--On Thursday, September 13, 2012 15:10 -0700 Dave Crocker
<dcrocker@xxxxxxxx> wrote:
> On 9/13/2012 3:08 PM, Barry Leiba wrote:
>>>>> The IETF Chair may decide to removed an I-D from the
>>>>> public I-D archive.
>>>
>>> This defines the IETF Chair as Chief Censor, with no written
>>> policy guidance. That is, deletion is at the whimsy of the
>>> Chair.
>>>
>>> Is that really what we (and the Chair) want?
>>
>> I very much agree. I'm happy with the decision being the
>> consensus of a board, but not giving it to an individual.
>
>
> Not that you've said something to exclude this, but I want to
> make sure we don't lose the second part:
>
> I believe we /do/ need a written policy that has been
> reviewed by legal counsel. Even with a group -- versus
> individual -- we should not create possible charges of
> censorship up to the personal whims of the moment.
I'm not sure, Dave. I see the theoretical risk, but the number
of times that problem has occurred in the past is, I believe,
zero. I'd be more concerned about it if we were talking about
removing active documents rather than taking something out of
the publicly-available archive of expired documents. And a
decision by the IESG to remove a document would presumably be,
like all other IESG decisions, subject to appeal.
If the IESG started taking down documents just because they
didn't like them, I'd also think we'd quickly try out the recall
procedure for the first time. But, again, the IESG has had many
years of experience with I-Ds they don't like (and many such
documents), during most of which time no one questioned their
ability to remove a document from public view (active or
archived) and, AFAIK, we have seen absolutely no abuse of that
type.
I'd be ok with a "written policy that has been reviewed by legal
counsel" if I thought there were even reasonable odds of its
adequately covering all the cases. But I fear that such a
policy would exclude some important case and thereby get the
IESG in trouble as we tried to work with (or work around, or
revise on an emergency basis) a policy that turned out to
prevent us (or even the IESG or IETF Trust) from exercising good
sense and doing what is either obviously right or as advised by
Counsel when Counsel is presented by a particular situation.
Proposal: I think it would be fine to have a policy in which:
(1) A document would be taken down as quickly as
possible in response to a legitimate court order with
approval of Counsel.
(2) A document to which, in the opinion of Counsel, the
IETF might not have clear copyright title, and for which
a DCMA take-down notice or equivalent was received
unless the IESG, with advice and consent of Counsel,
advised against such a takedown.
(3) The IESG could take anything else down for cause
at its discretion after:
(3.1) Seeking and receiving advice of Counsel
(3.2) Announcing intention to do so, and the cause,
on the IETF-announce list and allowing a brief
period for objections.
(3.3) Considering any advice and objections before
making a final decision.
(3.4) Subject to the above, the IESG is encouraged
to look with favor on requests to remove already-
expired documents from the public archive from the
party responsible for the document (authors of
individual drafts; WGs for WG drafts). We
recognize that such removal doesn't cause the
documents to disappear from the Internet, but
responsible parties making such requests about
their own documents presumably have reasons that
we should treat respectfully.
(4) Any of those actions would be subject to appeal if
members of the community felt there were important
issues that had not been considered. An appeal could
request that the advice from Counsel be made available
to the community. The IESG would be required to either
comply with that request or explain why not.
I don't like it --my personal position is much closer to Joe's
or Sam's-- but, if there is consensus that we have to have a
written policy, one like the above at least doesn't tie our
hands when a situation arises that we haven't anticipated.
While there is some possibility of DoS attacks embedded in it,
it remains important that we haven't seen huge numbers of these
requests (or even small numbers). Indeed, one of my fears about
the amount of discussion we've had (and any formal policy at
this point) is that it may encourage such requests. Too late to
worry about that now.
best,
john