RE: questions on WG Letter Ballot
Roger,
I agree with Howard's responses. I would like to add some addtional
information on a couple and answers to the two he left to WG chairs.
> (3) During recirculation, is it customary (or permissible, or
> advisable) to allow voting by members who failed to participate in
> the first round? This seems problematic for several reasons.
As Howard says, the recirc is open to the original ballot pool whether or
not the person voted on the initial ballot. However, it is important that
the recirculation letter specifically states the subject of the recirc which
is technical changes to the draft as a result of the comment resolution and
any unresolved negatives. Normally a summary list of the technical changes
is included in the ballot package. A voter who didn't respond the first time
around shouldn't be using the recirc to comment on the original material
(and of course voters who voted the first time shouldn't be submitting new
comments on the original material). Otherwise people could use the recirc to
infinitely delay approval of a draft.
Of course, one will sometimes get recirc comments that are outside the
subject of the recirc. Perhaps someone honestly found a problem that they
missed until after the original ballot closed. In that case, the Working
Group will have to decide whether the problem is so severe as to merit a
delay while it is fixed. This is a bit easier decision on WG ballot then on
Sponsor ballot because one can agrre to submit it as a Sponsor ballot
comment and resolve it then.
>
> (4) Would such votes be countable for the purposes of computing the
> return rate?
One doesn't conduct a recirc unless the initial ballot closed so it doesn't
matter to the return rate, but any additional ballots are normally reported
as part of the ballot tally after the recirc.
>
> (5) The 802 Rules say that WG 'Membership may be lost if two of the
> last three Working Group letter ballots are not returned, or are
> returned with an abstention other than "lack of technical expertise."
> This rule may be excused by the Working Group Chair if the individual
> is otherwise an active participant.' In deciding how to apply this
> rule, I'd appreciate hearing the policies and experiences of other
> Working Groups.
Generally one should be very strict in applying this rule because failure to
return ballots can significantly delay the work of the WG. This particularly
applies to those who do not return their ballots at all. When I applied the
rule, I only counted ballots not returned against voting status. Abstains I
counted as a returned ballot regardless of reason so that was the one area
where I allowed some leeway to the voters. Frankly, I would rather someone
abstain than sending in a Yes vote without reading the draft so I do not
like the abstention part of the rule. Other than that, I never excused the
rule. I cannot recall anyone asking me to excuse it. It would take a pretty
strong excuse like "I was in a coma" for me to excuse this rule.
If one is strict in applying the rule and does some hound dogging, one will
find that people rarely fail to return two out of three ballots other than a
few habitual offenders.
>
> (6) If (as in (3)) someone skips the original ballot but is allowed
> to vote in the recirc, would you count them as having returned the
> letter ballot for the purposes of membership retention (5)?
>
I don't recall any cases where someone failed to respond on two initial
ballots but responded on one or both of the recirculations. My inclination
would be to give credit based on whether the ballot was returned by the
initial ballot close as non-return of ballots causes significant problems
for the working group if ballots don't close.
-----Original Message-----
From: Howard Frazier [mailto:millardo@dominetsystems.com]
Sent: Friday, December 29, 2000 6:21 PM
Cc: stds-802-sec@ieee.org
Subject: Re: questions on WG Letter Ballot
Roger,
I'll answer what I can. In some cases, I will go beyond the requirements of
a WG ballot and talk about subjects that are of importance to RevCom. My
general advice is that you run your WG ballots just like sponsor ballots.
It's good practice, and it offers the least chance for screw-ups or
questions.
1) Tailor the duration of the recirc to the size of the material being
examined.
10 days is a good minimum, as long as it applies to the amount of time that
the voters actually get to look at the material. I think it is reasonable
to add a day or two of turn around time for email. If you are recirculating
a
small amount of material (under 100 pages) then 10 days is a reasonable
amount
of time. If you are recirculating a large amount of material, then increase
the duration of the recirc proportionately.
The most important thing you can do on a recirc is to CLEARLY IDENTIFY THE
MATERIAL THAT HAS CHANGED SINCE THE INITIAL BALLOT. Include a table which
summarizes the changes, including subclause, page, and line numbers. Then,
highlight the changes using strike through and underscore. Note that color,
while attractive on a screen, usually looks lousy on a printed copy. So,
I advise against using color to highlight changes.
Clearly written cover letters are a vital part of the ballot package.
