[cabfpub] Proposed Ballot 183 (Ballot process)

Ryan Sleevi sleevi at google.com
Mon Jan 9 17:01:41 MST 2017


Responses inline

On Mon, Jan 9, 2017 at 3:14 PM, Virginia Fournier <vfournier at apple.com>
wrote:
>
> VMF:  Per Section 2.2(d), the voting period is exactly 7 calendar days
> (the “Forum shall provide seven calendar-days for voting”).  The review
> period is “at least seven calendar days.”  However, I don’t think we should
> allow expansion of the review period given the many comments we’ve heard
> about speeding up the process.
>

I don't believe the Forum has interpreted this, historically, as you have,
which is why I highlighted it.

Perhaps as part of this Ballot we can find an alternate, or otherwise
specific, way to highlight that the Forum is providing exactly no more and
no less than 7 days.

We've also historically allowed for extensions to the review period,
particularly around the goal of "floating a ballot" to gather feedback.

I don't think we have any objections to setting a fixed 7 day vote and 7
day review, with no more and no less time afforded, but was hoping we can
find some way to clarify that, to avoid the (related) issues.


>    c) Allow both ballots to proceed
>      c.1) How do we determine what the final text is if both ballots
> conclude at the same time?
>      c.2) Should we require that the 'more recently proposed' ballot
> include provisions about any existing ballots?
>
> VMF:  Generally, we should not allow two ballots modifying the same exact
> text to proceed at the same time.  To be clear, what I mean by this is that
> there shouldn’t be two ballots running simultaneously on Section 3.2.2.4.5,
> for example.  However, you could have two ballots running simultaneously on
> Sections 3.2.2.4.5 and 3.2.2.4.6.
>
> If, for some reason, there are two ballots running simultaneously on the
> same section (e.g., Section 3.2.2.4.5) in the alternative, we should
> indicate in the ballots what happens if they both pass (as we did with
> Ballots 180 and 181).
>

Should and Will are two very different things.

For example, one way to resolve this would be introducing a clause in
2.3(c) to explicitly prohibit such overlap. If we do work to prohibit such
overlap, it would also argue that we should make it clearer that both
voting and review cannot be extended beyond the 7 day minimums that the
Forum has historically treated them as.


> 2.3(h) - Is it correct to suggest that "both teleconference and
> face-to-face meetings" be reworded as "either teleconference or
> face-to-face-meetings"? I'd be curious to know if this change presents any
> disagreement, but my understanding is that our Quorum has been taken as the
> union of the past three.
>   a) One possible issue with this Quorum has been the example of the "IPR
> Task Force". As the Bylaws provide no such definition of what a "Task
> Force" is, we must conclude either it's
>      1) a "Working Group" (which requires the production of Ballot to
> charter)
>      2) a regular Forum meeting (meaning that Quorum is defined over two
> weeks - 1 regular meeting, 1 TF meeting, 1 regular meeting - rather than
> the four weeks it takes for 3 regular bi-weekly meetings)
>      3) or a meeting of independent parties outside the CA/Browser Forum
> (which means no IP protections apply).
>  b) Otherwise, we have the possibility of measuring quorum as only those
> that have participated in both teleconferences *and* the past three
> meetings, which does not seem a correct read
>
> VMF:  It appears that this group has been called an "IPR Policy committee”
> before.  https://cabforum.org/2012/11/08/2012-11-08-minutes/  Perhaps we
> are using the wrong nomenclature.  This was not intended to be a WG, but a
> group of concerned members who have been trying to resolve this issue that
> has brought the Forum to a standstill.  There is nothing in the Bylaws that
> prevents such meetings.
>

Right, sub-groups and committees aren't prohibited (and referred to, in
passing, in 5.2(a) and 5.2(b)).

