[Standards] recommendation on glottolog codes?

[Sebastian Drude]

Thanks Dieter, and hi, all.

 

Indeed, I am involved in TC37 SC2, and actually leading the task force in WG1 to develop ISO/AWI 21636 <https://www.iso.org/standard/71300.html>  (a technical document rather than a standard, as it is not yet implemented anywhere, although a CMDI profile has been drafted).

 

This is to be a framework for describing inner-language variation, in particular identifying the dimensions of linguistic variation and recognizing them as in principle independent. That means, each instance of language use and each element in a language resource are a point in a multi-dimensional space.  The dimensions we are proposing are: 

1)     space (dialects etc.), 

2)     time (periods, epochs, etc.), 

3)     medium (e.g., oral/multimodal, written, …), 

4)     social space (sociolects, jargons, …), 

5)     [formality of] situation (registers), 

6)     person (individual varieties), 

7)     proficiency (learners interlanguage stages)

8)     performance peculiarities (for stutter, lisping etc.)

 

The new technical document will only cover the dimensions, NOT the values for each language in each dimension, so it does NOT attempt to cover any specific dialects of any language.

For that, Glottolog would be a point to start, although its coverage of language-internal variation is generally speaking (and admittedly:  <http://glottolog.org/glottolog/glottologinformation#dialects> “[…] we spent little effort on making dialect classifications consistent and on providing references for dialects.”) rather weak. Multitree is sometimes better (some resource represented in Multitree, that is), but lacks a well-established system for persistent identifiers and good justifications for dialect/language distinctions, among other things.

 

Of course, in an ideal world there would be a comprehensive database where also dialects and varieties in other dimensions could be easily registered and get a persistent code, and CLARIN, for one, would be an institution that would have enough standing and recognition to host and promote such a database, as would possibly the LinguistList or a small number of other networks or institutions, but at this point I do not see any institution, including CLARIN and its members and centres, willing to commit to set up and maintain such a system which needs quite some work – technical support as well as a revision process backed by a competent panel.

 

For languages (not varieties such as dialects etc.), I think a combination of ISO and Glottolog is the best what we can recommend at this time.

 

Given the well strengthened role of SIL and the US-stakeholders in ISO, I fear it will be hard to get any official recognition for Glottolog within the ISO framework, but if anyone has a good suggestion about how to go about it, I would be delighted to support this.

 

These are my comments on this…

 

Sebastian 

-- 

Sebastian Drude, Director

The Vigdís International Centre for Multilingualism and Intercultural Understanding 

The University of Iceland – Brynjolfsgata 1 – 107 Reykjavík – Iceland

Email sdrude at hi.is – Phone +354 695.6784

 

-----Original Message-----
From: Dieter Van Uytvanck [mailto:dieter at clarin.eu] 
Sent: Wednesday, 24 May, 2017 14:08
To: Piotr Bański <banski at ids-mannheim.de>; standards at lists.clarin.eu; Beeken, Jeannine C T <jeannine.beeken at essex.ac.uk>; Sebastian Drude <sdrude at hi.is>
Subject: Re: [Standards] recommendation on glottolog codes?

 

[I believe this thread was stalled - sending this again, now with Sebastian's work email added]

 

(trying to answer 2 mails at the same time, as 2 external participants are added)

 

On 05/05/2017 10:42, Piotr Bański wrote:

 

> Now, turning to the matter at hand directly. With both the name of 

> Martin Haspelmath and institutions such as SIL International and MPI 

> Jena in the background, and given the scale of the work behind 

> Glottolog, I personally share Dieter's sentiment, although as a 

> Committee, I guess we should better use a documented process for that, 

> and in the best of worlds, I would love to see that process pushed 

> onto ISO, and us giving a stamp of approval to a proposal that would 

> already be part of an ISO track (even if it were in the early stages 

> of the ISO process, with all the question marks entailed by that). 

> With my ISO hat on, I just can't imagine that no ISO activity would be 

> taking place in this regard, unless TC37 SC2 is in deep slumber.

> @Committee: Do we have anyone here with a direct channel to TC37 SC2 

> open, please?

 

[on  ISO/AWI 21636  =  <https://www.iso.org/standard/71300.html> https://www.iso.org/standard/71300.html]

 

I believe Sebastian (in CC) knows more about this.

 

> @Dieter: would you be willing to share the use case that prompted the 

> question? Or was it just general musing, given the existence of Glottolog?

 

Jeannine Beeken (CESSDA) asked what the status was of the "old" CLARIN recommendations on language codes: using ISO-639-3 and ISO-639-1.

 

I answered that these recommendations still hold, but that in certain cases there might be no suitable ISO-639-3 code. In such cases glottolog might be an alternative I suggested, with the promise to check with our standards committee.

 

Now ISO/AWI 21636 might also be an alternative, but that depends on the status - I'm sure Sebastian can comment.

 

best regards,

--

Dieter Van Uytvanck

Technical Director CLARIN ERIC

 <http://www.clarin.eu> www.clarin.eu | tel. +31-(0)850091363 | skype: dietervu.mpi

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clarin.eu/pipermail/standards/attachments/20170524/451d6101/attachment-0001.html>