[steering-discuss] Community bylaws

Charles-H. Schulz wrote (14-11-10 13:49)

I have a question about the Board of Directors. I see that the draft
says all members are chosen every year (from the Community Members)
and that one can be member only twice consecutively.
Has been considered that this leads to a situation where each year
people have to get used to the tasks, the other board members etc. so
that maybe it is a bit inefficient?

Well, that is a good question. My personal take was at first for a 2
years mandate. Then some others thought that 6 months would be good. I
sliced the apple into two :slight_smile:

I have seen many council and board tables, but never met rules with less then three years for membership.
Of course nowadays all and everything (and especially we) is much faster, but IMO two years is the bare minimum to prevent spoiling lots of time on learning each other and the work so often, and to reach some continuity and stability.

In line with this, I would propose split elections: Appr. 50% of the seats each year.

Best,
Cor

Volker Merschmann wrote (16-11-10 22:14)

I have a question about the Board of Directors. I see that the draft says
all members are chosen every year (from the Community Members) and that one
can be member only twice consecutively.
Has been considered that this leads to a situation where each year people
have to get used to the tasks, the other board members etc. so that maybe it
is a bit inefficient?

In this point I'm with Cor.
The duties of the BOD as written in
http://wiki.documentfoundation.org/CommunityBylaws#Board_of_Directors
are of high importance for the development of the whole project and
therefore long-term-orientated. They will be the premium contact for
other projects, authorities, commercial contributors etc.
I would recommend to drop the re-election limit in favor for a
continous work of the BOD.

A limit to re-election is good to get fresh blood and to prevent people becoming part of the furniture.
And two years twice makes four. So that is pretty OK to benefit from learning a job, the contacts etc. IMO,

Cor

Hi ,

Thorsten Behrens wrote (17-11-10 16:05)

Charles-H. Schulz wrote:

possible solution: sponsor's paid developers shouldn't participate in
ESC as *single* persons. Contributions in their free time and
following cooptation in ESC should be carefully evaluated from BoD or
Membership Committee.

Hmm, I think it's a bit unfair to segregate corporate developers. On
the other hand, what I can propose would not apply to the ESC but to
the BoD: we can put a limit to members who are employees by the same
corporations to three persons max.

Hi Charles, all,

yes, that would be good I guess. The exact number is I guess not too
important, maybe have verbiage like "not more than 40% of the board"

I think max. 25%, with a maximum of 3 persons, is fine.

or something in place. It may also be helpful to further specify
affiliation with a corporation - e.g. there may be other ways than
employment, that influences decisions, like vested interest,
significant business relationships / partnership etc.

(see e.g. Gnome foundation bylaws, article VIII, section 2)

+1

Regarding the valid concerns about limiting director terms vs.
keeping capable people / having a working board most of the time:
what about staggering votes, like only elect 1/2 of the board each
time? That way, existing board members can onboard newcomers, and
there's no interruption / hand-over times. Other than that, I *do*
believe limiting the time a person can serve is beneficial, since it
makes sure fresh blood comes in.

Regarding membership status, I kinda favour the opt-in approach -
i.e. members that want to stay *have* to re-apply each year. Of
course, re-application will be a formality most of the time (thus
hopefully not contributing much to the load of the membership
committee), but it will automagically limit members to people with
active interest (and active email addresses ;)).

I agree with the ideas, and think my proposal (50% elected each years, max. two times consecutively) has the same advantages.

Regards,
Cor

Cor,

Volker Merschmann wrote (16-11-10 22:14)

I have a question about the Board of Directors. I see that the draft says

all members are chosen every year (from the Community Members) and that
one
can be member only twice consecutively.
Has been considered that this leads to a situation where each year people
have to get used to the tasks, the other board members etc. so that maybe
it
is a bit inefficient?

In this point I'm with Cor.

