Re: Comments on draft-ietf-kitten-gssapi-extensions-iana-02.txt

[Jeffrey Hutzelman <jhutz at cmu.edu>]

--On Tuesday, March 25, 2008 01:38:20 PM -0500 Nicolas Williams 
<Nicolas.Williams at sun.com> wrote:

> On Tue, Mar 25, 2008 at 01:27:10PM -0400, Jeffrey Hutzelman wrote:
>> --On Tuesday, March 25, 2008 04:32:40 PM +0000 Alexey Melnikov
>> <alexey.melnikov at isode.com> wrote:
>>
>> > Indeed, "obsolete" is very useful. Adding an entry as experimental ("I
>> > don't know if this is going to work, I might have to change the API
>> > later on") might be useful as well.
>>
>> Nope; it doesn't matter.  The purpose of the registry is to register use
>> of  the name, not to serve as a development tracking database.  I
>> suppose one  might define a whole taxonomy of interface stability
>> designations and what  they mean, but really, an IANA registry is not
>> the place for that.
>
> Two minutes earlier you indicated that obsolete is fine.  I assume
> that's still your view, but that you're arguing that for any other
> interface stability attribute we should not record it in the
> registration.

Pretty much.  Information about the stability or "status" or whatever turns 
the registry from a collection of mostly-static things into a dynamic 
database that is trying to do more than intended.  IANA is a registration 
service, not a database engine.  The primary purpose of a registry is to 
avoid collisions and to let you find out what a name names and where to 
look for more information.  Anything more is extraneous.

I'll accept an obsolete flag with reference, because it lets you provide 
information on where one should look instead of using the obsolete name, 
and there often _is_ no other place for such a pointer.

-- Jeff
_______________________________________________
Kitten mailing list
Kitten at ietf.org
https://www.ietf.org/mailman/listinfo/kitten