Phillips, Addison 2008-10-04 18.41:
The big danger with variants is using two or more that are mutually exclusive. '1901' and '1996' are mutually exclusive; there is no logical way in which both could ever be used in the same tag. [...]
Unless we say that year digits should only be used to "date" an existing language or norm - and not itself serve as reference to a any language or norm - we could claim that 1996 is a modification of 1901. And thus we could claim that "de-1901-1996" would be meaningful. (Though redundant.)We could claim all sorts of things, but the current RFC already spells this particular case out and there exists several years of usage--predating RFC4646--that say that 1996 is not a modification of 1901.
I don't think the RFC goes into detail about development of German ortography and how 1996 relates to 1901 in that sense. What you mean is that we cannot, per the RFC and its history, add 1996 to the 1901 subtag, in order to modify what the 1901 subtag means.
(Btw, I never suggested that de-1901-1996 was good tagging, I only discussed what is "obvious" in this case.)
By keeping a strict year meaning to year subtags, we can help that registrants avoid such things [that I proposed] as 1996groene and 1996witte.Why? More importFrom ltru-bounces at ietf.org Sun Oct 5 17:32:38 2008
Return-Path: <ltru-bounces at ietf.org> X-Original-To: ltru-archive at megatron.ietf.org Delivered-To: ietfarch-ltru-archive at core3.amsl.com Received: from [127.0.0.1] (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 8051A3A68E5; Sun, 5 Oct 2008 17:32:38 -0700 (PDT) X-Original-To: ltru at core3.amsl.com Delivered-To: ltru at core3.amsl.com Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id EEF453A68E5 for <ltru at core3.amsl.com>; Sun, 5 Oct 2008 17:32:37 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.69X-Spam-Level: X-Spam-Status: No, score=-2.69 tagged_above=-999 required=5 tests=[AWL=-0.091, BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id DXab2Y7ZO6yy for <ltru at core3.amsl.com>; Sun, 5 Oct 2008 17:32:36 -0700 (PDT) Received: from smtp.domeneshop.no (smtp.domeneshop.no [194.63.248.54]) by core3.amsl.com (Postfix) with ESMTP id 851423A68D8 for <ltru at ietf.org>; Sun, 5 Oct 2008 17:32:36 -0700 (PDT) Received: from cm-84.208.108.246.getinternet.no ([84.208.108.246] helo=s.local) by smtp.domeneshop.no with esmtpa (Exim 4.68) (envelope-from <lhs at malform.no>) id 1Kme26-0001dC-HZ; Mon, 06 Oct 2008 02:33:06 +0200 Message-ID: <48E95CC2.9030402 at malform.no> Date: Mon, 06 Oct 2008 02:33:06 +0200 From: Leif Halvard Silli <lhs at malform.no> User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.1b1) Gecko/20060724 Thunderbird/2.0a1 Mnenhy/0.7.4.666 MIME-Version: 1.0 To: "Phillips, Addison" <addison at amazon.com> References: <mailman.1595.1223013734.4981.ltru at ietf.org> <2E5435EE3D0B48C69C76968E47EFABA3 at DGBP7M81> <48E7215E.3060005 at malform.no> <4D25F22093241741BC1D0EEBC2DBB1DA014C7FAFDB at EX-SEA5-D.ant.amazon.com> In-Reply-To: <4D25F22093241741BC1D0EEBC2DBB1DA014C7FAFDB at EX-SEA5-D.ant.amazon.com> Cc: LTRU Working Group <ltru at ietf.org>, Doug Ewell <doug at ewellic.org> Subject: Re: [Ltru] Prefixes X-BeenThere: ltru at ietf.org X-Mailman-Version: 2.1.9 Precedence: list List-Id: Language Tag Registry Update working group discussion list <ltru.ietf.org> List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/ltru>, <mailto:ltru-request at ietf.org?subject=unsubscribe> List-Archive: <http://www.ietf.org/pipermail/ltru> List-Post: <mailto:ltru at ietf.org> List-Help: <mailto:ltru-request at ietf.org?subject=help> List-Subscribe: <https://www.ietf.org/mailman/listinfo/ltru>, <mailto:ltru-request at ietf.org?subject=subscribe> Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: ltru-bounces at ietf.org Errors-To: ltru-bounces at ietf.org Phillips, Addison 2008-10-04 18.41:
The big danger with variants is using two or more that are mutually exclusive. '1901' and '1996' are mutually exclusive; there is no logical way in which both could ever be used in the same tag. [...]
Unless we say that year digits should only be used to "date" an existing language or norm - and not itself serve as reference to a any language or norm - we could claim that 1996 is a modification of 1901. And thus we could claim that "de-1901-1996" would be meaningful. (Though redundant.)We could claim all sorts of things, but the current RFC already spells this particular case out and there exists several years of usage--predating RFC4646--that say that 1996 is not a modification of 1901.
