorial-lang/#negotiating
Negotiating 
 
language with the server
 
 <http://www.loc.gov/standards/iso639-2/langcodes.html>
http://www.loc.gov/standards/iso639-2/langcodes.html ISO 639 Codes for
the 
 
Representation of Language Names
</proposed>

"Good design is accessible design."

Dr. John M. Slatin, Director 
Accessibility Institute
University of Texas at Austin 
FAC 248C 
1 University Station G9600 
Austin, TX 78712 
ph 512-495-4288, fax 512-495-4524 
email jslatin@mail.utexas.edu 
Web  <http://www.ital.utexas.edu/>
http://www.utexas.edu/research/accessibility