[cabfpub] Proposed Ballot 183 (Ballot process)

Gervase Markham gerv at mozilla.org
Tue Jan 10 03:43:48 MST 2017

Hi Virginia and all,

Thank you so much for your hard work on this. The diagram is excellent -
please don't apologise that it wasn't made in software! It really helps
bring clarity. And the process you have outlined seems the right one in
principle, so I am confident we can work out the textual details in
fairly short order.

However, "fairly short order" may not mean "immediately"; I think we
should not be considering moving this ballot to a vote until discussion
has died down a little bit. That may only be a few days, and I hope it
won't be a long time, but I think participants should have the space to
discuss scenarios and do their analysis without feeling under a
deadline. After all, we don't want to be back here again any time soon,
so it's vital to get this as right as we can.

On 09/01/17 18:44, Virginia Fournier via Public wrote:
> Please see my responses to your comments below.  I’ve also attached an
> updated draft with redlines to show changes from the last draft.  The
> diagram is also included for those who may not have seen it.

1) One thing which surprises me is that the ballot contains no edits at
all to the text of the Bylaws outside the new section 2.3. Are there
really no other parts which need removing, updating or clarifying in
order to fix any of the issues which have been raised? As a simple
example, the new 2.3 says "This section applies to any ballot that
proposes a Final Guideline or a Final Maintenance Guideline". However,
2.2 says "Ballots will be conducted in accordance with the following
rules." That could lead to an interpretation that if the ballot proposes
an FG or FMG, both sections apply. It might make sense to update the
opening sentence of 2.2 to make it clear that it only applies to other
votes. In fact, it might make sense to have the FG/FMG section before
the section on other votes.

2) A common principle in computer science is DRY - Don't Repeat
Yourself. Having something in one canonical place reduces errors when
making updates. It is wise for us to repeat lots of common text? Or
would it be better to have one section for common rules, and then two
other sections for rules specific to each type of ballot? That way, the
common stuff is only in one place, following DRY.

One already-existing example of this problem is that 2.2 calls the
pre-voting period a "review period", and 2.3 calls it a "discussion
period". The "review period" language has an unfortunate name clash with
the IPR review period. We could fix this, but perhaps it demonstrates
that having two sections saying nearly the same thing is an unwise idea?

3) Section g) talks about two categories, "CAs" and "product suppliers",
but then later in the para talks about the "browser" category. The
category should have a consistent name. This error is also in the
existing section 2.2. Because we are repeating ourselves, we now have to
fix the problem in both places, or leave sub-optimal text in the
newly-added text.

4) At the moment, if the PAG recommends not proceeding, and the proposer
and endorsers wish to proceed anyway, they must nevertheless start the
process again. It seems to me that in a Patent Advisory Group, the clue
is in the name - it's Advisory to the membership, not binding.
Therefore, I think the decision to proceed with the ballot should be
that of the proposer and endorsers alone, and everyone else can vote
taking into account the PAG recommendation that will be available to
them. So I suggest removing the new section l) iii).

Reading the emails, I believe I am disagreeing with Kirk on this point :-)

5) How is it decided, and who decides, whether "material" changes have
been made to the ballot? Would it be easier to say "no changes"? If a
ballot has (non-material) drafting errors, the period of time between an
Initial Vote and a Second Vote is not the time to be fixing it.

6) d) says the discussion period is "at least seven days", but e) calls
it "the seven-day discussion period". Historically, the length of the
discussion period has been set by the drafters of the motion, although
it is normally seven days. I suggest removing the words "seven-day" in
e), and letting the definition in d) stand.


More information about the Public mailing list