Re: Making the RMS resolution a Secret Ballot
Le dimanche, 11 avril 2021, 23.10:53 h CEST Jonathan Wiltshire a écrit : > So I suggest the DPL directs under s5.1 something like: > > "The secretary shall delay publication of the the association between > identify and ballot on the tally sheet, for a period of not more than 90 > days, unless directed otherwise by a General Resolution of the Developers." > > A GR about future secret ballots should make provision for how to apply it > retrospectively to this one. This preserves the 3:1 requirement that there > would have been were it being decided in advance. That's a quite interesting approach, thank you for it. My only concern is that deciding on the privacy of vote ballots *given the aggregated results*, is bound to be influenced by these results. I wouldn't be surprised by voters wanting to see the list of names _more_, depending on the results. -- OdyX signature.asc Description: This is a digitally signed message part.
Re: Making the RMS resolution a Secret Ballot
Hallo, * Didier 'OdyX' Raboud [Sun, Apr 11 2021, 06:00:30PM]: > For what I'm concerned, I don't "insist on making the personal views on this > GR public", because it was always clear (to me) that all voters' personal > views on this GR would end up being made public by the secretary on our > website; this is how we have all experienced the publication of GR results for > at least a decade. I insist on "not changing how we collectively run through > the GR experience in the middle of a sensitive GR". This is not "dragging my > fellows into denuding themselves", because my expectation is that every voter > is well aware that their vote would be published, in this GR as in any GR. [1] I didn't have you (personally) in mind while writing this but some other voices on this thread. Those which go along the line "I know that my vote will become public and so it gives me the right to see what others think as well, and also to everybody else in the world". Sorry, it doesn't, that is twisted logics. I mean, what is the loss if you don't see other people's votes? There isn't much unless someone wants to discriminate people with different opinions. Inside and OUTSIDE of Debian. And not sure how you get the impression but I don't really object to have a visible list of people's opinions on certain subjects, as long as the purpose of this information is bound to the technical work in the project itself. The problem with this GR is, the Rubicon was crossed when a certain group decided to repurpose our infrastructure into a survey engine for topics which are outside of our turf. If we go this way, then (in my opinion) the defaults of the system should be adjusted to be more fair and fail-safe for political/delicate voting. Actually I was thinking about proposing another option, like "this GR should not exist until our processes are adjusted" but there was not enough time to think it through, especially since that certain group decided to shorten the discussion period. Do the math. > If you don't want to (or cannot afford to) see your personal views on this > GR to be made public, then don't vote. > > Don't misread me; it is a very very serious concern that external factors > (threats, pression, etc) make it so that some of our voters will not vote in > fear of retaliation. But I think that if we, as a project, accept to bend our > traditional and constitutional procedures under that external pressure through > emergency exceptional measures, we also make the project more vulnerable to > future external pressure; we also weaken this GR's results too. See above, I don't see how adding a little bit of confidentiality would suddenly expose us to external pressure. But it would improve the GR quality by eliminating the possibility of retaliation from outsiders. For those who want a GR to be a messenger for showing their opinion to the public, we probably should implement a checkbox in the ballot meaning "My vote should become public". > We must protect our members from harassment; we must defend the project from > external influence; and we must call out external pressures in the strongest > terms possible. Threats against Debian project members because of their public > opinions in the project are *not acceptable*, ever [1], and we must stand, as > a project, against those threatening our members and our community. The > problem we face here is the pressure and the threats against project members, > not the publication of the GR votes. Let's not forget that. Well, we can state this the whole day long but you are talking about factors which are probably outside of our control and which are hard to prove and punish. Best regards, Eduard.
