Message-Id: <6.2.3.4.2.20050912112700.05348ca0@mail.jefsey.com>
Date: Mon, 12 Sep 2005 13:29:51 +0200
To: "Peter Constable" <petercon@microsoft.com>, <ietf@ietf.org>
From: "JFC (Jefsey) Morfin" <jefsey@jefsey.com>
In-Reply-To: <F8ACB1B494D9734783AAB114D0CE68FE0712E303@RED-MSG-52.redmon
	d.corp.microsoft.com>
References: <F8ACB1B494D9734783AAB114D0CE68FE0712E303@RED-MSG-52.redmond.corp.microsoft.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Cc: ltru@ietf.org
Subject: RE: RFC 3066 bis Libraries list
Precedence: list
Sender: ietf-bounces@ietf.org
Errors-To: ietf-bounces@ietf.org

At 06:38 12/09/2005, Peter Constable wrote:
> > From: "JFC (Jefsey) Morfin" <jefsey@jefsey.com>
> > RFC 3066 Bis imposes new constraints on the existing language tags
> > software libraries.

Dear Peter,
whatever the way you want to say it, these libraries have now to meet 
new specs they had not to meet before. I do not know if you already 
developed some piece of software, but this means there will be more 
lines of codes in RFC 3066 bis librairies and more possible bugs. 
Please do not confuse constraints of accuracy and quality on the 
libraries, constraints on the langtags, and constraints on the users 
who would be now to accomodate for ever a non end to end 
non-interoperable (except may-be in e-commerce) security leaking system.


Before approving a "BCP", and the whole IETF in not opposing the 
Draft must investigate the supposed existing running code. To show 
make sure that users are (will be) supported correctly. This concerns 
for example the _added_ information (scripts), the way they address 
"x-tags" private use, the filtering and negociatioin performed, or 
langtags such as qac-qark-aa, etc. They must also know what are the 
other libraries they could use. This is this much needed review I 
plan since the WG-ltru neither considered it, nor considered 
describing library validation tests. IMHO it is more positive than 
biaised considerations over RFC 3683.

RFC 2026 talks about the need to consider the interests of all the 
affected parties and of openness and fairness: I try to consider and 
support them all in an open and fair practical way, hence the reason 
of my call on this list, so every library can be quoted. As the most 
knowledgeable and the leading person IRT that project, you should 
help me and provide a list of all the librairies you know. The same 
about the authors? Or is there a problem I miss?

jfc

>We need to be more careful in describing the proposed revision to RFC
>3066 (aka RFC 3066 bis) wrt exiting libraries that conform to RFC 3066:
>every tag valid under the terms of RFC 3066 bis will be recognized by an
>existing library written to conform to RFC 3066. Not every tag that
>*could* be recognized by such a library would be valid under RFC 3066
>bis, but every tag actually valid today under RFC 3066 is also valid
>under RFC 3066 bis.
>
>
>The draft was written with careful attention to ensuring compatibility
>with existing libraries written to support RFC 3066. The draft can be
>said to impose new constraints that existing libraries would not impose;
>I don't see how it could be said that the draft imposes new constraints
>on those libraries.
>
>Peter Constable



_______________________________________________
Ietf mailing list
Ietf@ietf.org
https://www1.ietf.org/mailman/listinfo/ietf