Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Debian decides to allow secret votes

lwn.net

Post reply on HN
Print view View original

221–230 of 280 posts

Re: Debian decides to allow secret votes

#221

Post by kelnos »

toast0 wrote: post #92
spondyl wrote: post #9

A brief summary because I had to re-read this a few times to figure out what was going on. Debian had a vote with 4 options: Option 1 "Hide identities of Developers casting a particular vote" Option 2 "Hide identities of Developers casting a particular vote and allow verification" Option 3 "Reaffirm public voting" Option 4 "None of the above" (NOTA) The objection raised here is that Option 3 and Option 4 can be consi…

> If those two were represented as a single "NOTA" option, that option would have won the ballot in terms of votes. If the ballot were different, the results would be too. You can't take the results of one ballot and mush them around to support a hypothetical result; see also counting of the 'popular vote' for US president, a ballot which has never been tried, and whose results are unknowable. A vote for 4 says (to m…

> A vote for 4 says (to me, not a qualified voter) I don't like the current voting, and hiding identities is also undesirable; if this choice is selected, the status quo must be continued until a new alternative is voted upon, but there's clearly desire to do so.

That's a difference without distinction, though. If option 4 were to win, then, what would they then do? The exact same thing as if option 3 were to win -- keep the status quo -- no? The idea that an option 4 win might also get people to consider other voting systems isn't really relevant; you don't vote to consider things that you will then put to vote. You just consider them, and then have a vote. So the "legal" outcome of this vote would be identical if either option 3 or 4 had won.

IMO the DPL is correct that this election was constructed poorly: they should have had either option 3 or 4 in there, not both. But I think that might just be a "too bad" situation: since there are no provisions in the constitution for a "do over", I think the only reasonable thing to do is to accept the results of the election, change the voting system, and then propose a new vote to potentially change it back, but this time more carefully consider the options on the ballot.

Unfortunately, because of the supermajority requirement, this probably won't work as intended. Just because options 3+4 on the current vote (had they been combined into a single option) may have been enough to cause the supermajority requirement to fail, it doesn't mean that there will be a supermajority (in a new vote) to switch back to public voting.

I think they screwed up, but in order to preserve confidence in their system, they'll have to suck it up and live with it.

Re: Debian decides to allow secret votes

#222

Post by throwawaybutwhy »

layer8 wrote: post #219
charcircuit wrote: post #24

Earlier quoted context omitted.

By definition "none of the above" can't be the same as any of the above options.

Option 1 is also a superset of option 2, strictly speaking. The whole set of options doesn’t really seem well-defined.

...which is working as intended. This situation has all the hallmarks of a silent coup in the making.

Re: Debian decides to allow secret votes

#223

Post by kelnos »

asiachick wrote: post #216
LeifCarrotson wrote: post #193

Earlier quoted context omitted.

Stated differently, this makes states matter and rural voters matter in direct proportion to the number of people living there, because all people are equal. It's not right that a voter in Maine or New Hampshire, which have 1.3 million people and are roughly even between the two major parties, has a high probability of changing the outcome with their vote, while a voter in California or Texas - which have 30 times as…

This isn't so different in the EU is it? Smaller countries don't want their needs ignored by the bigger countries so they get more seats relative to their population than big countries. Maybe it's hard to imagine the United States of American to become 50 independent countries but isn't the way you keep them united is by promising the smaller ones that they won't be trampled on by the big ones?

> isn't the way you keep them united is by promising the smaller ones that they won't be trampled on by the big ones?

Yes, but in the US, at least, it's gotten to the absurd opposite where the minority is more or less assured a majority of votes. In order to avoid tyranny of the majority, we've more or less codified tyranny of the minority.

Re: Debian decides to allow secret votes

#224

Post by Gollapalli »

blueflow wrote: post #130
sjmm1989 wrote: post #118

Earlier quoted context omitted.

Stopped using Debian a long time ago, and this is just making me glad I did exactly that. What kind of hairbrained nimrod came up with that idea!? You can't have mixed voting methods and still call it fair in any respect. Any chance there could be tampering should be considered not just to be happening, but to be assured to happen. "Secret ballots" of any sort when everyone else is not partaking in said "Secret ballo…

I stand behind Debian to do votes in secrecy. During the RMS issue, I've seen many people who did not have their ferocity under control. This went as far as people wanting to have the "others" removed from the Debian project. People not being able to cast their vote without having to fear repercussions is justification enough for secret votes.

Is there a good rundown over what people were actually getting mad about with RMS? RMS has done a lot for free software, and while I think the whole GNU/Linux naming pasta is a bit silly, what RMS did with emacs, gcc, and copyleft is genuinely heroic in terms of accomplishment and impact.

So far as I can tell, it's "RMS defended Marvin Minsky" and "RMS said something about sex with teenagers not being super terrible many years ago".

Also, was the Minsky thing related to Epstein? Because if so, it's hard to blame Minsky for falling victim to a international intelligence operation that also ensnared presidents, financiers, and movie moguls, all of whom are far more shrewd and worldly than any mathematician has a right to be. And the right thing would be to blame the foreign power that was doing the ensnaring, IMHO.