Re: What does FD Mean
* Adrian Bunk [2021-04-11 20:53]: I am not saying people were stupid. Okay, that was hyperbolic. But you have to admit that you don't seem to put much confidence in people's ability (or willingness) to read the explanations that come with each ballot. I am by no means a voting system nerd and found them quite understandable. It can be hard to vote correctly in a voting system that is very different from what you are used to in real life, unless you are a nerd in voting systems. If you ask me as someone who has never used a ranked vote before, it is not that hard. The hard part is how the winner is determined from the votes. I don't think many people will bother to independently verify the results. -- ⢀⣴⠾⠻⢶⣦⠀ ╭╮ ⣾⠁⢠⠒⠀⣿⡁ │ Timo Röhling │ ⢿⡄⠘⠷⠚⠋⠀ │ 9B03 EBB9 8300 DF97 C2B1 23BF CC8C 6BDD 1403 F4CA │ ⠈⠳⣄ ╰╯ signature.asc Description: PGP signature
Re: Making the RMS resolution a Secret Ballot
On Fri, Apr 09, 2021 at 10:59:16AM -0700, Russ Allbery wrote: > I support this approach and believe the DPL should decide under 5.1(3) > that Debian will not publish the association between identity and ballot > for the RMS resolution. I too support the mechanism to achieve it but not the permanence, without consultation with the full project. We should be mindful that a change to the foundation documents to make future votes secret would require a 3:1 majority, and I think it would look odd to outsiders that we've just decided the rules don't apply to this vote. Given the anticipated GR on future secret ballots and the ability for 2K developers to override the DPL anyway, 90 days should be sufficient to achieve a consensus. So I suggest the DPL directs under s5.1 something like: "The secretary shall delay publication of the the association between identify and ballot on the tally sheet, for a period of not more than 90 days, unless directed otherwise by a General Resolution of the Developers." A GR about future secret ballots should make provision for how to apply it retrospectively to this one. This preserves the 3:1 requirement that there would have been were it being decided in advance. (I'm conscious about the extra work this creates for the secretary. Kurt, I really appreciate how you're handling all this and I'm sorry to be asking for more complexity.) -- Jonathan Wiltshire j...@debian.org Debian Developer http://people.debian.org/~jmw 4096R: 0xD3524C51 / 0A55 B7C5 1223 3942 86EC 74C3 5394 479D D352 4C51 signature.asc Description: PGP signature
Re: What does FD Mean
Le dimanche 11 avril 2021 à 20:53:33+0300, Adrian Bunk a écrit : > On Sun, Apr 11, 2021 at 06:50:36PM +0200, Timo Röhling wrote: > > > > Besides, I am still unconvinced and mildly offended by the assumption > > that people who voted 1--- were too stupid to do it right. > >... > > I am not saying people were stupid. > > It can be hard to vote correctly in a voting system that is very > different from what you are used to in real life, unless you are > a nerd in voting systems. I've discussed with another Front Desk member about adding a question on our voting system in the nm templates. The idea being to make sure if people have questions, they get some answers, and otherwise relevant pointers to doc. -- Pierre-Elliott Bécue GPG: 9AE0 4D98 6400 E3B6 7528 F493 0D44 2664 1949 74E2 It's far easier to fight for principles than to live up to them. signature.asc Description: PGP signature
Re: Making the RMS resolution a Secret Ballot
Le dimanche 11 avril 2021 à 18:12:56+0200, Micha Lenk a écrit :
> Hi Pierre-Elliott,
>
> Am 11.04.21 um 14:27 schrieb Pierre-Elliott Bécue:
> > > Those who insist on making the personal views on this (non-technical!!!)
> > > GR public should be ashamed of dragging their fellows into denuding
> > > themselves for no good reason.
> >
> > I am really worried that someone could say that holding to our values,
> > respecting our processes and avoiding to create potentially
> > indesirable precedents is "no good reason" and grounds to be ashamed.
>
> I haven't seen anybody suggesting to violate our processes or alike. What
> I've seen is a request to make use of 5.1(3) of our constitution, but I
> consider that a (granted, rarely used) part of our processes.
>
> What I see is a tension between our social contract and our code of conduct.
> The social contract paragraph three ("We will not hide problems") talks
> about doing our business in public, yet refers to our bug reports. On the
> other hand our code of conduct paragraph five ("Be open") clarifies that
> public methods of communication are preferred, "unless posting something
> sensitive." Reading the many mails on this context let me think that we have
> a rough consensus to consider the RMS GR results something sensitive. So, I
> consider the publication of the vote results for this GR a violation of our
> code of conduct...
>
> We haven't been in such a situation before (e.g. I am not aware of any
> previous GR about personal related issues), so I think we as a project need
> to find a balance for handling such situations in the future. Long term to
> me this could mean to change our constitution, e.g. allowing secret votes
> also for non leader votes in the future. Yet this doesn't help in the
> current situation.
>
> Pierre-Elliott, given the current social contract and the code of conduct,
> would you mind to elaborate a bit how you see our processes violated by
> letting the current leader make use of our constitution 5.1(3) to publish
> the RMS GR results as an anonymized tally sheet like in our yearly leader
> elections?
The CoC, although essential, doesn't supersedes the Constitution. It
actually is superseded by the Constitution.
4.2.3 reads "after the vote the Project Secretary lists all the votes
cast".
We are trying to use another part of the constitution to override this
part of the constitution.
Given how we do usually, I think this 5.1.3 usage is not consistent with
why it has been written, and therefore I think we are trying to break a
well-established process out of fear.
And I think this could give incentive to the outside to apply more
pressure the next time to see how much more we can bend.