My point was trying to bring clarity as to what constitutes a Forum Meeting
for purposes of determining quorum. I highlighted this because "previous
three meetings" leaves it ambiguous as to whether meetings of WGs are
considered (including that of any sub-groups and committees), as well as
any meetings of sub-groups or committees of the Forum at large. This is
significant because the Bylaws don't speak at all to the formation of
sub-groups or committees, and as such, a 'pessimistic read' could suggest
the following scenario:

1) Ryan and Gerv declare they're forming the "Ryan and Gerv are Cool Folks
Committee"
2) Ryan and Gerv announce to the public@ list a draft agenda for the next
three meetings, which will happen at 2017-01-10 12:00 GMT, 13:00 GMT, and
14:00 GMT, for which the sole agenda items are reading the anti-trust
statement and talking about how cool Ryan and Gerv are.
3) Ryan and Gerv put forward Ballot 185, which substantially changes the
BRs.

For purposes of Quorum, do the "Ryan and Gerv are Cool Folks Committee"
count as "the last three meetings"? If so, then quorum would be defined as
"1 participant".

I don't believe that's what's intended here, and for purposes of quorum,
we've historically measured it as "Those participating in Forum-wide calls
and Forum-wide F2F meetings", with the call being determined by regular
schedule and the F2Fs announced by the Chair well in advance.

Do you see the concern? I highlight it precisely because of the ambiguities
regarding what the IPR call counts as, and whether it counts towards
quorum. This is why I raised the concerns in the past regarding language of
"Task Force" and "Working Group", due to the ambiguities they introduced.



> 2.3(i) - The Forum leaves it ambiguous as to who is directed to complete
> this action, as well as what implications this has, if any, for the
> 'finalization' of the ballot if it's failed to be followed.
>   1) For example, if the Chair is on vacation during the close of the
> voting period, does the Vice-Chair take this action?
>   2) If either the Chair or Vice-Chair fail to take this action, is the
> Ballot still considered valid? Invalid?
>
> VMF:  This section was taken from Section 2.2 of the Bylaws, so this
> language has already been in the Bylaws for some time.  Has such a problem
> occurred?  We could always change it to read “the Chair or his/her
> delegate,” which of course could be the Vice-Chair.
>

Yup. I think that's a fine modification, but I think we still have the
ambiguity as to what the state is for the Ballot if the Chair or the
delegate fails to complete the action.

A particular concern here is the combination of this ambiguity and 2.3(j),
that the "Review Period" is extended beyond the 30-to-60 days in a way that
leaves members ambiguous as to the IP state. In particular, the question is
when does the Review Period formally start, and whether it's keyed off a
calendar event (e.g. 2 business days) or a specific action of a specific
person (the Chair or their delegate).

See below for how I expand on that concern.


> 2.3(j) - This is better in that it directs the Chair to take a particular
> action, but fails to consider what implications, if any, happen if the
> Chair fails to take this action.
> VMF:  When I worked at Sun, Scott McNealy was fond of saying, “We are all
> adults, and we have to trust each other to do our jobs.”  I think that
> applies here.  We’re just not going to be able to cover every possibility
> and eventuality in the Bylaws.  At some point, we have to trust that people
> are going to do the right thing.
>

So far, the Forum is fairly bad at that. Apologies if I'm unable to extend
that trust, since I think the discussions of the IP policy have highlighted
that some members have very different - and significant - interpretations
of what we took as common sense and standard operating procedures.

 1) Assume the Chair is on vacation again. For purposes of IP Review, does
> the Period begin on that second business day, or does it begin when the
> chair returns?
> VMF:  Again, I think this can be fixed by adding “the Chair or his/her
> delegate” for situations when the Chair is on vacation or otherwise
> unavailable.
>
> 2) Assume the Vice-Chair sends out the message. For purposes of the
> Bylaws, does that constitute notice of the IP Review Period?
> VMF:  Yes, if the Chair has delegated his/her authority to the Vice-Chair.
>

The issue I have with this (Combined with the remarks on 2.3(i)) is that it
leaves us in an ambiguous situation if the Chair/Vice-Chair fail to do so.

