Translations

M. T. Carrasco Benitez writes:
 > Should translations request be included in the same model as linguistic 
 > versions.  Translations request (to humans or machine) could be 
 > considered as "virtual linguistics versions"; i.e., the do not exist at 
 > this moment, but they can be available.
 > 
 > Transliterations could be included in this model.
 > 
 > If this is the case, some nomenclature should be defined.  Example:
 > 
 >  it-ht   (Italian, human translation)
 >  it-mt   (Italian, machine translation)
 > 
 > Tomas
 > 
 > 

This could get very interested in negotiating the language quality.
e.g. Accept-Language: it:q=1.0, it-mt;q=0.7, it-ht;q=0.5
(If you can't get an original Italian document make sure the machine
translation is better than the rough human draft.)

Having a standard measureable q-value may be all that is really 
required? e.g. it-mt;q=0.5 == it-ht;q=0.5