The duties of the BOD as written in
http://wiki.documentfoundation.org/CommunityBylaws#Board_of_Directors
are of high importance for the development of the whole project and
therefore long-term-orientated. They will be the premium contact for
other projects, authorities, commercial contributors etc.
I would recommend to drop the re-election limit in favor for a
continous work of the BOD.

A limit to re-election is good to get fresh blood and to prevent people
becoming part of the furniture.
And two years twice makes four. So that is pretty OK to benefit from
learning a job, the contacts etc. IMO,

I fully second both your suggestion (the one above + the renewal at 50% each
year).

Best,
Charles.

Hi Gianluca,

This is to say: both BoD and ESC with a fixed number of members should
be elected from "members" and "members-developers"

  That might work; its not a terrible idea. Then again - I am not aware
of any other project where the maintainers are elected :wink:

So corporations and governments would have a direct role in both the
political and technical bodies without a predominance of anybody.

  I -really- think that an informal approach, where large contributors
are invited to join the ESC (don't like the name frankly), and we have a
healthy mix of volunteers works best.

  Clearly the elected representatives can replace the ESC if it goes bad
- a useful check on their influence.

Let's say corporation X and government Y gain enough members to control
the ESC. Economically speaking, they can form a cartel and exclude any

  Lol :wink:

No enlargement of the ESC would prevent such situation, because no
enlargement would be permitted at all by the dominant members.

..

Just like Oracle with OOo.

  It is possible in theory. Of course, elections are not without their
potential problems too: creating division, disgruntled loosers, and
campaigning over political points in the (often fragile) volunteer
developer community

However, the ESC isn't simply a technical meeting place. In a software
project, devs do *the* work, so they have the real power.

  Sure - so we need to make the ESC reflect those doing the actual work,
but this is non-controversial IMHO. Even Oracle managed to do this
reasonably well: we had IBM, Novell, RedHat, Canonical etc. there - the
problem there was not a lack of representation.

And ESC, IMO, with its unknown number of members and cooptation, is
more likely open to external and uncontrolled bid for power.

  As/when the ESC turns bad; it should be easy to see. Then we can get it
kicked out and re-formed by the board.

  This is how GNOME works; thus far the 'release team' has not gone bad,
and plenty of people have come and gone through the team.

Good suggestions, though I'd like a fixed number of ESC members too.

  On the contrary; I would like it flexible; so if RedFlag joins, or IBM
joins, we can immediately offer them permanant representation (eg.)
rather than having to pick who to kick off :wink: or waiting for another
election cycle.

  Meritocracy is great, but a quiet, relational process works rather well
too IMHO, and having a small group of people who can actually decide
things and work together effectively is really useful.

  ATB,

    Michael.

Hi,

May I ask here then, the term 'release team' is used above and this is a
question that has been on my mind for a few days.

I suppose this is a question specifically for Michael, but not
exclusively so.

The ESC, do you see this as a very active group, for instance working as
the release team, meeting often and looking at individual issues?

Thanks much,

Drew

Hi Drew,

May I ask here then, the term 'release team' is used above and this is a
question that has been on my mind for a few days.

  Ah - the GNOME 'release team' is about as close as GNOME gets to having
a central body that stewards direction: mostly by deciding what new
modules to include into GNOME. LibreOffice is rather different - and I
think would benefit from having a central group of component maintainers
and companies to co-ordinate development.

I suppose this is a question specifically for Michael, but not
exclusively so.

  :-)

The ESC, do you see this as a very active group, for instance working as
the release team, meeting often and looking at individual issues?

  Wrt. looking at individual issues, probably not - unless they have wide
reaching consequences; but to better co-ordinate on the burning issues
of the day, and have a final say on things like:

  "Do we port entirely to Java" (I think 'no' but ... :wink:

  And to be a responsible backstop for technical issues - which often
simply require a decision - any decision being far better than none.

  Wrt. having a separate release team for LibreOffice - that makes a lot
of sense too, though I don't know that it needs to be that formal, and
should include all those whose job it is to get builds out there.

  Regards,

    Michael.

