ietf
[Top] [All Lists]

Re: Mail sent to midcom (fwd)

2001-02-01 16:40:02

I believe most IETF WG mailing lists restrict automatic posting to
those subscribed and a list of other from addresses. As a practical
matter, in this age of spam, that is considered "open" and, if not in
place, is commonly demanded by a consensus of the WG.  Every WG is a
little different and I believe these sorts of policies should be
determined on a WG by WG basis by the WG chair, the WG consensus, and
the AD.

From:  Lloyd Wood <l(_dot_)wood(_at_)eim(_dot_)surrey(_dot_)ac(_dot_)uk>
Date:  Thu, 1 Feb 2001 21:48:13 +0000 (GMT)
Reply-To:  L(_dot_)Wood(_at_)eim(_dot_)surrey(_dot_)ac(_dot_)uk
To:  James M Galvin <galvin(_at_)acm(_dot_)org>
cc:  ietf(_at_)ietf(_dot_)org, midcom-admin(_at_)ietf(_dot_)org, 
poised(_at_)lists(_dot_)tislabs(_dot_)com,
            Scott Bradner <sob(_at_)harvard(_dot_)edu>, Allison Mankin 
<mankin(_at_)east(_dot_)isi(_dot_)edu>
In-Reply-To:  <0G8300D6ZJ4GO7(_at_)eListX(_dot_)com>
Message-ID:  
<Pine(_dot_)GSO(_dot_)4(_dot_)21(_dot_)0102012105100(_dot_)16206-100000(_at_)regan(_dot_)ee(_dot_)surrey(_dot_)ac(_dot_)uk>

On Thu, 1 Feb 2001, James M Galvin wrote:

Although it is true that RFC2418 does not explicitly permit the "review"
of messages submitted to elists from non-subscribers,

correct.

It doesn't explicitly prohibit it either.

it is in fact an accepted practice on IETF elists.

(or elitists, as they perhaps should be known?)

I like to request a public list of those lists. It would be
interesting to spot any patterns in the data.

I don't think there is and don't see why there would be a centralized
list of specific WG mailing list policies.  Both the TRADE and XMLDSIG
WG's, which I chair and co-chair respectively, block most
non-subscriber posting.  I consider this to be consistent with RFC
2418 and in both cases the policies were installed early in the WG
existence after numerous complaints by WG members about spam.

So much so that the IESG has published a statement regarding the
policy and procedures of such practices:

    http://www.ietf.org/IESG/STATEMENTS/moderated-lists.txt

Note the use in that statement of the terms 'might be needed',
'persistent' and 'excessive'. midcom's list policy clearly does not
fall within that scope (a single email from me to midcom is neither
persistent nor excessive).

I interpret moderation to be the human review and approval of all
messages that get to the list.  Any WG that permits unrestricted
posting after you subscribe is not moderated.

Speaking for myself, I wish that all IETF elists could and would adopt
the practice of reviewing all non-subscriber submissions for at least
obvious irrelevance.

The IESG statement does not cover that. RFC2418 does not cover that.

Just to be clear, I'm making a distinction between moderation and review
to reject obvious irrelevance. 

irrelevance and its obviousness is in the eye of the beholder. I would
be extremely worried if messages were suppressed because e.g. a WG
chair had deemed a matter to be settled and would not permit further
discussion because further discussion is irrelevant (thinking of e.g.
recent heated EF technical debate on diffserv).

As long as WG chairs are trusted to determine WG consensus, I don't
see why they can't determine if a message is obviously irrelevant to
the tasks for which a WG was created.  I'm not claiming a WG chair
would never make a mistake but their decision are subject to
review. Anyway since those who subscribe can post whatever they want,
judgement usually doesn't enter the picture.

That's where this is heading.

In that context, I agree with you that the phrasing in the
notification message you received could be improved, but I think
it's an unfair leap from "reviewing messages" to "midcom is not
open" without even asking what the actual policy and practice is
and confirming whether or not the AD and IESG are aware of it.

Just to be clear: midcom's policy falls outside stated IESG policy,
and we've agreed that it is not permitted by RFC2418.

As far as I can see, midcom's policy is unrelated to the IESG policy,
which covers manual moderation of all posting.  Nor do I see any
agreement that it is not permitted by RFC 2418.  It depends on what
"open" means.  The definition of words changes with circumstances.  In
the country or the early Internet, an open house might have no lock or
guard and a host might, as many did, have a "guest" account with no
password.  In the middle of Manhattan or the modern Internet, open
means you have locks and guards and you usually check out people who
haven't gone through a procedure to get on an access list.

midcom is reviewing messages, based on email address and then
content. midcom is not open. There's a direct causal relationship
between those two statements.

I believe your definition of open differs from mine.

Melinda's
note makes it clear, at least to me, that the policy is consistent with
the spirit of RFC2418 

see your first sentence...

In which he said that such a policy was not explicitly permitted in
2418 which is not at all saying that it is prohibited.

and the IESG statement indicated above.

which it isn't.

Speaking as Co-Chair of this working group, unless you have a specific
request for a change to RFC2418 or the IESG statement,

I'd like a statement that RFC2418 will be adhered to by mailing lists.

In my opinion, it is being adhered to. In any case, it is not the job
of any particular WG, even POISSON, to be a policeman over other WGs.

I don't see any basis for continued discussion of this point on
the poised elist.

so future emails will be filtered by the moderator?

The poised mailing list is not moderated.

If you object to how the midcom elist is operating you need to take that
up with the midcom-admin and the relevant AD.

done. on cc. On open IETF lists, I have the right to post what you
deem to be rubbish, and you have the right to choose to ignore me (and

No, you do not have an unlimited right to post what Jim Galvin
considers rubbish, for certain definitions of rubbish, on the poised
mailing list, since he is a co-chair of the working group.  Nor do you
have any unlimited right to post what Fred Baker, as IETF chair,
considers rubblish, for certain definitions of rubbish, on the IETF
list. If these facts mean that the poised and ietf lists don't meet
your definition of open, so be it.

the satisfaction of doing so). midcom's policy limits those rights a
priori without consensus or even persistent complaints from list
members.

regards,

L.

www.isoc.org: 'ISOC is active in areas such as censorship'. No kidding.

Jim
co-Chair of the POISSON Working Group

<L(_dot_)Wood(_at_)surrey(_dot_)ac(_dot_)uk>PGP<http://www.ee.surrey.ac.uk/Personal/L.Wood/>

Donald



<Prev in Thread] Current Thread [Next in Thread>