Re: Comments on draft-roach-bis-documents-00

[Stephen Farrell <[e-mail]>]

Hiya,

On 09/05/2019 12:49, Joel M. Halpern wrote:
> A big part of the problem I see, which I see addressed by this draft, is
> that as currently constituted we can not fix many existing protocols
> without doing a major modification to address security and / or privacy
> issues.

I guess we may have different definitions of "fix":-)

I totally agree that it's not feasible to address all
legacy security and privacy issues in all bis documents.
I don't agree that bis documents ought get a free pass
wrt such issues where they exist.

> 
> Hence the bit of text being discussed about not having DISCUSS or
> Abstain on the basis of not fixing existing security and privacy issues
> is important.
> 
> ADs are free to comment, and every WG I have seen takes such comments
> seriously.  So if the AD thinks there is a simple improvement in
> security and / or privacy that ought to be included, they can ask about it.

In my little head, that is what most security/privacy DISCUSS
ballots on bis drafts tend to be. Adam's draft seems to me to
call for ADs to not ask/chat/discuss such issues, and I don't
agree with that level of non-discussion being the expectation.

Cheers,
S.

> 
> Yours,
> Joel
> 
> On 5/9/19 6:28 AM, Stephen Farrell wrote:
>>
>> Hiya,
>>
>> I've read the draft and like it except for the bit
>> SM quotes below.
>>
>> On 09/05/2019 09:34, S Moonesamy wrote:
>>> Hi Adam,
>>>
>>> There is the following sentence in Section 3.2 of
>>> draft-roach-bis-documents-00:
>>>
>>>  Â Â  "IESG members SHOULD NOT issue DISCUSS or ABSTAIN ballot positions
>>>  Â Â Â  based on unchanged text except as described in Section 3.3."
>>
>> I'm fine with the idea that the IESG would mostly just
>> review the diff, and the IESG do need to respect the
>> fact that existing RFC text has IETF consensus, but I
>> don't think it's ok to try to force ADs to ignore security
>> or privacy issues that the proponents of a bis would
>> like to ignore. I read the text above as doing that. I'm
>> not saying that all such things ought always be fixed in
>> bis drafts as we clearly do not do that, but a SHOULD NOT
>> DISCUSS seems wrong.
>>
>> Separately, if enough ADs ABSTAIN then the draft should
>> have a problem. ABSTAIN ballots weren't that common when
>> I was on the IESG so unless that's changed a lot I don't
>> think that clause is useful or advisable.
>>
>> So if that text stays in, I would hope that ADs would
>> ignore it and try do what they consider correct. That may
>> be another argument to not have a SHOULD NOT - just say
>> that the goal is to keep reviews to the diff.
>>
>> Lastly, I'd leave out that text because even if it were
>> what the IESG wanted, it ought be in the discuss-criteria
>> IESG statement and not in an RFC/BCP that derives from
>> this draft. (It is good that it's there now, so I can
>> whine about it though:-)
>>
>> Cheers,
>> S.
>>
>>
>>>
>>> Why is an ABSTAIN an issue?
>>>
>>> What about IESG member "comments"?  Can those comments be ignored?
>>>
>>> Regards,
>>> S. Moonesamy
>>>
>>>
>