[Accessforall] [Architecture] Codes for languages in ISO 24751 and the registry

Christophe Strobbe strobbe at hdm-stuttgart.de
Fri Oct 5 06:18:52 EDT 2012



Am Do, 4.10.2012, 21:53 schrieb Denis Anson:
> I suspect that some of the subtleties of this discussion are extraneous to
> the issue at hand.

In eng-GB they would say I have a bee in my bonnet about languages ;-)


> While there are some 6500 "living" languages in the
> world, the issue for AT selection is the languages that are supported by
> the AT, which I suspect to be a much smaller number.


The discussion on ISO 639-x, IETF BCP 47 etc is not simply a question of
which standard allows us to identify more languages, but also which one
provides more precision. More recent parts of ISO 639 recognise that,
e.g., ar (Arabic) and zh (Chinese) should be treated as macrolanguages,
and provided new tags for the distinct individual languages encompassed by
the old tags. "Chinese" and "Arabic" cover a huge number of speakers.
Sometimes, the more recent three-letter codes don't provide advantages
over the older two-letter codes, but in other cases they will. For
example, if you want to convert Mainland Mandarin texts, Taiwanese
Mandarin texts or Cantonese texts to their corresponding Braille scripts,
marking the source text zh-CN, zh-TW, or zh-HK, respectively, seems to be
sufficient. But in the area of speech synthesis, the eGuideDog project is
working on Ekho <http://www.eguidedog.net/ekho.php>, a TTS for Cantonese,
Mandarin and Zhaoan Hakka (and Korean). Hakka is identified as 'hak' in
ISO 639-3 but does not have a code in older parts of ISO 639.



> According to
> <http://www.ling.gu.se>, around 3000 of those languages are
> spoken by fewer than 10,000 people, and the likelihood of any AT being
> personalized for that small a population is minimal.  We need to be able
> to link the user to an AT that shares at least one language with them.

As you say, the number of languages supported by AT is much smaller than
the number of living languages.
But the ISO 24751 needs & preferences are not only about matching users
with AT: we also want to match resources in specific languages (including
subtitles, captions, etc) with languages that a user understands. I think
that is a much wider range than what is supported by AT.


> Outside the US, the incidence of multilingual people is fairly high, I
> think.


For example, in Europe, multilingualism is so important that we had a
"Commissioner for Multilingualism" in the European Commission between 2007
and 2010 (later  re-merged into the education and culture portfolio; the
EU also maintains a multilingualism portal at
<http://europa.eu/languages/>).

Best regards,

Christophe


P.S. The distinction between language and dialect is not simply a
linguistic question but heavily influenced by politics. Hence the
statement: "a shprakh iz a dialekt mit an armey un flot" - "a language is
a dialect with an army and a navy":
<https://en.wikipedia.org/wiki/A_language_is_a_dialect_with_an_army_and_navy>.

>
>
> Denis Anson, MS, OTR
> Director of Research and Development
> Assistive Technology Research Institute
> Misericordia University
> 301 Lake St.
> Dallas, PA 18636
>
> voice: 570-674-6413
> fax: 570-674-8054
>
> danson at misericordia.edu<mailto:danson at misericordia.edu>
>
>
>
>
> On Oct 4, 2012, at 3:43 PM, Colin Clark
> <colinbdclark at gmail.com<mailto:colinbdclark at gmail.com>>
>  wrote:
>
> +1. This issue is clearly more subtle than I understood, and I'm learning
> a ton from Christophe's reports.
>
> Colin
>
> On 2012-10-04, at 3:23 PM, Gregg Vanderheiden wrote:
>
> Christophe, can you take ownership of this  -- and work with everyone to
> find a resoultion?
>
> ---
> Colin Clark
> Technical Lead, Fluid Project
> http://fluidproject.org
>
> _______________________________________________
> Architecture mailing list
> Architecture at lists.gpii.net
> http://lists.gpii.net/cgi-bin/mailman/listinfo/architecture
>
> _______________________________________________
> Architecture mailing list
> Architecture at lists.gpii.net
> http://lists.gpii.net/cgi-bin/mailman/listinfo/architecture
>


-- 
Christophe Strobbe
Akademischer Mitarbeiter
Adaptive User Interfaces Research Group
Hochschule der Medien
Nobelstraße 10
70569 Stuttgart
Tel. +49 711 8923 2749



More information about the Accessforall mailing list