Current requests

Luc Pardon lucp at skopos.be
Wed May 20 18:06:25 CEST 2015


Allow me to beat a few dead horses.

On 20-05-15 05:18, Peter Constable wrote:
> If a user needs to tag some document as gl-ao1990, there is nothing whatsoever that prevents them from doing that now, without the change Luc has requested, and that tag would be fully valid. 

Yes, that's true. But that brings the question: why bother with prefixes
at all? Something doesn't make sense here.

And look at it from the procedural side (assuming that a registration
request has come in, as in this case). BCP47 3.5 states very clearly (my
emphasis) that this request can _only_ be rejected "because of
significant objections raised on the list". That rule is there for a
reason.

In other words: _absence_ of significant objections means it _must_ go in.

Now, how can we tell an applicant "oh, don't worry, you don't need it,
go tag as you please and leave us alone" and then maintain with a
straight face that this was a significant objection that made it really,
really impossible for us to approve the request?

>
All that is lacking by non-addition of the prefix tag is drawing
attention to gl-*-ao1990 as a use case that's significant enough to have
been called out in the registry.

Yes again. But there is a more sinister aspect to it. The non-addition
(or even the absence) may also be interpreted as a political statement
by IANA. It might be seen as the IANA taking sides in debates that are
sometimes hotly contested.

For the request under discussion, see:

   https://en.wikipedia.org/wiki/Reintegrationism

If IANA rejects gl-ao1990, it actually says "writing Galician in a
Portuguese orthography is NOT RECOMMENDED" (BCP 4.7 3.1.8 negated) and
"in our view, this practice it too insignificant to draw attention to it".

I can't see how the people at IANA would think that a good idea.

So it's better to err at the safe side.


Of course, all of the above means that I fully agree with you when you say:

> Having said that, my perspective is that if there is some identifiable user community for gl-ao1990, then I think that's a sufficient condition for listing gl as a prefix for ao1990.

Luc Pardon


More information about the Ietf-languages mailing list