Right - thanks. Your description is what I understood as their function,
the 'release team' reference simply caused be to say 'Huh?'.

Ciao,

Drew

Cor Nouws schrieb:

Charles-H. Schulz wrote (14-11-10 13:49)

I have a question about the Board of Directors. I see that the draft
says all members are chosen every year (from the Community Members)
and that one can be member only twice consecutively.
Has been considered that this leads to a situation where each year
people have to get used to the tasks, the other board members etc. so
that maybe it is a bit inefficient?

Well, that is a good question. My personal take was at first for a 2
years mandate. Then some others thought that 6 months would be good. I
sliced the apple into two :slight_smile:

I have seen many council and board tables, but never met rules with less
then three years for membership.
Of course nowadays all and everything (and especially we) is much
faster, but IMO two years is the bare minimum to prevent spoiling lots
of time on learning each other and the work so often, and to reach some
continuity and stability.

In line with this, I would propose split elections: Appr. 50% of the
seats each year.

+1

Bernhard

Hi guys,

Cor Nouws schrieb:
>>> Has been considered that this leads to a situation where each year
>>> people have to get used to the tasks, the other board members etc. so
>>> that maybe it is a bit inefficient?

  This is fairly normal, and there is usually both change and continuity
in things like the GNOME board. Also, old-timers are usually around and
willing to help out mentoring / getting people up-to-speed.

>> Well, that is a good question. My personal take was at first for a 2
>> years mandate. Then some others thought that 6 months would be good. I
>> sliced the apple into two :slight_smile:

  I like a year-long term; it seems a good balance.

> In line with this, I would propose split elections: Appr. 50% of the
> seats each year.
+1

  So - I havn't got to looking at this in detail yet; but I strongly
recommend a 'fair' voting scheme - such as used by GNOME - ie. STV. This
makes it very difficult for a contributor with 51% of the votes to get
100% of the seats [ which 1st past the post assures ].

  However - the obvious benefits of STV are really watered down by a
smaller electorate due to rounding errors; obviously, if (using STV) you
elect one person at a time, you have some of the first-past-the-post
problems.

  Then, there is the admin overhead of elections, and the problems of
getting people to vote more regularly.

  Thus, overall - I would strongly recommend a single, big vote, once per
year to elect everyone - and not worry about the continuity issues: they
tend to fix themselves. The electorate tends to value such things as
"experience" in the candidate's statements.

  HTH,

    Michael.

Hi, :slight_smile:

Having been given permission to proofread and revise the initial
draft, I presumed it would be OK to do the same for subsequent
amendments. I hope I did not overstep myself there; if I did, please
say so and I will, of course, desist. However, I came up with a
revised text as below (it simply states exactly the same things, but
re-worded).

It seems that there are still some big ambiguities that would need to
be resolved:

"The Chairperson is elected by a special electoral college comprised
of the BoD, the AB and and the ESC (however, ESC members who are also
members of the BoD can only cast one single vote in this election,
regardless of their membership of both bodies). The vote by this
special college is not decided by the votes of the individual members
taken as a whole; instead, each respective body holds a vote among its
members, and returns a nomination of one candidate (a specific list of
names, or one name only, will have been submitted by the BoD and the
AB). The three bodies therefore arrive at a shortlist of three
nominees. If one of the three nominees has a majority within the
shortlist (has two votes out of three, or is a unanimous choice), the
outcome is deemed to be decisive and the electoral process is
concluded. However, if three different people are nominated, then a
conciliation process takes place, with the aim of eliminating one
nominee and making a choice between two nominees only. The
Chairperson's term of office is two (2) years, but he/she can serve as
many terms as are seen fitting."

1) "(however, ESC members who are also members of the BoD can only
cast one single vote in this election, regardless of their membership
of both bodies)": So which body do they cast their vote in? How and
when is that decision taken? The choice could change the outcome of
the voting.

2) "(a specific list of names, or one name only, will have been
submitted by the BoD and the AB)": How would the list be drawn up?
Perhaps you need at least a cross-reference to another clause in the
bye-laws that resolves that question? If there's only one name, then
there would be no point in voting at all...