Imagine a Ballot concludes on January 7 for a maintenance guideline. We
have several possible interpretations of what constitutes the review period:
2017-01-07 -> 2017-02-06 (30 days from conclusion of ballot)
2017-01-08 -> 2017-02-07 (30 days from the scheduled announcement of the
results - noting this begins on a weekend)
2017-01-09 -> 2017-02-08 (30 days from when the Chair sends out the email,
if they sent it w/in 1 day of conclusion)
2017-01-11 -> 2017-02-12 (30 days from the "Within two business days", if
the Chair/Vice-Chair waits until the last minute)
2017-01-14 -> 2017-02-15 (30 days from "When the Chair got back from
vacation and realized they forgot to send the email")

You can understand how understanding, clearly, when the review period ended
is significant - especially given that we've received notifications in the
past on the 'day of' conclusion. My goal is to ensure we have an
unambiguous process, and seemingly the best way to ensure that is by
guaranteeing that state changes happen based on specific calendar events,
rather than specific actions of the chair/vice-chair. To what extent we can
formalize that, I'm open to suggestions. I would argue that for purposes of
Exclusions, we either make it explicit on a calendar event, or bound it in
some way (e.g. "the Review Period shall begin when the Chair sends the
email or within 5 business days of X, whichever occurs sooner")


>  3) We previously had the very unfortunate issue where the Chair was
> employed by the same company as the one submitting the Exclusion Notice,
> and there was some ambiguity with regards to whether the submission was
> made by the deadline. In line with Kirk's previous remarks, do you believe
> there's any path to correct this with the Bylaws, or would you view this as
> necessitating a change to the IPR Policy (specifically, Section 4.3)
>
> VMF:  I’m not sure how it would be fixed.  From my perspective (not
> Apple’s) and without offering any legal advice, it seems that if a member
> doesn’t properly submit an exclusion notice, then it has no effect.  If the
> member ever tries to assert an “excluded” patent in an infringement case,
> they’d have to explain to a judge/jury why they didn’t follow the proper
> process for the exclusion, and why they should be able to assert
> the “excluded” patent after not following the process.  It might be a
> challenge.
>

The previous proposed resolution was to require Exclusion Notices be
submitted to the public at cabforum.org list. This avoids some of the
ambiguity, but seems to materially overlap with the IPR Policy (Section
4.3). My hope was that we might find a way within the bylaws to accomodate
this, without the corresponding IPR Policy change, but it may be that the
latter is unavoidable.

Equally, the 'uncertainty' aspect is one we could also accept - and it
would be a matter for the Courts to evaluate the evidence as to whether the
Exclusion was valid. If that's the direction we want to go, it seems useful
if we can quantify anything that would make that a more concrete, and less
nebulous, matter to ascertain.


>
> 2.3(j) - Again, assume the Chair fails in any of these steps, what do you
> believe happens / should happen?
>   1) Chair fails to distribute Exclusion Notices within two business days
> after the Review Period closes
>   2) Chair distributes an incomplete set of Exclusion Notices
>   3) Chair's MTA represents to the person making the Exclusion Notice that
> mail has been successfully queued for delivery (e.g. the mail server
> reports to the sender that it's accepted)
>      a) Chair encounters a mailbox corruption issue and loses exclusion
> notices
>      b) Chair's antivirus system later quarantines the mail so that it's
> not kept in their inbox, but instead moved to a quarantine
>
> VMF: The Bylaws are going to be as long as War and Peace if we try to
> address everything that could possibly go wrong.  Earthquakes, floods,
> locusts, meteors.  The general rule should be that if the Chair “fails” for
> some reason to send out the Exclusion Notices, it does not affect the
> validity of the Exclusion Notices.  However, if the member does not file
> the Exclusion Notice properly, it would affect the validity of the notice.
> I hope the difference is clear.
>

We're in this situation now precisely because the Bylaws were not adhered
to. I can only imagine how much this has cost all of our respective member
organizations in time and legal fees because of it. My hope is that with an
ounce of prevention, we might gain a pound of cure.

