Request for variant subtag fr 16th-c 17th-c Resubmitted!

Peter Constable petercon at
Tue Dec 19 03:35:19 CET 2006

From: ietf-languages-bounces at [mailto:ietf-languages-bounces at] On Behalf Of Ciarán Ó Duibhín

> As a linear list of languages, ISO 639 may divide up 
> varieties which belong together from another point of 
> view, and which would be kept together in a different 
> linear classification.  

Categories with scope = macrolanguage, as introduced with ISO 639-3, exist for this very purpose.

> What has been proposed is to give some support to 
> alternative classifications 

That may be, but it doesn't change the fact that "fr" and "frm" have different semantics, that the intended semantics of "fr" are "Modern French", and that the intended semantics of "frm" are "Middle French".

> As I understand it, the semantics of "fr-variantx" 
> and "frm-variantx" would be that the texts so tagged 
> belong with "fr" and "frm" respectively according to 
> the ISO 639 definition of those prefixes.

That is plausible only if "fr-variantx" and "frm-variantx" have distinct semantics; but it was my impression that "1606Nict" was intended to denote one specific varietal range. 

I did not have opportunity (or, admittedly, interest) to read the entire, drawn-out thread, so perhaps it's all been explained and I've failed to notice. If the intent is that a particular document could be tagged either "fr-1606Nict" or "frm-1606Nict" simply based on personal preference, then that indicates to me that the semantic is just one concept, and so there should be one tag. 

But if the idea that one might have two documents A and B such that, if "fr-1606Nict" were appropriate for A and "frm-1606Nict", for B, then the opposite would *not* be appropriate, then that indicates to me that the semantics are *two* concepts; and if there are two concepts, then the registration can document what they are.

> People who wish to retrieve fr but not frm, or 
> vice versa, continue to do so.

If there is just one concept to be declared, then IMO users would be much better off having agreement on one best practice as to how to tag the content rather than saying people can tag in two different ways that are assumed to be equivalent in spite of principles that argue otherwise. One leads to interoperability; the other hinders, even opposes, interoperability.

> Isn't it likely that there will be more instances 
> of this sort, and that when users have tired of 
> retrieving areas of overlap through disjunctions
> involving differently named subtags which mean 
> the same thing, that there will be a demand to make 
> the names variantx and varianty the same?

They should not only make variantx and varianty the same, they should make mmm-variantx and nnn-varianty the same, particularly if, by first principles, if mmm <> nnn, then mmm-variantx and nnn-variantx should *not* mean the same thing.

Peter Constable

More information about the Ietf-languages mailing list