Request for variant subtag fr 16th-c 17th-c RESUBMISSION

Addison Phillips addison at
Mon Jan 22 21:02:06 CET 2007


I think you missed my point. I'm not arguing that there isn't a
difference. Or that language isn't variable or hasn't evolved etc. I 
don't mean to show disregard for the variations of French. It's just 
that they aren't what concerns me here: I have no doubt such variations 
are real and exist. Very similar kinds of things can be said of (for 
example) English: the English of Shakespeare, for example, is typically 
"translated" into a more modern form, despite being considered to be 
"modern English" (as contrasted with Middle English). It doesn't matter, 
since I'm not debating or disagreeing with the distinction. My point was 
that it wasn't clear what to do looking ONLY at the requested records 
(as a registry user would do).

In particular, I note that you don't need the subtags '1606nict' and 
'1694acad' to distinguish late Middle French from early modern French 
documents. 'fr' and 'frm' do that quite nicely.

If the point of '1694acad' is to distinguish "Early Modern French" from
more modern "Modern French", that's fine. But the distinction (perhaps
necessarily) is quite fuzzy: some documents will be closer to Middle
French and some closer to Modern French (or proto-modern French) and so
forth. Your registration request seems more focused on differentiating
these documents from "late middle" documents. Similarly, the point of
'1606nict' would be to distinguish "Late Middle French" from "regular"
Middle French. Tagging transitional documents is tricky business already
(there is an element of--perhaps arbitrary--judgment necessary).

Identifying language variation is a tricky business below a certain
level of granularity. For most uses, these subtags will be unnecessary.
But certain communities will find some utility in them. The
problem is creating sufficiently clear descriptions and distinctions to
allow them to be generally useful in those domains. And, for 
registration purposes, demonstrating that there won't be many other 
potential subtags with different (perhaps arbitrary) distinctions for 
this same time period (that is, register quickly and repent at leisure).

Finally, I would echo Doug's response, which is that the Description 
fields are extremely long: they are more suitable for Comments. And I 
would avoid placing URIs into your comments/description, since URIs tend 
to be fungible over time (the "Cool URIs" dictum [1] notwithstanding)

If I were to request these records, I would tend to make them look more 
like the below. Note the omission of additional notes such as "used in 
Americas", "not completely stable" and so forth: these things go without 
saying or belong in external references. Note also that I removed the 
additional bits from the subtags--I would suggest that these might not 
be necessary and the year-only subtags have at least some history in the 
registry for indicating beginnings/ends of orthographic regimes:

Type: Variant
Subtag:  1694
Description: Early Modern French
Prefix: fr
Comments: 17th century French, as catalogued in the "Dictionnaire de
    l'académe françoise", 4eme ed. 1694; frequently includes
    elements of Middle French, as this is a transitional period

Type: Variant
Subtag: 1606
Description: Late Middle French (before 1606?)
Prefix: frm
Comments: 16th century French as in Jean Nicot, "Thresor de
    la langue francoyse" 1606;




Addison Phillips
Globalization Architect -- Yahoo! Inc.

Internationalization is an architecture.
It is not a feature.

More information about the Ietf-languages mailing list