3) "However, if three different people are nominated, then a
conciliation process takes place, with the aim of eliminating one
nominee and making a choice between two nominees only.": That could
give rise to a difficult situation... IMHO, you would need to
establish a clear procedure for this, to avoid some tense deadlocks in
the future...

HTH.

David Nelson

Hello David,

Hi, :slight_smile:

Having been given permission to proofread and revise the initial
draft, I presumed it would be OK to do the same for subsequent
amendments. I hope I did not overstep myself there; if I did, please
say so and I will, of course, desist. However, I came up with a
revised text as below (it simply states exactly the same things, but
re-worded).

thank you a lot for this!!!

It seems that there are still some big ambiguities that would need to
be resolved:

"The Chairperson is elected by a special electoral college comprised
of the BoD, the AB and and the ESC (however, ESC members who are also
members of the BoD can only cast one single vote in this election,
regardless of their membership of both bodies). The vote by this
special college is not decided by the votes of the individual members
taken as a whole; instead, each respective body holds a vote among its
members, and returns a nomination of one candidate (a specific list of
names, or one name only, will have been submitted by the BoD and the
AB). The three bodies therefore arrive at a shortlist of three
nominees. If one of the three nominees has a majority within the
shortlist (has two votes out of three, or is a unanimous choice), the
outcome is deemed to be decisive and the electoral process is
concluded. However, if three different people are nominated, then a
conciliation process takes place, with the aim of eliminating one
nominee and making a choice between two nominees only. The
Chairperson's term of office is two (2) years, but he/she can serve as
many terms as are seen fitting."

1) "(however, ESC members who are also members of the BoD can only
cast one single vote in this election, regardless of their membership
of both bodies)": So which body do they cast their vote in? How and
when is that decision taken? The choice could change the outcome of
the voting.

Right, that sounds clunky so let me clarify: members of the ESC who are
also members of the BoD only vote at the BoD and not at the ESC. Is it
better?

2) "(a specific list of names, or one name only, will have been
submitted by the BoD and the AB)": How would the list be drawn up?
Perhaps you need at least a cross-reference to another clause in the
bye-laws that resolves that question? If there's only one name, then
there would be no point in voting at all...

I can clarify that, but in essence I guess 1)people will nominate
themselves to the BoD and 1)that the BoD as well as the AB can nominate
someone.

3) "However, if three different people are nominated, then a
conciliation process takes place, with the aim of eliminating one
nominee and making a choice between two nominees only.": That could
give rise to a difficult situation...

Yes. :slight_smile:

IMHO, you would need to
establish a clear procedure for this, to avoid some tense deadlocks in
the future...

Well, I think we can submit the Chairman's choice to the popular
election then (understand the TDF contributors).

Best,
Charles.

Hi there,

It seems that there are still some big ambiguities that would need to
be resolved:

  Well detected :slight_smile:

"The Chairperson is elected by a special electoral college comprised
of the BoD, the AB and and the ESC (however, ESC members who are also
members of the BoD can only cast one single vote in this election,

  Oh - wow, what is the ESC doing in that mix. I would prefer that the
board simply elect the chairman, who is just a member of the board that
has some special meeting management role :slight_smile:

  Hopefully that de-complicates the whole process; then again I havn't
read the proposal in full recently.

  I believe there is a -huge- danger of over-engineering any constitution
- particularly when you get engineers near it :slight_smile: and ending up with
some huge joke like the OO.o governance - where obscure rules seemed to
breed in dark corners :slight_smile:

  HTH,

    Michael.

Hello,

Hi there,

> It seems that there are still some big ambiguities that would need
> to be resolved:

  Well detected :slight_smile:

> "The Chairperson is elected by a special electoral college comprised
> of the BoD, the AB and and the ESC (however, ESC members who are
> also members of the BoD can only cast one single vote in this
> election,

  Oh - wow, what is the ESC doing in that mix. I would prefer
that the board simply elect the chairman, who is just a member of the
board that has some special meeting management role :slight_smile:

  Hopefully that de-complicates the whole process; then again I
havn't read the proposal in full recently.

