Re: [Ltru] Re: Update ltru-initial-05

Dear Frank,
Interesting points, we would have discussed and hopefully addressed 
had we studied in common our Charter and made a clean sheet work.

At 03:13 16/10/2005, Frank Ellermann wrote:
>r&d afrac wrote:
>  > I thought the idea was to eventualy merge width CLDR?
>
>Why do you think that ?  I'm just curious.  It's less than
>a week that I pulled another public en-UK stunt instead of
>en-GB in a Usenet newsgroup... <sigh />

I just read the charter, looked at who are the authors, read the list 
of the Unicode officers and members, joined Unicode, discussed with 
people from their organisations, been contacted by some. Been 
inquired by industry people who are no members. By concerned Govs. 
Discussions inside ISO and other SSDOs. Exposure after the second LC helped.

The point is not what people plan. But what the problems are and what 
are the possible solutions and budgets.

> > Every time I asked about the dissemination system and its
> > size, usage, etc. I was responded "like Unicode"?
>
>http://www.unicode.org/Public/UNIDATA/NamesList.txt 768 KB

I am afraid you missed the point. Please access the CLDR project.

- http://www.unicode.org/cldr/
- Directed by Mark Davis
- quoted in our charter
- none of the simple questions asked on this list about their IPRs 
was responded
- the determination to exclusive - any IETF only project would have 
rejoiced at "0-".

>Yes, that's about twice (?) the size I'd expect for a 3066ter
>registry,  For 3066bis it's less than 80 KB.  If you still
>don't believe it let's wait until it's created, you can then
>verify that it's really hosted by IANA and less than 80 KB.

RFC 3066bis is of low interest irt. size. The size and use for "IANA" 
management is what they call RFC 3066 ter, quater, etc. A realistic 
forecast is between Unicode or the like and the DRS (an ISO 11179 
conformant Distributed Registry System).

> >> I'd gues that less than 1000 persons worldwide have ever
> >> heard of "3066bis".
>
> > Please do not say that after an IESG 3066bis LC.
>
>=> John reviewed the drafts.  Bruce explained why he didn't.
>Probably some others also looked at it, but didn't bother
>to comment.

Thank you for this testimony. I agree. I suppose that a few tried but 
find it quite obscure (no definition, etc.) and abandonned.

> > I would say that at least 500 knows it through my team
>
>Those folks also didn't bother to comment it (assuming that
>they exist, I don't share that opinion)

I am not sure what "assuming that they exist, I don't share that 
opinion" may mean? Which opinion.

I am not interested in commenting on RFC 306 bis, just in being able 
to use it. To remind our need was my job. You are the experts. We are 
the users. I am interested in sharing only if I see you do not 
deliver what I need.

I will recall two chanels I quoted (but I run/share in some others):

- AFRAC which works on the way to make every common information 
avalaible to users - so is interested in the technical quality of the 
deliverables; AFRAC is a small French R&D team with connections to 
various lists such as ISOC, Eurolinc, etc.

- NICSO which is a users / govs / small ccTLDs / community oriented 
think-tank. It broadcast specialised notes. Obviously they are not 
read by all the destinees.

I think a global awareness of 500 (people knowing there is a major 
problem, where and I will keep them posted) is conservative. But this 
people are not interested in a "possible". Now the document is 
approved, it enters into their field as a political move or as a 
legal violation. Politics are now interested. Lawyers will be 
interested when it starts being enacted.

> > Now, you would also imply that may be less than 100 ever
> > read it?
>
>Yes.

Thank you for this evaluation.

> > So that IETF would go by minority consensus?
>
>No.  But those are TAObis / PECSI / general topics.  I'm still
>trying to find out how it's supposed to work - especially if
>there are conflicts, or if somebody only claims that there's
>a problem.

This is a good question. I was enrolled in the principle of a WG in 
giving conditions which were not respected. A charter came which was 
not respected. There is no way to really block the work of a WG even 
if you think it detrimental to the IETF. This is IMHO the main problem.

The only way left is I think the one I was able to follow:
- to make sure the document is the least harming one in the case it 
would be enforced. But you saw what it means.
- to try to work out an external consensus on a least confusing maket solution
- to prepare and engage the legal, market, political mechanic to 
block the error to try to enforce a practical consensus.

The problem is, if there is only one person, there may be an error, 
but at least there will not be confusion over a debate. The problem - 
as I experimented it - is when because there is only one (or a very 
limited kernel of) person who think paying attention to the IETF is 
worth the effort. If the problem is not that important for external 
people, we have a standardisation confusion. If there is a conflict 
and if it is vital to the external people or if they are no 
interested in other IETF issues, their only solution they have is to 
kill the IETF.

I hope that now the IESG has approved the document that danger is 
over. But it has been more than real. I never hidden it. And this is 
very easy now. This is why PECSI is important.

> >> E.g. I have a link <http://www.iana.org/assignments/lang-tags/>
>
> > This seems an odd discussion because "lang-tags" is a
> > directory of 92 files. Not a registry of hundredth and then
> > thousands of entries?
>
>It was only an _example_ for a public link to the relevant part
>(my personal POV) of the old registry = "stuff I won't find in
>the source standards".  I could as well link to a single file
><http://www.iana.org/assignments/language-tags> or to the HTML
>view <http://www.iana.org/assignments/lang-tag-apps.htm> of
>this directory.
>
>For a fun effect try:
>http://purl.net/xyzzy/-AWFID/inlinktext:%2522language%2520tags%2522

This is good. But this is a searcher page (good idea). What we need 
is an operational registry supporting real applications/usages.

> > You seem to consider the langtags will not really be used
> > and known?
>
>What I consider is that it may take years until the last user
>of language tags will note that RfC 3066 has been replaced.

I am not as pessimitic as you are. The work achieved is of interest 
for the "globalization" (end to end ASCII English 
internationalization + localisation). I understand that CLDR is 
progressing well. Let assume it is in line with their Q2.2006. 
Globalization will then have most of its tools since they keep the libraries.

I suppose the WSIS, meetings like Berlin, the UNICODE meeting will 
give it some momentum. Then the meta-spam issue will make it known. 
So a solution will have to be proposed by mid-next year. That 
solution is the "0-/1-" userspaces and ISO 11179 conformance. ISO 
639-3/6 will have been published. This gives a good schedule and 
permits to work.

The point now is for us to publish the langroot, to work on the DRS 
servers logic and better document and support "globalisation". For 
the WG-ltru it is to finish the work over filtering. But it has to do 
it seriously in caring about the legal implications. The hope is the 
Internet Governance obliges the IETF solution to be the default of 
the common one the soonest to reduce costs, confusion and 
balkanisation as in IDNA.

The WG-ltru also has to do it in avoiding mad threads like the "route 
66" one. The approval of the IESG protects now the IETF against DoS 
(I hope) but will put more focus on this mailing list by more people.

Take care.
jfc

_______________________________________________
Ltru mailing list
Ltru@ietf.org
https://www1.ietf.org/mailman/listinfo/ltru