Am Montag, 18. Mai 2020, 14:34:55 CEST schrieb Simon Lees:
On 5/18/20 8:37 PM, Richard Brown wrote:
> On Sat, 2020-05-16 at 14:34 -0700, Fraser_Bell wrote:
>> Although, I must say, the thought that a minority can cause a full
>> re-election of the elected Board Officials deeply distresses me.
>> In that case, if that happens, I personally will be considering
>> up my openSUSE Membership and leaving openSUSE.
Gerry, I have to admit that I disagree with your opinion about the 20%
rule - but nevertheless, I'd be be sad to see you leaving. Even if, and
especially because ...
>> so this is a matter of Principle with me.
... I really like people who stand for their principles. We need more of
them, not less ;-)
(Besides that, I like your music ;-)
> I think the point you are neglecting to consider
is that the 20%
> is not "The membership should be proactively asked to see whether
> of them disagree with the democractically elected Board and to
> trigger a reelection"
> But the 20% rule is more "if 20% of the membership is driven enough
> to, themselves, proactively push for re-electing the whole Board,
> then that re-election is triggered"
That's something that is not defined in the rule. Of course we can
speculate in both directions how it was meant ;-)
I know from some associations in germany that their rules state that
members have a right to get the list of members if they want to start a
non-confidence vote so that they can contact all members. (Without this
rule, the board could keep the membership list secret and easily prevent
the non-confidence vote to succeed.)
That rule is even more extreme than "the membership gets asked" - and
still, I haven't seen a non-confidence vote in the associations I know
in all the years I remember.
> Given we're a project where 1 person can
contribute and change a lot
> technically, if the Board is ruling in a way to, speaking frankly,
> piss off 20% of the membership enough to drive them to campaign
> against the Board, then maybe re-electing the Board is the best
> option for community harmony.
Given some of your mails in the last months, I'm surprised to say:
I couldn't agree more (well, except s/maybe// ;-)
> That said, I think the rule was never written
with the idea of the
> election committee getting involved.
I'm not sure if the people who wrote this rule thought about how it
would work in practise. I'd guess that they hoped there would never be a
need for this rule to be used - and IMHO that's the most likely
explanation why the 20% rule doesn't go into all details.
> I certainly always envisioned
> that such cases requiring far more proactive engagement from the
> disgruntled 20% than just 1 person calling for a vote of no
> confidence..a call that wasn't even seconded on the mailinglist..
I'm not going to re-read the full discussion, but I remember some mails
that were quite close to seconding that call. And, see below, stating
that in public clearly isn't easy.
When looking at this the problem we have is with
things like mail
spoofing etc The only way we can actually be sure that the opinion's
we are getting are actually from members is with Helios and its then
obvious that its best that the board isn't handling that process.
There's another thing I'd even consider more important than the
technical difficulties of requiring people to send mails:
the chilling effect.
I can imagine that a requirement to send a mail to a public mailinglist
(including public archives) would stop several people from signing the
non-confidence vote/petition. Not because they aren't pissed off enough,
but because they are afraid of getting punished (in whatever way) for
stating their opinion.
Just as a hypothetical example: let's say Richard would send a public
statement that the board pissed him off and should resign. Some board
members are his colleagues, and I'm somewhat sure that they would
(diplomatically spoken) like him less after reading such a statement. As
a result, working together for sure won't be easier afterwards - even if
it's completely unrelated to what the board did.
Of course this isn't limited to colleagues working in the same company.
It also applies to community members, and I'd hate to see friendships
breaking because of "what? you voted against the board?!"
IMHO _that_ is the most important reason why using Helios makes sense -
it limits the amount of people who can see who voted against the board
to the election commitee (I hope using voter aliases gets enabled to
ensure this!) and limits the chilling effect.
Actually I'd even love to see a second option to choose, maybe "ignore
me". It wouldn't have an impact on the end result (we'd still need 20%
of the members asking for a re-election), but if only a few people
choose it, it would make it impossible even for the election commitee to
know who voted against the board because "$person voted" would no longer
be equal to "$person voted against the board".
Don't get me wrong: I trust the election commitee.
Nevertheless, I'd still prefer a solution that doesn't even need any
trust when it comes to such sensitive data - data that is probably more
sensitive than which candidates you vote for in a board election.
But maybe I'm just too paranoid from years of AppArmor development and
several so-called "AppArmor WTF moments" ;-)
There are good reasons why elections have secret ballots.
A non-confidence vote is more sensitive, therefore it also has to have
secret ballots (with the obvious exception that a brave person needs to
publicly start it).
think the election committees compromise of a petition
does a good job of keeping with the spirit of the rule and
requiring people to step up if they are unhappy even if it
undoubtly does lower the bar for a re-election
Yeah with the perspective of an upcoming foundation I think whatever
we come up with needs to have only one clear interpretation
Agreed, and ideally it should also include a timeline - two months are
clearly too long. (Not meant as critism of the election commitee - I can
imagine that it wasn't easy to come up with a good way to handle this
without having any guidelines in our election rules.)
Even with issues
around email addresses etc I think documenting that 20% of membership
must +1 an email to openSUSE project would probably be workable.
I completely disagree with requiring +1 mails to opensuse-project
(unless you want to make the board really untouchable by abusing the
chilling effect - but then, it's easier to simply drop the 20% rule ;-)
Using a petition with non-public list of signers is the only sane way
how the 20% rule can work in the way it's intended, because being afraid
of getting punished (in whatever way) for signing the non-confidence
vote should never ever be a reason not to sign.
 AppArmor WTF moments are the moments when you check your audit.log
or use aa-logprof to update your AppArmor profiles, and see that a
program wants permissions you'd never expect it needs.
As an extreme (and luckily made up) example: if ping would require
read access to your home directory, that would qualify ;-)
// If non-crazy input manages to reach this code path,
// we should consider it a bug.
[from MediaWiki 1.27.4 languages/LanguageConverter.php]
To unsubscribe, e-mail: opensuse-project+unsubscribe(a)opensuse.org
To contact the owner, email: opensuse-project+owner(a)opensuse.org