  I believe there is a -huge- danger of over-engineering any
constitution
- particularly when you get engineers near it :slight_smile: and ending up with
some huge joke like the OO.o governance - where obscure rules seemed
to breed in dark corners :slight_smile:

So, I'm going to rewrite the ESC part; it will at least simplify the
chairman story. But having him/her part of the BoD would also nix its
role I fear.

Best,
Charles.

Hi Charles, :slight_smile:

thank you a lot for this!!!

No problem. I've got a watch on the page, and will visit whenever
there's a change then. :wink:

1) "(however, ESC members who are also members of the BoD can only
cast one single vote in this election, regardless of their membership
of both bodies)": So which body do they cast their vote in? How and
when is that decision taken? The choice could change the outcome of
the voting.

Right, that sounds clunky so let me clarify: members of the ESC who are
also members of the BoD only vote at the BoD and not at the ESC. Is it
better?

I understood what you meant, no problem there. The ambiguity is how
the decision is taken about which body they vote on... Especially as
throwing their vote in on one body or the other could maybe weight the
election in one direction or another, and change the result. My
suggestion was that it would be good to lay down unambiguous rules for
this...

2) "(a specific list of names, or one name only, will have been
submitted by the BoD and the AB)": How would the list be drawn up?
Perhaps you need at least a cross-reference to another clause in the
bye-laws that resolves that question? If there's only one name, then
there would be no point in voting at all...

I can clarify that, but in essence I guess 1)people will nominate
themselves to the BoD and 1)that the BoD as well as the AB can nominate
someone.

OK, I get the idea. Perhaps a separate, short paragraph explaining
that might be good? I could draft one tomorrow and submit it in a
standalone edit that will be easy to identify and roll back/modify if
it doesn't quite say what you want...?

3) "However, if three different people are nominated, then a
conciliation process takes place, with the aim of eliminating one
nominee and making a choice between two nominees only.": That could
give rise to a difficult situation...

Yes. :slight_smile:

I guess this is the point that, IMVHO, might be in most need of an
unequivocal procedure, as it could give rise to controversial
situations...

HTH.

David Nelson

Hello David,

Hi Charles, :slight_smile:

> thank you a lot for this!!!

No problem. I've got a watch on the page, and will visit whenever
there's a change then. :wink:

:slight_smile:

>> 1) "(however, ESC members who are also members of the BoD can only
>> cast one single vote in this election, regardless of their
>> membership of both bodies)": So which body do they cast their vote
>> in? How and when is that decision taken? The choice could change
>> the outcome of the voting.
>
> Right, that sounds clunky so let me clarify: members of the ESC who
> are also members of the BoD only vote at the BoD and not at the
> ESC. Is it better?

I understood what you meant, no problem there. The ambiguity is how
the decision is taken about which body they vote on... Especially as
throwing their vote in on one body or the other could maybe weight the
election in one direction or another, and change the result. My
suggestion was that it would be good to lay down unambiguous rules for
this...

yup. But after Michael's points, I also think we might clarify and
simplify all this a great deal. In a nutshell

1) the ESC does not get to vote, it's not elected, and it's a technical
body. The AB can propose candidate(s), but cannot vote.
2) BoD appoints the CH, by vote or by consensus. People can nominate
themselves and send their nomination to the BoD no later than 2 months
before the election date. The AB can also nominate one or several
candidates and sends the name(s) to the BoD no later than 2 months
before the election.

That way, it's easier and faster. Any thoughts?

Best,
Charles.

Sounds much cleaner & simpler to me :slight_smile:

  ATB,

    Michael.