Whether you agree or not, I honestly don't really care, because as I
already said: I won't make a fuss about it if the DPL tries to make the
vote secret and the secretary follows.
With best regards,
--
Pierre-Elliott Bécue
GPG: 9AE0 4D98 6400 E3B6 7528 F493 0D44 2664 1949 74E2
It's far easier to fight for principles than to live up to them.
signature.asc
Description: PGP signature
Re: What does FD Mean
On Sun, Apr 11, 2021 at 06:50:36PM +0200, Timo Röhling wrote: > > Besides, I am still unconvinced and mildly offended by the assumption > that people who voted 1--- were too stupid to do it right. >... I am not saying people were stupid. It can be hard to vote correctly in a voting system that is very different from what you are used to in real life, unless you are a nerd in voting systems. > Cheers > Timo cu Adrian
Re: What does FD Mean
* Adrian Bunk [2021-04-11 19:06]: Such a change would not remove any voting options since votes like 1--- have equivalent espressions like 1222. I disagree. Not on theoretical grounds, but if you assume that someone votes 1--- because they are too lazy to read the instructions, how do you think they will respond to an error message telling them they forgot to rank *all* options? Most likely, they'll vote 12345678. Besides, I am still unconvinced and mildly offended by the assumption that people who voted 1--- were too stupid to do it right. But for the sake of your argument, a much simpler solution would be a change in Devotee to report back the vote in normalized form, e.g. "Your vote has been recorded with the following effective rankings: 1222 1 Foo 2 Bar 2 Baz 2 Further Discussions If this is not your intention, please vote again with a corrected ballot." Cheers Timo -- ⢀⣴⠾⠻⢶⣦⠀ ╭╮ ⣾⠁⢠⠒⠀⣿⡁ │ Timo Röhling │ ⢿⡄⠘⠷⠚⠋⠀ │ 9B03 EBB9 8300 DF97 C2B1 23BF CC8C 6BDD 1403 F4CA │ ⠈⠳⣄ ╰╯ signature.asc Description: PGP signature
Re: Re: What does FD Mean
On Thu, Apr 08, 2021 at 10:51:26AM +0100, Barak A. Pearlmutter wrote: > Hey Adrian, Hi Barak, > > When looking at the tally of the latest systemd vote,[1] > > there are plenty of votes like > > 1--- > > > > It is obvious what these voters wanted to express, > > and that their ballot was wrongly filled due to a > > lack of understanding how our voting system works. > > That's really interesting. > > Maybe we should cobble up a GUI tool to download & correctly fill out > & sign & email a ballot? Instead of numbering items (which requires > knowing how to count) people could drag them into an order, etc. This > would allow inactive or non-uploading DDs, ones who can't manage (or > can't be arsed) to fill out and gpg-sign a ballot, to express their > valuable opinions. It could look for the right key using the > devscripts mechanisms, like ~/.devscripts DEBSIGN_KEYID and > environment variables and such. Maybe check if any available private > keys appear on the Debian keyring. That way people who haven't needed > their key in years could still vote. > > (Not sure if joking.) after thinking about it for a few days, I suggest the following change to the Constitution: +All options must be ranked. -Not all options need be ranked. -Ranked options are considered preferred to all unranked options. Voters may rank options equally. -Unranked options are considered to be ranked equally with one another. Such a change would not remove any voting options since votes like 1--- have equivalent espressions like 1222. But it would help voters who are not used to ranking from real-world elections. > --Barak. cu Adrian
Re: Making the RMS resolution a Secret Ballot
Hi Pierre-Elliott,
Am 11.04.21 um 14:27 schrieb Pierre-Elliott Bécue:
Those who insist on making the personal views on this (non-technical!!!)
GR public should be ashamed of dragging their fellows into denuding
themselves for no good reason.
I am really worried that someone could say that holding to our values,
respecting our processes and avoiding to create potentially
indesirable precedents is "no good reason" and grounds to be ashamed.
I haven't seen anybody suggesting to violate our processes or alike.
What I've seen is a request to make use of 5.1(3) of our constitution,
but I consider that a (granted, rarely used) part of our processes.
What I see is a tension between our social contract and our code of
conduct. The social contract paragraph three ("We will not hide
problems") talks about doing our business in public, yet refers to our
bug reports. On the other hand our code of conduct paragraph five ("Be
open") clarifies that public methods of communication are preferred,
"unless posting something sensitive." Reading the many mails on this
context let me think that we have a rough consensus to consider the RMS
GR results something sensitive. So, I consider the publication of the
vote results for this GR a violation of our code of conduct...
