[cabfpub] [EXTERNAL]Re: Ballot XXX: Update Discussion Period
sleevi at google.com
Mon Dec 11 16:46:05 UTC 2017
On Mon, Dec 11, 2017 at 11:37 AM, Dimitris Zacharopoulos <jimmy at it.auth.gr>
> Perhaps I misunderstood Kirk's original intent so please correct me if I'm
I certainly understood Kirk's proposal to be about the discussion period,
since it's specifically requested to be included in a ballot about the
> IMO the "editorial changes" proposal is independent to ballots being
> discussed or voted on. The proposal is that at any time (*not during an
> official ballot discussion or voting period*), if someone detects a typo
> or an incorrect reference in a CA/B Forum document that fits the definition
> of an "editorial change" (as noted in the W3C Process Document provided by
> Virginia), that member may just send an e-mail to the public list saying
> what the problem is and a recommended correction, and that this is an
> "editorial change".
> Examples that would probably qualify as "editorial changes" from past
> 1. "certificaet" to "certificate"
> 2. The name of the document "baseline requirements"
> 3. Incorrect references when the BRs were converted to RFC3647 format
> Then, there are two possible routes:
> 1. If there is an objection made by a member, the change is no longer
> considered "editorial" and it has to go through a separate ballot process
> (changes proposed by a member, endorsed by two others, discussed and voted
> 2. If there is no objection, it will be included in the text of an
> upcoming ballot (whatever ballot that is), and this particular change will
> be marked in the introduction section of the ballot as "editorial change"
> or "errata", or whatever. It will be discussed and voted along with the
> rest of the ballot language. Then, two possible outcomes:
> 1. The ballot passes, and so are the "editorial changes"
> 2. The ballot fails, so the "editorial changes" will have to wait
> for a next ballot
> These "editorial changes" will always have to catch a "ballot train" in
> order to be formally accepted. They can be introduced by any member when
> there is no formal ballot discussion or voting period.
I don't believe that was the proposal at all (and certainly not in line
with some of the other changes), but it also introduces its own problem.
For example, if someone proposes that "X" is an editorial change the day
before a ballot, does that mean no objections have been received? If it
rides the train to the next ballot, does that mean if an objection is
raised during that ballot, it gets removed from the ballot? What happens if
both an editorial change and a ballot affect the same area?
> "Worst case scenario" I can think of:
> 1. The forum is discussing about a new ballot and the formal
> discussion period starts at day X
> 2. A member introduces an "editorial change" *one day* before day X.
> 3. The official discussion period for the new ballot begins, including
> the text with the "editorial changes" at day X
> 4. Members have 7 days of official discussion to object to the
> "editorial changes", in which case the ballot author and endorsers will
> either remove these changes before the voting period begins or let them be
> and risk the ballot failing.
> This was my point. In order to facilitate the 'objection' phase, we
rapidly converge upon a hijacked voting process - in which we allow 7 days
for review/objection, and/or then 7 days for voting.
> I don't mind working on a separate ballot for this and let Gerv's ballot
> go ahead.
Yes. This seems a most productive outcome is to separate these two.
> Would people support this?
> Do you see any other risks in this process?
A number of risks, still unaddressed. I think on a more fundamental level,
we need to ask ourselves what the cost of the current process is, relative
to when we've had to exercise it. By my tracking, it's only been excercised
once - all other bits turned out to not be editorial in nature. I'm curious
if other folks have a different perspective and can highlight different
ballots that achieved this, or highlight anything editorial that remains
unaddressed due to it being 'too much overhead'. I think the complexity and
harm for the process, in light of those two, greatly overshadows any
potential benefit, and the process challenges alone are significant.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Public