On 6 Mar 2014, at 17:22, Gérard Lang-Marconnet <[log in to unmask]> wrote:
> In fact Michael Everson is proposing no less than:
> 1: “Fr[ee]zing" part 1, that is directly contrary to the current text of ISO 639-1, so so needing to make essential changes in the text of ISO 639-1;
One does not have to make changes in a text to stabilize and “freeze” a standard. There are procedures for that.
> 2: "Retiring" part 2, that is directly contrary to the current text of ISO 639-2, so needing to make essential changes in the text of ISO 639-2; and what about the standardization of MARC in ISO 639-2/B ?; and what about collective language code elements in part 2 ?
In my opinion, 639-2 is an inadequate subset of ISO 639-3. The /T vs /B lists provide a small number of aliases, which are not problematic.
What *about* the collective language code elements? What is your point?
> 3: "Modifying the process for the attribution of ISO 639-3 code elements, in the case "of entities that are not controversial" (under the definition that ME does not find these entities controversial ?)",
> so needing to make essential changes in the current text of ISO 639-3;
I don’t believe so. Does the text of that standard state that the RA will only process requests once a year? (I don’t have a copy to hand.) If it does, then I’d revise it. If it doesn’t, then it’s a matter for the RA’s procedures, which do not need to be altered by an ISO ballotting process.
The text if ISO 15924 does not specify the turnaround time that the 15924/RA uses. The RA just decided its policy.
> 4: "Changing part 5, so that this standard could become used by anyone, including ME, or transferring the collective language code elements of part 2 in part 5, or even retiring part 5", all of that being directly contrary to the current text of ISO 639-5, so needing to make essential changes in the text of ISO 639-5;
If it isn’t being used, it’s because there’s no need for it.
> 5: "Retiring, or maybe completely changing ISO 639-6, that is directly contrary to the current text of ISO 639-6, so needing to make essential changes in the text of ISO 639-6.
I would propose WITHDRAWING ISO 639-6.
> So, if essentially changing the texts of parts 1, 2, 3, 5 and 6 of the ISO 639 standard does not constitute a "business case for a new ISO 639", I really would like to know what sort of proposition could be seen as constituting such a "business case” !
It would have perhaps been more helpful had you responded to the actual points I made. Here they are again.
ISO 639-6 is not useful because it’s a mess.
The granularity of ISO 639-6 would not seem to be useful in a practical way.
What is the risk in “estranging” small and remote” languages in terms of 639-3? Which “small and remote” languages do you mean? (By “remote” I suppose you mean “distant from wealth”.)
Which small languages are increasing in importance, in what sense?
While I am unhappy with the working timetable of the ISO 639-3 RA, and believe that its responsiveness can be improved for entities which are not controversial, I don’t see that the process is so broken given the nature of the changes that have been made in the last while. [You responded t
Is 639-5 actually in use by anyone? Who? Where? [I did NOT suggest changing it so that it becomes useful. I asked if anyone were using it.]
The Wikipedia provides a single database for access to any code element. No ISO version is required.
The market is met reasonably well by Parts 1, 2, and 3. Wider implementation of Part 3 would be directly helpful to me, as I note.
> however, on the other end of the spectrum, the code elements are the only way to reliably refer to small languages with only sparse resources,
What? Recently I’ve been working with Irish and Cherokee and Osage, all small languages with rather different types of resources available. The code elements I have needed are all in parts 1, 2, and 3 (ga/gle/gle, chr/chr, osa/osa.
We don’t need “code elements” to describe the attributes of such languages. We need an encyclopaedia for that. And we have one.
The revision and expansion of un-used parts of this standard sounds like a kind of “if you build it they will come” standardization, though. I still don’t see a reason to revise all of the parts of ISO 639.
Michael Everson * http://www.evertype.com/
|