Re: [idn] URL encoding in html page

[John C Klensin <klensin@jck.com>]

--On Tuesday, 02 April, 2002 09:27 -0800 Dave Crocker
<dhc@dcrocker.net> wrote:

> At 10:57 AM 4/2/2002 -0500, John C Klensin wrote:
>> While the IETF does not design or standardize user interfaces,
>> it seems to me that discussion about user interfaces and what
>> functions they do, or do not, require is entirely appropriate.
>> If we specify a wire ("network interchange") standard that is
>> intended to provide some functionality, and that functionality
>> cannot be rationally implemented in a user interface,
> 
> The only problem is that this extended discussion is that it
> is not serving that purpose.
> 
> It is not assessing the UI issues correctly, it is confusing
> protocol issues with implementation issues, AND it is trying
> to specify UI designs.  None of which is constructive to an
> IETF working group effort.
> 
> Frankly, both of these problems are persistent in IDN.  We
> start with a topic that has a generic relevance to the work --
> although we often engage in that discussion far past the time
>...
> John, as you well know, IETF working group progress requires
> constantly enforcing the scope restrictions specified by the
> charter.  Ensuring forward progress often has more to do with
> limiting scope of discussion than with anything else.
> 
> Perhaps it is silly of me to worry about the difference
> between pleasant academic discussions, versus practical
> engineering work that results in practical protocols developed
> in a timely manner.
> 
> For some reason, I have been under the impression that the
> latter is the purpose of the IETF.

Dave, assume, at least for the purposes of this discussion, that
I agree with you about at least the first part of this (I do
agree about much of it).   I've been disappointed that the WG
Chairs have, for whatever reason, been ineffective in shutting
off useless and redundant lines of discussion. And I recognize
that the problem is largely not theirs, but that a significant
fraction of the active participants on the WG list seem to want
or need to move forward with it no matter what -- this
collection of cats seems determined to not be herded, and I
doubt that either of us could do much better were we sitting in
the Chair.

I wish, too, that the WG would simply omit something and go
silent -- what is disturbing to me about this round of
discussions is how much they are going over old ground, without
contributing any new information or insights -- and let the IESG
do something, and move on (more on this below).   There too, we
probably agree, at least in large measure.

What I am objecting to is your tactics.  I don't think we make
progress by making up rules and trying to enforce them, or by
bending the rules to your (or even my) personal convenience.  To
take one of the clearer examples, I found the "user interface"
discussion during the Minneapolis "keyword" BOF interesting and
helpful, not because anyone was suggesting that we standardize a
UI, but because it provided real (and new, to at least some of
us) information and details about the context in which some of
this work would be used.  I think it would have been helpful to
spend more time teasing protocol requirements out of that
presentation, but the time in the BOF was overscheduled and we
had to spent time dealing with criticisms of the fact that the
presentation was permitted instead.  

Holding a WG to its charter is certainly appropriate, but

(i) The charter doesn't say what you seem to think it does.  The
wording "...specify the requirements for internationalized
access to domain names and to specify a standards track protocol
based on the requirements..." was very carefully chosen.  It
does not say "produce a solution that gets Unicode names into
the DNS". I think reasonable people can read what it does say as
including discussions about what is actually required and where
in the stack to do it (those two options were certainly
considered in-scope when the charter was written) and even what
character repertoires are required for us to consider the
"access" to be "internationalized".   The charter goes on to say

		The scope of the group is to investigate the possible
		means of doing this and what methods are feasible given
		the technical impact they will have on the use of such
		names by humans as well as application programs, as well
		as the impact on other users and administrators of the
		domain name system. 

Which, it seems to me, makes understanding of application (and
even UI) requirements on topic, and probably even makes consumer
confusion issues on topic.
		
		A fundamental requirement in this work is to not disturb
		the current use and operation of the domain name system,
		and for the DNS to continue to allow any system anywhere
		to resolve any domain name.

Which, however wrongheaded or misguided some of us think the
UTF-8 discussion to be, and however ill-advised a few of us
think one or two of the clarifications in RFC 2181 are, makes
that arguably in-charter too.


(ii) It is possible to argue that once a WG is chartered to
produce a protocol, it must produce a result, and the IESG must
accept that result because the WG put in a lot of work on it and
reached some sort of consensus.  But that makes the IESG as an
approval body, and IETF Last Call, nearly pointless.  It also
suggests that a WG should not be shut down until and unless it
reaches consensus.   And those conclusions either lead to
guarantees that the IETF will periodically eject very
poor-quality work or that some WGs will go on forever.  I can't
accept either of those outcomes, and I think you have argued
against both of them in other contexts.  Otherwise, it seems to
me to be much better that a WG be able to conclude that it
cannot produce a reasonable protocol (or a protocol that
satisfies all of the requirements as it sees them) and ask to be
shut down than that the IESG have to do that against the WG's
will (or perceived charter).  I don't think that outcome is
desirable here, but, if it is to be even possible, there must be
the possibility of discussing it and it seems to me that you
have been advocating suppressing discussions along those lines
too.

I don't see that as constructive, and I don't see it as
consistent with either the charter or with open exchange of
views in the IETF.

    john