Phonetic orthographies

Gerard Meijssen gerardm at
Fri Nov 24 20:19:58 CET 2006

Peter Constable schreef:
> I'm not so sure about this. I suspect that often there are differences of orthography that readers can cope with (even if they have strong feelings about orthography differences). 
> For example, there are probably no English readers for whom differences such as colour/color are an obstacle to making effective use of documents. The same would apply to change in German orthography, and I suspect much the same would be true in many more cases of alternate, practical orthographies for language.
> I don't think the same is true in general of the difference between practical orthographies and phonetic transcriptions. The average English speaker would not recognize as English a phonetic transcription of English using any commonly-used transcription -- make such a document effectively useless to them.
> That, IMO, is comparable the effective usefulness of getting a document in a language that as known but in a script that is not known.
> Peter
When a person is reading a document, he or she does not care at all what 
tag is giving to the language of the content. It is readable to him or 
it is not. When it is not, it is not useful. Computers cannot read 
anything. For them, a tag can make a difference between processing a 
text or not. I would not be surprised however when there are a 
significant amount of websites that identify themselves as English and 
are anything but English. I do know of applications that have lost a lot 
of their usefulness because they use the "Standard" languages and 
thereby limit the scope of their application even when the application 
would be useful for languages that the Standard does not recognise.

I am also convinced that persons who speak English to any degree have no 
use whatever for phonetic transcriptions of English. An IPA 
transcription would be useless anyway if all the Meta data indicates is 
that it is English or British / American.. such a Standards compliant 
distinction does imho not significantly identify it for a linguist to be 
truly useful.

Am I being difficult here ?

>> -----Original Message-----
>> From: ietf-languages-bounces at [mailto:ietf-languages-
>> bounces at] On Behalf Of Eric Muller
>> Sent: Tuesday, November 21, 2006 11:26 PM
>> To: Doug Ewell
>> Cc: ietf-languages at
>> Subject: Re: Phonetic orthographies
>> Doug Ewell wrote:
>>> I hear the following two lines of argument:
>>> [...]
>> 3. In the context of (written) language identification, we don't really
>> care about scripts per se, we care about orthographies. It is true that
>> for those languages where there are multiple orthographies we care to
>> distinguish, it is often the case that each orthography uses a different
>> script (Mongolian in Cyrillic vs. Mongolian in Mongolian vs. Mongolian
>> in Latin), so scripts look like the "primary" attribute. But that is
>> just an illusion; witness the fact that pretty soon we are forced to
>> treat as separate scripts Hans and Hant to maintain that illusion (no to
>> mention that there is no way to account, using scripts, for the two
>> modern written forms of modern Greek).
>> Rather than fight about what a script is (Mark), or equivalently what
>> ISO 15924 is about (Ken), and since the world of language tags is
>> isolated of other standards via its registry anyway, why not just say
>> that the four-chars subtags are orthographies and be done with it? That
>> does not preclude to use (and reuse) the subtag "Latn" as a convenient
>> name for those situations where there is just one (or a single dominant)
>> orthography that uses the Latin script.
>> Eric.

More information about the Ietf-languages mailing list