A seemingly easy way to resolve this, as suggested above, is to indicate as
much as possible that these progressions happen on a fixed cadence based on
the calendar, or, in the absence of that, set an upper bound on the time
for the Chair to advance it before it 'automatically' addresses.


> 2.3(k) - For avoiding any ambiguity, historically the interpretation was
> that a Ballot is "in force" when a new versions of the Guideline are made
> public on the website. Do you believe this section supports that
> interpretation?
>
> VMF:  I’m sorry - I don’t understand this one.
>

This relates to Section 4.1 of the IPR policy ("Prior to the approval of")
and the past ambiguities regarding this.

The question is "What event constitutes approval", given that we have a
sequence of events progressing, some of which are gated on human factors.
We can basically pick wherever, but we should have agreement.


>
>   a) One interpretation is that "the Initial Vote is deemed to be final
> and approved" the moment the Review Period closes, but
>     1) What happens if the Chair failed to distribute Exclusion Notices
> per 2.3(j), due to the reasons I mentioned. Is the Ballot's finality keyed
> on 2.3(j) being successful?
>
> VMF:  I have a hard time imagining this happening.  Wouldn’t people start
> asking 2 calendar days after the close of the Review Period if they hadn’t
> seen the Exclusion Notices?  Why would they wait the whole 30 or 60 days to
> ask?
>

>   b) Another interpretation would be that it would be once the Chair sent
> out the mail, regardless of what was published on the public website
>   c) Another interpretation would be that it is only once it's published
> on the public website (e.g. once all of 2.3(k) has been met)
>
> VMF:  Sorry, I just don’t see this happening.
>

As recently as this week, we saw an issue where a members mail bounced.
Similarly, during the previous set of exclusion notices, we saw one such
exclusion notice flagged as spam. It absolutely happens, and it would be
great to think of what steps we can take to mitigate this or otherwise make
it unambiguous as to what should happen.

Currently, the Chair represents a "Single Point of Failure" - if any
exclusion notices are failed to be received/published, we enter in to a
very ambiguous state. For the purposes of avoiding ambiguity, we should
strive to see what 'little tweaks' we can add.


> 2.3(l)(i) - This requires that a PAG to be formed, but there's no
> guarantee that a PAG will be formed, in either the Bylaws or the IPR. That
> is, given that the PAG is voluntary, what actions, if any, happen if there
> are no volunteers for the PAG (for example, due to concerns regarding
> knowing infringement)?
>
> VMF:  I don’t see this as an issue.  Usually the members who have filed
> Exclusion Notices will want to be on the PAG.
>

Usually, but not always.


>  And those who want the ballot to pass would participate in the PAG.
>

That's not at all true.


>  The PAG could also be a 15 minute phone call with members deciding they
> don’t want to do an in-depth patent review - it doesn’t have to be anything
> specific.
>

While I agree that it's unlikely, it's at least useful to know what view
members take as the necessary steps. I'm particularly trying to avoid a
repeat of Ballot 180/181/182, for which two members applied an
interpretation to the Bylaws and threw out four years of practice. I'm not
saying that we didn't need to solve these issues, and I've certainly beat
to death that particular horse, but you can hopefully understand why
spelling out some process to avoid ambiguity is a net-win for the Forum.

Heck, even if we want to punt everything to "Any exceptional situation
shall be resolved by Forum Ballot" would be fine. We've had situations like
this in the past (where the Bylaws were ambiguous), most notably with
respect to the election of Chair/Vice-Chair, and the resolution was "Don't
do anything until we have a ballot to codify process". That's very
different than how 180/181/182 were handled, but arguably a pattern that
would hopefully resolve things to everyone's satisfaction.