We haven't been in such a situation before (e.g. I am not aware of any
previous GR about personal related issues), so I think we as a project
need to find a balance for handling such situations in the future. Long
term to me this could mean to change our constitution, e.g. allowing
secret votes also for non leader votes in the future. Yet this doesn't
help in the current situation.
Pierre-Elliott, given the current social contract and the code of
conduct, would you mind to elaborate a bit how you see our processes
violated by letting the current leader make use of our constitution
5.1(3) to publish the RMS GR results as an anonymized tally sheet like
in our yearly leader elections?
Regards,
Micha
Re: Making the RMS resolution a Secret Ballot
Le dimanche, 11 avril 2021, 01.02:18 h CEST Eduard Bloch a écrit : > Those who insist on making the personal views on this (non-technical!!!) > GR public should be ashamed of dragging their fellows into denuding > themselves for no good reason. Just clarifying one thing here, to make sure there's no misunderstanding. All non-election Debian GRs have had the tally sheet of nominal votes published, thereby making the personal views of all voters public. [0] ( Looking at the last GR; https://www.debian.org/vote/2019/vote_002.en.html links to https://www.debian.org/vote/2019/vote_002_tally.txt, which has your vote: "V: 67512348 blade Eduard Bloch". ) For what I'm concerned, I don't "insist on making the personal views on this GR public", because it was always clear (to me) that all voters' personal views on this GR would end up being made public by the secretary on our website; this is how we have all experienced the publication of GR results for at least a decade. I insist on "not changing how we collectively run through the GR experience in the middle of a sensitive GR". This is not "dragging my fellows into denuding themselves", because my expectation is that every voter is well aware that their vote would be published, in this GR as in any GR. [1] If you don't want to (or cannot afford to) see your personal views on this GR to be made public, then don't vote. Don't misread me; it is a very very serious concern that external factors (threats, pression, etc) make it so that some of our voters will not vote in fear of retaliation. But I think that if we, as a project, accept to bend our traditional and constitutional procedures under that external pressure through emergency exceptional measures, we also make the project more vulnerable to future external pressure; we also weaken this GR's results too. We must protect our members from harassment; we must defend the project from external influence; and we must call out external pressures in the strongest terms possible. Threats against Debian project members because of their public opinions in the project are *not acceptable*, ever [1], and we must stand, as a project, against those threatening our members and our community. The problem we face here is the pressure and the threats against project members, not the publication of the GR votes. Let's not forget that. -- OdyX [0] For as long as I remember at least, but certainly since I gained voting rights in 2011. I just checked, the 1999 vote about logo licenses already had a public tally sheet; https://www.debian.org/vote/1999/result_0002 [1] From https://lists.debian.org/debian-devel-announce/2021/04/msg1.html > VOTING SECRECY > This is a non-secret vote. After the voting period is over the details on > who voted what will be published. [2] Critics against one's opinion of course are. signature.asc Description: This is a digitally signed message part.
Re: Making the RMS resolution a Secret Ballot
Le 11 avril 2021 01:02:18 GMT+02:00, Eduard Bloch a écrit : >Those who insist on making the personal views on this (non-technical!!!) >GR public should be ashamed of dragging their fellows into denuding >themselves for no good reason. I am really worried that someone could say that holding to our values, respecting our processes and avoiding to create potentially indesirable precedents is "no good reason" and grounds to be ashamed. No one tries to shame you for defending a switch to a secret vote. Maybe you could just do the same. -- Pierre-Elliott Bécue From my phone
Re: Making the RMS resolution a Secret Ballot
On 11/04/2021 08:53, Kurt Roeckx wrote: > As secretary, I do not intend to make the vote secret. Given Debian's current rules that is probably correct. > Personally, I would like to see the vote secret. Me too. > I think the best way to make the vote secret is that the DPL makes a > decision. We have a process that can deal with people not agreeing > with that decision. That seems to be the sensible way forward, those who disagree can use the mechanism already in place to overturn his decision. Colin -- Colin Tuckley | +44(0)1223 830814 | PGP/GnuPG Key Id G8TMV | +44(0)7799 143369 | 0xFA0C410738C9D903
Re: Making the RMS resolution a Secret Ballot
On Sun, Apr 11, 2021 at 09:29:21AM +0900, Charles Plessy wrote: > I agree for the GR vote to be secret. I understand others came to a > different conclusion. I trust Kurt for making the right decision. I > will not complain about it. As secretary, I do not intend to make the vote secret. Personally, I would like to see the vote secret. I think the best way to make the vote secret is that the DPL makes a decision. We have a process that can deal with people not agreeing with that decision. Kurt
