W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > April to June 2002

RE: controlled use of language

From: lisa <lisa@jctech.co.il>
Date: Mon, 20 May 2002 16:54:30 -0400
Message-Id: <5.1.0.14.2.20020520165418.04b84750@localhost>
To: w3c-wai-gl@w3.org

Let me clarify,

This is not about removing ambiguity by controlling the words and style
of the author.
This is about creating a technique of meta data and markup, that
clarifies their choice of word usage.

RDF metadata and xml.

add on some tools and most folks could use it

all the best,

lisa


---Original Message-----
From: Al Gilman [mailto:asgilman@iamdigex.net]
Sent: Thursday, May 16, 2002 10:14 AM
To: Sean B. Palmer; lisa Seeman; William Loughborough
Cc: w3c-wai-gl@w3.org
Subject: Re: controlled use of language


At 12:01 PM 2002-05-16, Sean B. Palmer wrote:
 >[Snipping WAI PF since, IMO, language is neither a protocol nor a
format
 >:-)]

What we are talking about is the _protocol_ which governs the coupled
use of controlled and uncontrolled vocabularies.  And multiple
definitions of vocabularies with different control protocols.  Some are
controlled through consensus, publication and registration [Dublin
Core], some through feedback methods, responding to demographics of
vocabulary control in populations at cognitive risk [CD word lists].
The controlled vocabularies are an extreme form of "lowest level
language" in the sense in which TimBL uses that notion.  Hence a form of
format.  Ergo: it's all about protocols and formats.

Al

Pinpoint example:

I tried to find a basis for "spellcheck your content" in WCAG 2.0.

The "check you work" guideline only talks about orthodox use of markup,
not of text.  Wrong.  There are unmarked encodings in the text that need
to be valid, too.  See the "use real words in standard spellings" thread
dealing with SGRS [search on "orthographic" in www-voice list].  This
has to be in our graph of clauses, too.

And this is a technique which follows from all of "use clear and and
simple expression," "compatibility with AT" and "use technology with
hygiene."  Multiple heredity -- you can't file it safely under either.

Al


 >From: "William Loughborough" <love26@gorge.net>
 >
 >> So to "Be prepared for it to be a long process." is a
 >> matter of *forever* because it ain't gonna happen.
 >
 >Whilst I agree with the "who's going to police the police? The
coastguard?"
 >sentiment, I think that Lisa is not (just?) hinting towards a
controlled
 >language (which, like Orwell's Newspeak, is absurd), but rather/also
 >towards regularized interfaces for creating Web content.
 >
 >Sidenote: whilst this is at a tangent to any useful discussion, there
have
 >of course been numerous studies conducted on simplifying languages. A
good
 >starting point is the AECMA's "Simplified English":-
 >
 >http://www.userlab.com/Downloads/SE.pdf (PDF only; sorry)
 >also: http://www.aecma.org/Publications/SEnglish/sengbrc.htm
 >
 >which is all well and good, but is aimed towards people who have a
 >relatively high understanding of English in the first place, and can
spend
 >the time and effort working on a simplified vocabulary; i.e. applies to
a
 >minute fraction of the population.
 >
 >But onto more promising avenues of discussion. I like the DAML
 >(http://www.daml.org/tools/) approach of having content forms, where
people
 >can create homepage metadata by searching and/or selecting (Prof. Jim
 >Hendler has talked about this). It was also one of the aims of UWIMP,
which
 >is the service that William and I planned many months ago.
 >
 >If we are to create William's much prophecized ambient information
space,
 >it must be possible for the people with the information to publish what
 >they know with minimal costs and maximum benefits to them. This is only
 >going to be possible through small incremental changes, although if
some
 >organization like Google or Alexa gave over their already vast data
stores
 >to the Web community, that would help a lot (why pay for a server when
 >Google and Alexa index it for free anyway?).
 >
 >It's an ongoing battle. Metadata formats such as RDF are there to help
too,
 >and a standardized deployment of RDF-in/from-HTML is long overdue
 >(hopefully not another manifestation of the six month phenomenon). RDF
is
 >good because the Semantic part of the Semantic Web is there to satisfy
the
 >(much ballyhooed) "semantic pragmatic" nature of machines.
 >
 >But marking up text itself to be "unambiguous" is rather HumanMLish in
 >character... the waste of time criticisms are justifiable for
corollaries
 >of "the language moves faster than the editor". Not only that, but
language
 >is art - if you move towards unambiguity, you may lose the essence of
many
 >of the great works. One's mind may drift back to Jefferson's famous "we
 >hold these truths to be sacred & undeniable", and wonder what he
thought of
 >Adams' and Franklin's (if their) modifications.
 >
 >--
 >Kindest Regards,
 >Sean B. Palmer
 >@prefix : <http://purl.org/net/swn#> .
 >:Sean :homepage <http://purl.org/net/sbp/> .

---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.361 / Virus Database: 199 - Release Date: 5/7/2002


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.361 / Virus Database: 199 - Release Date: 5/7/2002
   
Received on Monday, 20 May 2002 16:49:34 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:19 GMT