Be clear, be explicit be complete. Put a date on all correspondence with
balloters, your WG, and RevCom.
2) 50% is 50%.
3) The voting pool cannot change during a ballot (even in the event of the
death of one of the balloters). Therefore, all of the people who were
eligible to vote at the time a ballot was issued are eligible to vote
on a recirculation.
4) Since you have to achieve the desired return rate on the initial ballot,
it shouldn't really matter if people vote for the first time on a
recirculation ballot. However, you would count such ballots as part of
the return rate. Note that once you have achieved the required return
rate on the initial ballot, it is not necessary to achieve a 50% return
rate on recirculation ballots. On a recirc, the voter's previous ballot
stands unless they respond on the recirc.
5) I'll let WG chairs respond to this one.
6) I'll let WG chairs respond to this one.
Some things you didn't ask, but I'll offer anyway.
A) Get a group of "hound dogs" lined up. These are people who will follow
up with voters to get them to turn their comments and ballots in as early
as possible. Don't rely on email alone for this process. The personal
touch, in the form of a phone call, is very important. By all means,
send reminders via email, but if you want your ballot to close, and
(most important) you want to get the comments in as early as possible,
a phone call will help.
B) What if you don't get the required return rate by the close of ballot?
Well, this is not good, but not fatal. If you miss by a couple of ballots,
then get on the phone and pester the voters who have not returned a ballot.
If you miss by a mile (which actually happens some times) then you need
to get your hound dogs in high gear. I believe that you are allowed up
to sixty days after the close of ballot to try to achieve the required
rate of return.
C) Get the comments in early, and start working on them early. Get a team
set up to handle the comments as they come in, and produce responses
and get feedback from the balloters. You should get your WG to approve
official responses, and you should tell the balloters that an early draft
of a response from an editor or "Ballot Resolution Committee" is subject
to change, but you will find that the more you work with the balloters to
develop a reasonable response, the easier it will be to resolve their
comments.
D) Consensus does not require unanimous approval! Don't bend over backwards
to accomodate some one who is obviously at odds with the consensus of the
group. You can go to RevCom with unresolved negative comments, as long
as the WG has responded to the comment and recirculated it.
E) It is a very good idea to diligently track the status of comments from
negative balloters. If the WG and the balloter reach agreement on the
resolution of a comment, get the WG to vote to that effect, and get the
balloter to sign a piece of paper (or send an email) acknowledging that
the comment has been resolved. Then, once all of their comments have
been resolved, get the balloter to sign a piece of paper (or send an email)
stating that the ballot should be changed to "affirmative with comments".
In other words, don't wait for and don't count on the recirc process.
People forget to send in recirc ballots all the time. RevCom will accept
email from the balloters, or signed pieces of paper.
F) Do everything you can to explain the balloting, comment, response, and
resolution process to your editors, your WG and the balloters. Some of
them will be old pros, and some will be total novices. During 802.3z
I produced a FAQ that helped guide people through the maze. The FAQ is
available at:
http://www.ieee802.org/3/z/savefaq.html
I make no warranties or guarantees that this FAQ is up to date, or even
accurate, but hey! it's free, and it's already written!
Howard
"Roger B. Marks" wrote:
>
> Folks,
>
> I have some questions on the Working Group Letter Ballot rules, both
> written and unwritten. I'd appreciate your guidance:
>
> (1) The 802 Rules don't specify a recirculation duration. Is there
> any objection to 10 days (all-electronic)?
>
> (2) The 802 Rules specify "at least 50 percent of the members
> voting." Does the SEC in practice have a stricter demand?
>
> (3) During recirculation, is it customary (or permissible, or
> advisable) to allow voting by members who failed to participate in
> the first round? This seems problematic for several reasons.
>
> (4) Would such votes be countable for the purposes of computing the
> return rate?
>
> (5) The 802 Rules say that WG 'Membership may be lost if two of the
> last three Working Group letter ballots are not returned, or are
> returned with an abstention other than "lack of technical expertise."
> This rule may be excused by the Working Group Chair if the individual
> is otherwise an active participant.' In deciding how to apply this
> rule, I'd appreciate hearing the policies and experiences of other
> Working Groups.
>
> (6) If (as in (3)) someone skips the original ballot but is allowed
> to vote in the recirc, would you count them as having returned the
> letter ballot for the purposes of membership retention (5)?
>
> Thanks for your help.
>
> Happy New Year!
>
> Roger