>
> 2.3(l)(ii) - How are material changes determined? Is that the discretion
> of the Chair? The discretion of the proposers? The Forum at large?
>
> VMF:  “Material changes” is generally understood to mean changes other
> than correcting errors/typos or making minor editorial changes for clarity.
>

While I understand what it's generally understood as, in the past, the
Chair has proposed material changes as immaterial changes, to which some
members have objected, because it would have meaningfully affected the
interpretation. My hope is we might provide tighter guidance.


> 2.3(l)(ii) - How is it determined that the PAG has provided a
> recommendation? Is it through action of the Chair? Through a formal Ballot
> of consensus?
> VMF:  The PAG will provide a recommendation to the members when it
> concludes its analysis, delivered by the PAG Chair.
>

I don't think this provides much confidence in the face of contention,
which I think we might reasonably expect with as 'simple' a PAG as that for
182.


> 2.3(l)(ii) - What happens if the PAG declines to provide a recommendation
> in a timely fashion? For example, a PAG is convened, ten members are
> initially active, but response tapers and the PAG does not formally
> recommend any action. What occurs to the Ballot? What occurs to any future
> ballots?
> VMF:  It may not be possible for the PAG to provide a “ imely" response.
> It depends on many factors, which are outside the Forum’s control, such as
> negotiating license agreements with third parties, analyzing other
> technology to see if it can be used instead of an excluded item, etc.  If
> members are interested in seeing the Ballot pass, they need to stay
> involved in the PAG so the ballot doesn’t languish.  I’m not sure why there
> would be an affect on future ballots - it’s not FIFO.
>

This ties back in to the earlier question on 2.3(d)


> 2.3(l)(ii)(B) - For purposes of clarity, is it fair to state the the
> Ballot has failed for IPR (as a distinct case from failing for lack of
> quorum or failing due to lack of Yes votes)?
> VMF:  No, it hasn’t failed.  It is being reconsidered in light of the
> PAG’s recommendation.
>

Is there a path you can see to provide any other status other than what you
just said? I would prefer the Ballot be clearly indicated as Pass/Fail,
especially if (and I believe we do) need to take steps to correct the
issues related to 2.3(d).


> 2.3(l)(iii) - It would seem this step would allow the PAG to
> 'indefinitely' keep a ballot hostage, even if it is successfully voted on
> by members. This relates to the ambiguity of recommendation (per the
> questions re: 2.3(l)(ii)), but I'm struggling to find anything that would
> prevent the following: Ballot proposed, Forum votes yes, IP exclusion
> received, PAG formed, only the excluder joins the PAG (thus leaving no
> Chair), excluder (does not make a recommendation in a timely fashion or
> recommends not proceeding). In the case of not making a recommendation, it
> seems the Ballot is "hung" forever. In the case of recommending not
> proceeding, it seems that the cycle can continue indefinitely, until
> additional members join the PAG (so as to recommend some other action). In
> that case, we potentially have a 'hung PAG' (imagine 2 members with 50/50
> split), and this is why the questions about the nature of a PAG are
> relevant.
>
> VMF:  If only the excluder is on the PAG, this tells me that the ballot
> isn’t important to anyone else.
>

That's a false equivalency. You're trying to measure importance by how
willing members are to take on legal risk by engaging in PAGs. It's not
fair to assume that, say, if a member is prohibited from joining a PAG due
to legal risk, that the resolution of the matter is not important or
significant to them.


>  If it’s only important to the excluder, then it might not matter to the
> rest of the Forum if it’s “hung” forever.  If there are others who are
> interested in seeing the ballot pass, then they should get involved in the
> PAG to help it move forward.  Members should not get to sit back and reap
> the rewards without participating/contributing.
>

As stated above, that's an incredibly uncharitable view towards members'
ability to participate in PAGs.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cabforum.org/pipermail/public/attachments/20170109/2193eed9/attachment-0001.html>


More information about the Public mailing list