On Wed, 19 Feb 2003, Charles Lindsey wrote:
> Sorry, but I am already on so many mailing lists that I cannot manage any
> more.
If you are unwilling to be a member of the IMAP mailing list, then please
do not propose changes to IMAP.
> However, I think our discussions so far have been useful, in that they
> show that if/when UTF-8 get into news (or email), some extension to IMAP
> will need to be part of the package, and we have established some features
> that would be needed in such a package and others that would need further
> discussion.
I am afraid that you have misunderstood. It is not "some extension to
IMAP". It is a massive mandate upon any entity -- IMAP servers, POP3
servers, SMTP servers, and SMTP clients -- which processes message
headers.
> I think the next job is to clarify that "if/when" bit, which means
> negotiating with the IESG. After that we can come back to writing an IMAP
> extension in full knowledge of what it will have to accomplish.
I can save you the trouble.
UTF-8 headers will not go into any standards-track document relating to
news. Doing so requires unacceptable burdens on other protocols. Any
document which attempts to standardize such a practice is doomed to
rejection.
I had hoped that you would voluntarily abandon this idea once the problems
had been explained to you.
> >Your message had a header of "Newsgroups: local.usefor" which is
> >unreplyable. There is no such newsgroup as "local.usefor" here at the
> >University of Washington.
> Now I find that remark odd. Like many others on this list, I gateway my
> mailing lists into local newsgroups for convenience of handling. Not
> surprisingly, the evidence of that is visible when gatewayed back into the
> list. But that should not affect your ability to reply.
Your message arrived to me in email with these headers:
Date: Wed, 19 Feb 2003 15:54:36 GMT
Message-Id: <[EMAIL PROTECTED]>
Newsgroups: local.usefor
Path: chl
From: "Charles Lindsey" <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Subject: Re: IMAP and Netnews
Since there is no mention of the [EMAIL PROTECTED] email
address, any reply that I send would not go to the Usefor mailing list.
Furthermore, the reply would try to send to a newsgroup called
"local.usefor" which does not exist here.
I was forced to edit out the bogus newsgroup, and edit in the mailing list
email address, manually in order to compose this reply.
Please be courteous and use properly replyable headers. It is unseemly
for a document editor to send unreplyable messages.
> The Newsgroups
> header currently has no defined meaning within email, though it is
> commonly used as an informational header to indicate that a message has
> been posted and mailed
Not only is that the only sane definition of that header, but mail/news
software used by millions of people worldwide use that definition.
> I emailed two copies of that message, one using the "cc to poster" feature
> on my news agent, which would have included "To: [EMAIL PROTECTED]"
> and one to the Usefor list using "To: [EMAIL PROTECTED]".
> Both would have had my usual "From: ...". Both fully compliant AFAICS.
Are you saying that you deliberately composed a header to prevent my reply
to go to the Usefor mailing list? Or is your software broken?
I expect that your future messages will have:
From: "Charles Lindsey" <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
cc: [EMAIL PROTECTED], [EMAIL PROTECTED]
in the headers.
> A quick Googling shows that you read a wide variety of groups, and
> post regularly.
> Which, sadly, is not the case with some others who have come here and tried
> to tell us how to do our job.
I hate to say this, but the more that I research about what Usefor is
doing, the more convinced I am that Usefor is not doing its job.
Perhaps you should have listened to those people. It sounds like they
knew what they were talking about. I get very suspicious of the
motivations of people who say stuff like "some others who have come here
and tried to tell us how to do our job."
How long has Usefor been in existance?
Why are there no goals and milestones on its charter page?
Why is there a complete duplication of RFC 2822 and the MIME RFCs in the
Usefor document, as opposed to using RFC 2822 and the MIME RFCs as
normative?
> I am unaware of any client that will
> "fail to swallow" 8bit input, though many clients will fail to "digest" it
> in that they will display nonsense characters.
Sigh. Please review the MIME working group minutes and discussions, in
which numerous problems with 8-bit were discussed. Your experience seems
to be extremely limited.
> But clients that interoperate with IMAP usually also have the capability
> to interoperate with POP3, SMTP, NNTP and maybe even UUCP. I have never
> seen any suggestion that those other servers are in any way obligated to
> "fix" things that the client is unable to swallow/digest.
All clients and servers of these protocols are required to comply. 8-bit
headers are non-compliant.
8-bit UTF-8 headers in NNTP will require extension negotiation and
downgrade handling in POP3 servers, IMAP servers, SMTP clients, and SMTP
servers. All of these protocol changes will have to be specified and
implemented before they can be permitted in NNTP or any other protocol.
> No, if it was "valid in" according to a revised standard, then it would be
> "valid out" also.
What about compatibility?
Compatibility with the installed base is a requirement for any proposal to
have a chance of passage. Even the major changes from IMAP2 to IMAP4
necessitated a compatibility mechanism, which is why to this day IMAP2
clients still work with IMAP4rev1 servers, and well-written IMAP4rev1
clients (such as Pine) still work with IMAP2 servers.
You can not change the defintion of "garbage in" into "valid in", unless
there is mutual agreement by both client and server, and mandatory
transform when there is disagreement.
> >I don't *want* to provide a downgrading service.
> Then don't.
You're contradicting yourself. You've just said that you are withdrawing
your unacceptable proposal for UTF-8 headers in news.
> No, Usefor has always been so geographically widespread that there seemed
> little point in arranging face-to-face meetings that few would be able to
> attend.
This is perhaps an explanation for why Usefor is having such difficulties.
This is all the more reason why there must be a meeting in San Francisco,
both to come up with a new charter and to get Usefor on track.
I'll contact the area director.
> But even if I were able to go (which I am not) I have no funding.
I too have to deal with funding issues. But as a document editor, I have
been compelled to attend IETF meetings when my presence was required.
If you are completely unable to attend IETF meetings, then I suggest that
you turn over the document editor function to someone else.
-- Mark --
http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
