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

Re: Cleaning House

From: Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>
Date: Sat, 05 May 2007 14:28:04 +0100
Message-ID: <463C8664.3060008@googlemail.com>
To: Lachlan Hunt <lachlan.hunt@lachy.id.au>
CC: "John Foliot - WATS.ca" <foliot@wats.ca>, www-html@w3.org, public-html@w3.org

Lachlan Hunt wrote:

> * What's the semantics you're trying to represent?
> 
> e.g. a taxonomic designation, a technical term, an idiomatic phrase from 
> another language, a thought or a ship name are commonly rendered in 
> italics, and there are other use cases for bold.  Pick  at least one and 
> then answer the remaining questions.
> 
> * Whats the use case for the semantics? (Why would authors use it? Real 
> world example content would be good)
> * What problems would a new feature solve?
> * Why are <b> and/or <i> unsuitable for the use case/problem?
> * What benefit is there for users?
> * What benefit is there for authors?
> * What benefit is there for implementers

I'll have a crack at some of these.

1. Taxonomic designations.

Suggested HTML: "We belong to the species <span lang="la">Homo sapiens 
sapiens</span>."

Alternatively substitute <name type="taxonomy"> for <span>.

Problem it solves: Talking UAs should pronounce it without emphasis in 
Latin pronunciation. Use of <i> results in emphasis and the wrong 
pronunciation. JAWS for example already supports the lang attribute.

Authoring tool implementation hints: "Name" then "Taxonomy" dropdown 
options. Taxonomical lookup wizards.

2. Technical term.

Suggested HTML: "Authors <term>should</term> use semantic markup."

Problem it solves: No single typographic convention exists in print, but 
I think should always be italic in English braille. Often not emphasized 
in speech. <i> would typically result in emphasis.

Additional benefits: You could specify a glossary and definitions for a 
document, then use <term> to indicate that UAs should refer to those 
glossaries and definitions when explaining a term, and to hint to users 
that such special definitions exist. Especially useful when applying a 
narrow meaning to ordinary language like MAY, SHOULD, MUST, etc. So 
users would be less confused and authors would have to do less work in 
explaining where to find definitions.

Authoring tool implementation hints: "Special term" dropdown option. 
Auto-markup of terms as you type using a supplied glossary. Tooltips.

3. Idiomatic phrase from another language.

Suggested HTML: "Semantic markup has no <span lang="fr">je ne sais 
quoi</span>."

Problem: Should be italic in Western print. Should not be stressed in 
ordinary speech. Putting these in <i> is hopeless. Talking UAs need to 
know which language in order to use the correct pronunciation. The 
italic formatting should, again, be the result of the lang attribute. I 
think this a good use-case for <span>; <foreign> wouldn't add anything.

Authoring tool implementation hints: "Switch language" command. 
Auto-recognition, spelling correction, and markup of other languages. 
Flag icons (perhaps).

4. Thought.

Suggested HTML: "<thought>I love markup</thought>, thought the web 
developer."

Problem: <i> and <b> are merely emphasized, but like quotations and 
direct speech, it would be much better if talking UAs could read 
thoughts in a different voice. <thought> and <speech> would be useful 
additions to HTML to go along with <q> and <blockquote> or <quote>.

Additional benefits: Would work rather well in HTML chat.

Authoring tool implementation hints: "Thought" dropdown option. Thought 
bubble icon.

5. Ship names

These are a real minefield of typographical uncertainty, but perhaps:

Suggested HTML: "I served on board the <name type="ship">Enterprise</name>."

Problem: Ship names are variously formatted in print and should 
typically italicized in English braille but <i> produces emphasis in 
speech where none is required.

Some precedents/discussions of this troublesome subject:

http://joeclark.org/YESaccessNOsgml.html

http://mh.cla.umn.edu/houdek2.html

http://www.cs.queensu.ca/achallc97/papers/s007.html

http://www.menota.org/guidelines-2/ch9/metrics_2-0-b.page#id2417660

--
Benjamin Hawkes-Lewis
Received on Saturday, 5 May 2007 13:46:58 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:44 UTC