About the “condorcet” votes

I just saw the vote of NOKUBI Takatsugu who posted it by error on debian-vote.

The vote ranked “Further discussion” second just after his preferred option. I find such a vote too *strong*. This vote really means “My option or none”, or in other words: I don’t want any compromise (which is quite strange given that his first option is the “compromise” option).

I believe that we should vote with the aim to have a winner in the end and as such we should avoid putting something below “Further discussion”. IMO the only valid reason to put something below is if that option would hurt Debian in one’s own opinion.

Just rank the options in your order of preference. I ranked “Further Discussion” last, it’s my way of accepting the diverging opinions within our project.

(So there’s nothing personal against Takatsugu, I just took the opportunity of his little mistake to point out something I find important)

BTW, if you haven’t voted, please do. Even if you don’t care about the outcome, vote and leave all the fields empty (or rank them equally), that way you’ll express that you’re happy whatever the outcome is and you won’t be part of a silent majority. And the outcome of the vote will be stronger.

This leads me to the following question: I wonder if we shouldn’t require DD to vote and if they don’t participate in 2 or 3 consecutive votes, they shall be considered by the MIA team… it would be a kind of implicit “ping of maintainers”.

Update: FYI, Takatsugu thanked me by private mail for the explanation and will recast another vote.


  1. Anonymous says

    I wonder if we shouldn’t require DD to vote and if they don’t participate in 2 or 3 consecutive votes, they shall be considered by the MIA team… it would be a kind of implicit “ping of maintainers”.

    DDs are not required to vote, nor should they be. However, the MIA team does look at voting participation as one indication of activity from a maintainer.

  2. says

    Please don’t judge other developers’ votes, this is a fascist tendency. They know how to vote, and they know how to rank their preferred options.

    NOTA means: “do nothing”. If you disagree with a proposal, you surely prefer to do nothing than accepting the proposal.

  3. says

    Josselin, given the kind exchange that I had with Takatsugu, it’s clear that not everyone understands fully the principles of Condorcet voting, and it’s quite understandable. Most of us never used Condorcet voting for something else than Debian and most of us never learnt Condorcet voting at school.

    But every time that I give *my opinion* you treat me of a fascist. Why the hell ? What’s so difficult about accepting that we have diverging opinions and stop insulting me each time that I open the mouth ?

    Of course, FD == NOTA == do nothing. But let’s face it: we have disagreement that could hinder our progress to release etch in time and it’s better to take a decision (whatever it is) than to not take a decision. So, for me, ranking further discussion last is the best course of action given that most views are quite fairly represented on the ballot.

    In response to Glandium, yes, IMO option 3 would not harm the project. Just like releasing sarge with the same documentation didn’t harm the project.

  4. says

    FD = further discussion, not do nothing. If everyone chooses to do nothing for a while, they’re ignoring the result: try again to find agreement.

  5. says

    Substancially, option 3 says that as far as the license allows (some) modifications, it doesn’t fail DFSG #3.
    Example of a license text that would allow (some) modifications: You may ditribute modified versions of this software provided that your modifications happen in the do_nothing() function.

  6. says

    Glandium, option 3 says “the GFDL conforms to the spirit of the DFSG”. It doesn’t mean that we’ll take the same decision with even dumber licenses. So my interpretation of option 3 is *not* “as far as some modifications are allowed, the license doesn’t fail DFSG #3” but rather “since the GFDL only forbids modifications of secondary part, we can decide that the GFDL doesn’t fail DFSG #3”.

    All that said, unmodifiable sections still suck and that’s why I’ve put Adeato’s amendment first.


  1. Rank FD High

    Any compromise found from that insincere voting will not last long. Voting closes this weekend: if you have ranked FD below options you dislike, revote now if you want to find a real compromise. …