Machine Translation

Peter Constable petercon at
Sun Sep 13 00:16:43 CEST 2009

Such a debate should be focused on particular usage scenarios. The scenario Debbie described, _if_ we think it useful for BCP 47 tags, could warrant a variant subtag but not an extension. Scenarios involving detailed systems for quality rating, _if_ we think it useful for BCP 47 tags, could warrant an extension but would make less sense for variant subtags.


-----Original Message-----
From: ietf-languages-bounces at [mailto:ietf-languages-bounces at] On Behalf Of Doug Ewell
Sent: Saturday, September 12, 2009 12:22 PM
To: ietf-languages at
Subject: Re: Machine Translation

Felix Sasaki <felix dot sasaki at fh dash potsdam dot de>: wrote:

> Other practical information which one could not pack into a broad data 
> category "machine translation" easily (to use Peter's terminology), 
> but not easily in the "language tag" field would be: name of system 
> that generated the translation (maybe several ones where used ...), 
> quality of the input, quality rating of the system (e.g. BLEU score).
> IMO these fine grained differences are necessary for making use of 
> this kind of metadata, and I don't see a clear use case for a broad 
> "machine translated" sub tag.

This is exactly why, IF this type of information is to be captured in a BCP 47 tag at all, it would be better to create an extension rather than registering a suite of variants, more or less haphazardly, which might impose too much specificity on some uses and not enough on others.

We should probably have the "BCP 47 or not" debate first, and only then, if it is decided to support this in BCP 47, worry about whether to use variants or extensions.

Doug Ewell  |  Thornton, Colorado, USA  | RFC 5645, 4645, UTN #14  |  ietf-languages @

Ietf-languages mailing list
Ietf-languages at

More information about the Ietf-languages mailing list