Re: Debian decides to allow secret votes

#225

Post by kevinpet »

Sesse__ wrote: post #29
dtech wrote: post #19

Earlier quoted context omitted.

They use ranked choice voting so this shouldn't matter, as far as I understand RCV eliminates the lowest option one-by-one, passing the votes to the next option on each ballot as if the eliminated option never existed. The discussion seems to be about ballots with 3 > 2 > 4, and when 3 was eliminated the votes went to 2, allowing it to pass. That's a bit strange because you'd expect a ballot to always have 3 next to…

Debian's vote system does not use instant runoff voting (which is eliminating the lowest options one by one). Instead, it uses Condorcet voting, which picks the options defeating all the others in pairwise competition. (If no such option exists, it gets more complicated. Also, the NOTA option is special and is used to check majority or supermajority requirements before the actual Condorcet process starts.)

Thank you. This is the first comment that clearly explained the issue.

Re: Debian decides to allow secret votes

#226

Post by techsupporter »

treeman79 wrote: post #187
dagw wrote: post #163

Earlier quoted context omitted.

or that people would making a different vote if it was counted differently? Two changes would happen. A lot of people who don't vote today because their state either heavily favours the 'wrong' candidate or the 'right' candidate will already win by a massive margin, would start voting, since all of sudden their vote matters a lot more. Secondly, and more importantly, candidates would campaign very differently. In tod…

This also makes many states cease to matter. Already a big issue that big cities vote one way while rural voters go another way.

> This also makes many states cease to matter.

States are abstract political entities; it's people who matter.

People are represented in the House by population (less so than they should be since we refuse to let the lower chamber grow in size) and in the Senate by state.

The President of the United States represents every United States citizen everywhere on the planet. The holder of that office should be elected by a majority, preferably with some form of expression of ranking preference, of those voters regardless of what political entity they are currently in or connected to.

Re: Debian decides to allow secret votes

#227

Post by qiskit »

the_lonely_road wrote: post #4

Apparently by a single vote and the legitimacy of the vote is in question due to some technicality I didn't care to reread until I understood. What I don't understand is why the project is NOT already secret voting? Do not most free democratic countries practice 'secret ballot' voting specifically because not doing so caused all kinds of problems like voter intimidation? I googled the first three that came to mind (f…

> Do not most free democratic countries practice 'secret ballot' voting specifically because not doing so caused all kinds of problems like voter intimidation?

Because the voting developers are part of "governance". They are like senators. Nobody allows secret votes in government. When senators vote, their votes are public for many reasons.

Re: Debian decides to allow secret votes

#228

Post by michaelmrose »

Debian is apparently bad at democracy despite putting a lot of thought into the matter. The entire problem has to do with conflating 3 questions with boolean answers as 3 sets of answers to those 3 questions.

1. Should we have a vote to amend yes no higher threshold to pass.

2. If yes should we hide identities of developers casting a vote yes no

3. If yes should we allow verification yes no

Arguably 3 is a ridiculous question so ought to be safely omitted.

Arguably it would be reasonable to require more than 10% to cast a vote as well.

All we know now is that a single digit percentage of Debian developers prefer to vote anon.

Given the low turnout there is little reason not to respect the present result but have another smarter vote.

Re: Debian decides to allow secret votes

#229

Post by mariusor »

danuker wrote: post #18
the_lonely_road wrote: post #4

Apparently by a single vote and the legitimacy of the vote is in question due to some technicality I didn't care to reread until I understood. What I don't understand is why the project is NOT already secret voting? Do not most free democratic countries practice 'secret ballot' voting specifically because not doing so caused all kinds of problems like voter intimidation? I googled the first three that came to mind (f…

I am not familiar with Debian's practices, but electronic voting can't work for a nation because: - centralization (there must be a central, corruptible place where the voters are authenticated or the votes are counted) - software is untrustable: https://www.win.tue.nl/%7Eaeb/linux/hh/thompson/trust.html The USA has had a Diebold voting scandal. Other countries are using a paper-based voting process which can be supe…

I do not think that we have investigated electronic voting enough to be able to say it's not feasible with such certainty. I wouldn't start with national high stakes ballots, but I'm confident that if enough smart people put their mind to it, a solution is possible.

Re: Debian decides to allow secret votes

#230

Post by mariusor »

grumbel wrote: post #51
charcircuit wrote: post #27

Earlier quoted context omitted.

Electronic voting systems using zero knowledge proofs are superior to paper votes because it's possible to check for yourself that your vote was actually counted and not ignored. Paper voting is extremely expensive to scale compared to a website that lets you vote. Everyone could get a notification on their phone when they are asked to vote on something as opposed to having to fill out paper and send it somewhere to…

> it's possible to check for yourself that your vote was actually counted and not ignored. That's a bug, not a feature. The point of not doing that with paper voting is that it makes selling your vote difficult, as nobody else can verify what you voted for. You on the other side know that you put the ballot in the box and can stay around to see if the votes in the box get accurately counted. With electronic voting yo…

> That's a bug, not a feature.

I think you're wrong. A bug would be if you could tell how a vote was cast, not if it was counted.

Post reply on HN