Hi, :slight_smile:

yup. But after Michael's points, I also think we might clarify and
simplify all this a great deal. In a nutshell

1) the ESC does not get to vote, it's not elected, and it's a technical
body. The AB can propose candidate(s), but cannot vote.
2) BoD appoints the CH, by vote or by consensus. People can nominate
themselves and send their nomination to the BoD no later  than 2 months
before the election date. The AB can also nominate one or several
candidates and sends the name(s) to the BoD no later than 2 months
before the election.

That way, it's easier and faster. Any thoughts?

Best,
Charles.

Yes, I get the idea. If it's alright with you guys, I'll figure out
how to draft that in, and will give a heads-up when I've done so (over
the next 24 hours, because I'm slave to a client for the coming
hours). Is that OK?

Also, I have an idea about a couple of legal experts I could contact
and, if they're willing, invite them to jump in on this thread and
maybe help arrive at some really bullet-proof bye-laws... should I do
that?

David Nelson

Hello David,

Hi, :slight_smile:

> yup. But after Michael's points, I also think we might clarify and
> simplify all this a great deal. In a nutshell
>
> 1) the ESC does not get to vote, it's not elected, and it's a
> technical body. The AB can propose candidate(s), but cannot vote.
> 2) BoD appoints the CH, by vote or by consensus. People can nominate
> themselves and send their nomination to the BoD no later  than 2
> months before the election date. The AB can also nominate one or
> several candidates and sends the name(s) to the BoD no later than 2
> months before the election.
>
> That way, it's easier and faster. Any thoughts?
>
> Best,
> Charles.

Yes, I get the idea. If it's alright with you guys, I'll figure out
how to draft that in, and will give a heads-up when I've done so (over
the next 24 hours, because I'm slave to a client for the coming
hours). Is that OK?

it's more than ok, thank you again!

Also, I have an idea about a couple of legal experts I could contact
and, if they're willing, invite them to jump in on this thread and
maybe help arrive at some really bullet-proof bye-laws... should I do
that?

Well, we do have lawyers to, for instance we have Gianluca here (but
I'm sure there are others) and also Florian Effenberger who despite any
evidence of the contrary, is not a system administrator :slight_smile:

Best,
Charles.

Hi Michael,

Michael Meeks wrote (22-11-10 11:50)

Hi guys,

Cor Nouws schrieb:

Has been considered that this leads to a situation where each year
people have to get used to the tasks, the other board members etc. so
that maybe it is a bit inefficient?

  This is fairly normal, and there is usually both change and continuity
in things like the GNOME board. Also, old-timers are usually around and
willing to help out mentoring / getting people up-to-speed.

Well, that is a good question. My personal take was at first for a 2
years mandate. Then some others thought that 6 months would be good. I
sliced the apple into two :slight_smile:

  I like a year-long term; it seems a good balance.

   What is a balance between a reasonable, common term (2 or 3 years) and non-sense (6 months)?

In line with this, I would propose split elections: Appr. 50% of the
seats each year.

+1

  So - I havn't got to looking at this in detail yet; but I strongly
recommend a 'fair' voting scheme - such as used by GNOME - ie. STV. This

   Can you pls tell what STV is?
   And what has a fair voting scheme to do with 100% each year or 50% each year?

makes it very difficult for a contributor with 51% of the votes to get
100% of the seats [ which 1st past the post assures ].

  However - the obvious benefits of STV are really watered down by a
smaller electorate due to rounding errors; obviously, if (using STV) you
elect one person at a time, you have some of the first-past-the-post
problems.

  Then, there is the admin overhead of elections, and the problems of
getting people to vote more regularly.

   Both voting each year for 100% of the seats, or each year for 50% of the seats, result in one election per year. With less counting in the letter case :wink:

  Thus, overall - I would strongly recommend a single, big vote, once per
year to elect everyone - and not worry about the continuity issues: they
tend to fix themselves. The electorate tends to value such things as
"experience" in the candidate's statements.

   The reasons you gave, that were clear for me so far, did not at all convince me.

   Maybe you can further explain what you think to gain with 100% of the seats each year?

Regards,
Cor