Re: consolidating FreeformClass profiles?

On 12/3/14, 6:07 AM, Dave Cridland wrote:
> On 3 December 2014 at 02:44, Peter Saint-Andre - &yet <[email protected]
> <mailto:[email protected]>> wrote:
>
>     Seeing no objections, I'm going to make this change.
>
>
> I'm not wholly sure I fully understand the change you're proposing,
> which is why I've not commented.

The change is to not define a separate profile but instead point to the 
OpaqueString profile from draft-ietf-precis-saslprepbis. We are working 
hard to reduce the number of profiles, since multiplication of profiles 
beyond necessity will increase the possibility of user confusion.

http://tools.ietf.org/html/draft-ietf-precis-framework-20#section-5.1

> I suspect I need to spend some time getting to grips with Precis, and
> what the distinction is between Precis and Stringprep. Help here - such
> as a discussion at the XMPP Summit at the end of January - would be very
> welcome.

Unfortunately, now is the time to understand it, because we are pushing 
hard to finish this work in both PRECIS WG and XMPP WG.

This overview might be useful:

https://stpeter.im/files/i18n-tutorial.pdf

I might even have time to make a few updates to that soon so that it 
addresses your questions.

The following readings might prove useful:

http://tools.ietf.org/html/draft-ietf-precis-framework-20#section-1
http://tools.ietf.org/html/rfc4690
http://tools.ietf.org/html/rfc6885

> The only thing I feel qualified to comment on is that resources in XMPP
> shouldn't have spaces stripped; I think that'd be a surprising change in
> a handful of cases.

I'm proposing that we use this profile for resourceparts:

http://tools.ietf.org/html/draft-ietf-precis-saslprepbis-12#section-4.2

There is no space-stripping there, but there is conversion of non-ASCII 
spaces to ASCII space.

Peter

-- 
Peter Saint-Andre
https://andyet.com/