Anonymous votes

Adding some discussion here, I'm personally ok with non-anonymous voting, but it may be worth considering the implications that non-anonymous voting vs anonymous voting have, particularly those situations in which a given individual may be part of multiple communities for which there might be outside influence.

For instance a committer which works for a company may have an interest in the outcome of a vote for which a committer (who is an employee of that company) has a viewpoint which is at odds with the companies desires. Not saying that does or even will happen, but it might be worth taking into consideration.

Can we switch this poll to non-anonymous one (or create a parallel one)?
I don't think technical votes should be anonymous

My position was always that decisions about people (i.e., elections) should be anonymous and any other polls/votes should be public.

However committers and individual communities are a little bit special in the sense that the people represent just themselves in these communities which differs from the other communities where your votes should be public also because you should vote on behalf of a company/organization and not just based on your individual interests.

@Nicola Tuveri Thank you for starting a discussion. I always prefer voting anonymously; in my opinion, the result from anonymous voting is more solid (although it might be surprising quite often). People participating in an anonymous poll cast their vote without fear of how others may feel about their position. I agree that there should be an open discussion that precedes every poll, where various reasons and views are shared among participants. In this poll (ansi-c -> c-99) the discussion was skipped. I think I'm just experiencing a learning moment here. Next time I will start with a discussion thread before proceeding to a poll.

@Dmitry Belyavsky There are two options: continue with the anonymous poll or scratch it and start from the beginning. It's just not fair to change the rules while voting is running. Do you want me to cancel this poll and start a new non-anonymous one?

@Tomas Mraz It did not come to my mind that we can distinguish between elections (about people) and technical polls. We can go that way so elections will be anonymous and all other polls non-anonymous.

If you don't mind, I'd like to continue with the anonymous poll here. Also, keep in mind that even the result (change to C-99) means nothing here because the decision about the switch will happen elsewhere. We (as committers) are going to ask advisory committees to change the technical policy so ansi-c will get replaced with C-99.

I'm not sure what's going to happen when the poll is over, if every participant will be able to see comments/questions at particular votes. Perhaps we can just do it the other way round, we can finish anonymous voting and then discuss the result. This, I think, also works.

We can continue this poll as anonymous, WFM

Anonymous voting has come up in Debian too, where only elections used to be anonymous. But this vote changed all votes to be anonymous: https://www.debian.org/vote/2022/vote_001

I really like the point shared in the link below and hope others take the time to read it. It genuinely makes sense to me and reflects what I think about this topic.

https://www.researchgate.net/publication/262855066_Against_the_secret_ballot_Toward_a_new_proposal_for_open_voting

OK I see the point in the paper:

Voting should be organized in a manner which means simply that no one
can be sure that how he votes will be hidden from his friends and associates.

it makes sense and I like it. However, I don't know how to set it up in the tool we are using. I'd like to organize a poll anonymously in the sense that no one can see how others vote until the poll is closed and the results are shown. Maybe I misremember, but during the election to the board, I could see how other people voted in my group and I could eventually (even unconsciously) adjust my poll accordingly. This is bad. I need people to cast their vote independently as much as possible. I expect people to do the homework, re-think the question, and cast their vote, such is the best for them and the project. I don't mind if the poll gets unanonymized after results are published, but I don't know how to set it up here.

OK I've just found it in the manual: one has to tick option 'Hide Results' In this case the tool will do exactly what I want. voting will not be anonymous but results are not shown until the voting period closes. After that everyone can see how each member voted.

( https://help.loomio.com/en/user_manual/polls/starting_proposals/index.html#hide-results)

I think this is the right way to do. Another learning moment for me.

I was surprised to see this particular vote being anonymous. My first thought was "darn, are the stakes that high?"

Our values are to be open and transparent and in discussions like this I think they should always be open and visible and not just shown at the end. The voting process encourages engagement if there are dissenting views and that is critically important for accurate measurement. If you can point people to an issue where you know they have a strong option but just don't participate in the general discussions (as they are less engaged in the project work) then you will get a better quality result. Otherwise it is often a tiny echo-chamber effect that is then used to justify an outcome that isn't actually supported by the wider community.

I would redo the poll as open and visible as everyone votes - as I think those who have already voted are motivated enough to do so again.

I also think polls of communities where there are lots more people engaged and much more representative of our users are more important than polls of the committers - but also getting the committer viewpoints on record and visible helps - and also to see the debate.

The other aspect is that where there is a discussion going on, people can and do change votes/opinions - and anonymous voting entirely removes that option.

@Tim Hudson

OK, I will start a new poll. Please let me know if I should proceed with:

a) Anton's system, where an individual's votes are shown when a poll is over

b) a system where an individual's votes are shown as the vote is cast.

I personally prefer a) as it makes it harder if an individual tends to vote with (or against) the crowd.

Thanks for the clarification.

@Sasha Nedvedicky It’s nice to see (a) named after me, but it shouldn’t be; and (b) what I’d like to see here too - Loomio allows for that type of poll.

To have people able to see the result and the patterns and be able to change based on others views and to encourage debate and engagement during the vote, only b) makes sense to me. I also think the current a) needs adjustment to show the running status (number of votes and expected result) at the very least to be useful for encouraging participation during the vote.

@Tim Hudson, not sure, though, that discussion on the topic has to be tied to the vote itself. Also, technically speaking, the way the poll is presented (taking https://openssl-communities.org/p/PS4NqW5C/switching-from-ansi-c-to-c-99- as an example), I see a series of statements, rather than a discussions. Not quite the same thing.

@Richard Levitte  I think in this platform we need to anyway create a thread for discussion, and it can be tied to a (new) poll.

I've started new poll here:

https://openssl-communities.org/d/3V4qoIdX/switching-from-ansi-c-to-c-99-discussion

thanks for participating again.