I don't think the RFC goes into detail about development of German ortography and how 1996 relates to 1901 in that sense. What you mean is that we cannot, per the RFC and its history, add 1996 to the 1901 subtag, in order to modify what the 1901 subtag means.
(Btw, I never suggested that de-1901-1996 was good tagging, I only discussed what is "obvious" in this case.)
By keeping a strict year meaning to year subtags, we can help that registrants avoid such things [that I proposed] as 1996groene and 1996witte.Why? More importantly, iantly, isn't that a matter for the registration process to determine/enforce, rather than the RFC?
Because several have pointed out that years could be a good thing to use in tags. Determine/enforce? No. Only advice.
This to me suggests to that, in reality, there is consesus that year subtags are strictly to be understood as years. Or else, as mentioned above, "de-1901-1996" could be meaningful.No. The two German subtags you cite above are mutually exclusive because they both refer to specific orthographies. That is, language tag variant subtags refer to variations in language, not to years.
I don't see that we fundamentally disagree when you say that the years refers to "variations in language, not to year". Perhaps we can agree that years refers to an /dated/ variation of some norm?
If 'de-1996" means "Norm of Written German, the 1996 reform", then I would say that "Norm of Written German" is represented by "de" (which also includes the suppressed '-latn'), and that "1996" represents "the 1996 reform".
This does not mean that "1996" only means the year - after all, it is registered as an extension of the "de" subtag. But if we go the full route in the other direction, then we could claim full autonomy and just use lang="1996".
The dated language norm in question here is "Written Hochdeutsch" - which is a norm that normally is referred to with the tag "de" or "de" plus a region subtag. I could in fact "(mis)use" "de-1996" to signify "written form". But that very fact does not mean that it is "1996" which means "written form". The "written form" is hidden in the "de" tag. Or else, we could not have applied "1996" to it.
Let's use a more useful example. 'boont' and 'scottish' are both registered variant subtags that can be applied to English (en). They are mutually exclusive because they refer to specific regional dialects/jargons. One cannot speak Boontling Scottish-wise or vice versa. Replacing 'boont' with '1890' and 'scottish' with (I dunno, let's pick a date) '1707' doesn't make the underlying language variations any different or mutually compatible.
If 'boont' was a development - aka (sub)dialect - of scotish, then 'en-scottish-boont' and/or "en-1707-1890" would be OK. But contrary to that artificial example, 1996 and 1901 refers to the written norm of the same German "dialect".
I think it is your intepretation of years (which is leaning towards an autonomy understanding of what the years represents) which is most at risk of allowing "de-1901-1996".
The normal assumtion is that a "naked" language subtag, such as 'de', refers to a language represented by a standardnorm. And so both 1901 and 1996 refers to a dated variant of the standard written norm of German.
This is inherently open to debate for each proposed variant, and I am sure ietf-languages will hold such debates whether we add a rule or not, so I do not support adding a new rule to attempt to regulate this behavior, especially not if we are trying to get through a Last Call.Of course it must be open to debate - we cannot know in advance what one want to register and how it relatest to "reality". And I do not think that the current lack of recommended region subtag prefixes for 1996 is a fundamental error. But I do think that it would be much more in line with the strict year meaning that you apply to "-1996" to list the region variants of "de" for which the 1996 dating is good.Really, when the prefix omits all mention of scripts or regions, we have the positive ability to use whichever onesmake sense,
Well, here, I must say that I have taken in what Doug said: the supress latin script for "de" is valid for 1996.
even if the registrants overlooked and omitted one. I suspect that remnant populations might use 1901 or 1996, depending, with "de-NA".
The keyword here is 'overlooked'.
I think the problem here is that you're suggesting that we should try to register what all of the "correct" potential tsn't that a matter for the registration process to determine/enforce, rather than the RFC?
Because several have pointed out that years could be a good thing to use in tags. Determine/enforce? No. Only advice.
This to me suggests to that, in reality, there is consesus that year subtags are strictly to be understood as years. Or else, as mentioned above, "de-1901-1996" could be meaningful.No. The two German subtags you cite above are mutually exclusive because they both refer to specific orthographies. That is, language tag variant subtags refer to variations in language, not to years.
I don't see that we fundamentally disagree when you say that the years refers to "variations in language, not to year". Perhaps we can agree that years refers to an /dated/ variation of some norm?
If 'de-1996" means "Norm of Written German, the 1996 reform", then I would say that "Norm of Written German" is represented by "de" (which also includes the suppressed '-latn'), and that "1996" represents "the 1996 reform".
This does not mean that "1996" only means the year - after all, it is registered as an extension of the "de" subtag. But if we go the full route in the other direction, then we could claim full autonomy and just use lang="1996".
The dated language norm in question here is "Written Hochdeutsch" - which is a norm that normally is referred to with the tag "de" or "de" plus a region subtag. I could in fact "(mis)use" "de-1996" to signify "written form". But that very fact does not mean that it is "1996" which means "written form". The "written form" is hidden in the "de" tag. Or else, we could not have applied "1996" to it.
Let's use a more useful example. 'boont' and 'scottish' are both registered variant subtags that can be applied to English (en). They are mutually exclusive because they refer to specific regional dialects/jargons. One cannot speak Boontling Scottish-wise or vice versa. Replacing 'boont' with '1890' and 'scottish' with (I dunno, let's pick a date) '1707' doesn't make the underlying language variations any different or mutually compatible.
If 'boont' was a development - aka (sub)dialect - of scotish, then 'en-scottish-boont' and/or "en-1707-1890" would be OK. But contrary to that artificial example, 1996 and 1901 refers to the written norm of the same German "dialect".
I think it is your intepretation of years (which is leaning towards an autonomy understanding of what the years represents) which is most at risk of allowing "de-1901-1996".
The normal assumtion is that a "naked" language subtag, such as 'de', refers to a language represented by a standardnorm. And so both 1901 and 1996 refers to a dated variant of the standard written norm of German.
This is inherently open to debate for each proposed variant, and I am sure ietf-languages will hold such debates whether we add a rule or not, so I do not support adding a new rule to attempt to regulate this behavior, especially not if we are trying to get through a Last Call.Of course it must be open to debate - we cannot know in advance what one want to register and how it relatest to "reality". And I do not think that the current lack of recommended region subtag prefixes for 1996 is a fundamental error. But I do think that it would be much more in line with the strict year meaning that you apply to "-1996" to list the region variants of "de" for which the 1996 dating is good.Really, when the prefix omits all mention of scripts or regions, we have the positive ability to use whichever onesmake sense,
Well, here, I must say that I have taken in what Doug said: the supress latin script for "de" is valid for 1996.
even if the registrants overlooked and omitted one. I suspect that remnant populations might use 1901 or 1996, depending, with "de-NA".
The keyword here is 'overlooked'.
I think the problem here is that you're suggesting that we should try to register what all of the "correct" potential tags arags are. But the generative nature of language tags is based on a quite different philosophy: we don't have to do the detailed linguistic research and weigh all manner of data to provide such things (as the old RFC 3066 kind of required).
Resian, its subdialects and the 1994 subtag was registered as a result of detailed linguistic research.
I feel that here you are saying almost the opposite of what you claim to be saying.
Example: Let's say they made a reform in Canadian French and chose "2008" as subtag for that reform. You then seem to be of the opinion that one could not put in the registry that the recommended prefix should be "fr" and "fr-CA". Because, before you would allow the Quebecians to be *that* spesific with regard to the prefix, you have a whole list of linguistic questions that must be answered before they should be allowed to claim that their own reform was only relevant to Quebec French.
So it seems like the 1996 subtag should have registered the region subtags it was good for - because, it was known which it was good for - it did not require investigation, it was 4 countries which agreed. And John has told that it is not forbidden to use 1996 with any other region tag than the ones it is recommended for. So, if we really agree about that, then, if it is relevant, one can just add 1996 to another region subtag, despite its lack of recommendation. (And if the usecase is really really relevant, one can just augment the list of prefixes in the next revision.)
-- leif halvard silli _______________________________________________ Ltru mailing list Ltru at ietf.org https://www.ietf.org/mailman/listinfo/ltru e. But the generative nature of language tags is based on a
quite different philosophy: we don't have to do the detailed linguistic research and weigh all manner of data to provide such things (as the old RFC 3066 kind of required).
Resian, its subdialects and the 1994 subtag was registered as a result of detailed linguistic research.
I feel that here you are saying almost the opposite of what you claim to be saying.
Example: Let's say they made a reform in Canadian French and chose "2008" as subtag for that reform. You then seem to be of the opinion that one could not put in the registry that the recommended prefix should be "fr" and "fr-CA". Because, before you would allow the Quebecians to be *that* spesific with regard to the prefix, you have a whole list of linguistic questions that must be answered before they should be allowed to claim that their own reform was only relevant to Quebec French.
So it seems like the 1996 subtag should have registered the region subtags it was good for - because, it was known which it was good for - it did not require investigation, it was 4 countries which agreed. And John has told that it is not forbidden to use 1996 with any other region tag than the ones it is recommended for. So, if we really agree about that, then, if it is relevant, one can just add 1996 to another region subtag, despite its lack of recommendation. (And if the usecase is really really relevant, one can just augment the list of prefixes in the next revision.)
-- leif halvard silli _______________________________________________ Ltru mailing list Ltru at ietf.org https://www.ietf.org/mailman/listinfo/ltru
Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.