[top-politics] Re: Software initiatives

echarp — 2006-10-24 10:44:39+0200

> > But, during the vote, any single point of failure ought to be > > removed to strengthen the whole thing. Consensus is the tool of > > choice to agree on what the expressed decisions are (as is currently > > the case). > > Just like to add that what possibilities and rights a bord should have > ofcourse have to very clearly specified and it shall only be more or > less technical issues about the voting process in relation to the > design intent, nothing more.

The voting system has to be defined for the voters to cast their vote. Anybody willing should be able to follow that through and, with the votes database, be able to come to the correct results.

The voting procedure should be defined… democratically! ;)

> So his has to be written clearly in a regulation document, otherwise > this board soon enough will contain old politicians seeing a way of > preserving their own power instead of acting simple clerks, working > more as servants to the voters.

[top-politics] Re: Software initiatives

ketty . — 2006-10-23 17:05:22+0200

On 10/23/06, MG <magnusgus@gmail.com> wrote:

> -I’m fine with that and such parallell running can be used. > But. for legal resons we cannot see this as test environment once using > it for making laws etc. > In that stage there must be a core soft with as little bugs as > possible.

You seem to assume the bugs of a groups of parallel softwares is all bugs existing in any of the softwares. But infact the bugs of the group is only the ones existing in all softwares. Do you see why it makes total sence to use not one but a group of parallel software for a prodution environment (making laws)?

[top-politics] Re: Software initiatives

Serge — 2006-10-23 15:17:53+0200

Hey,

I think your disagreement (echard and Magnus) stems from the voting methods you advocate so that in essence I believe your arguments actually complement each other. In the case of AD’s continous voting approach, there is a need for a much more specific piece of soft to be integrated in the system as it determines the closing of a vote dynamically. This need for integration and crosstalk is absent from echarp’s approach as he is thinking of vote verification after the fact (if I am understanding both your arguments correctly), which indeed could be done with something as simple as an excel spreadsheet populated with the voting results.

One could imagine that different counting pieces of code could be certified by delegates handling voting methods in the system (trustworthiness solved, Magnus?) and used in parallel so as to ensure that should one counting system fail or be compromised, other systems may show the discrepancy and raise a red flag (satisfying mode of operation, echarp?). This way you would have the certification needed for results to be trustworthy while enjoying the added security of having several counting options, discrepancies in which would be a reasonnable basis for investigation into a possible manipulation or irregularity.

Best regards,

Serge

[top-politics] Re: Software initiatives

Serge — 2006-10-23 14:59:03+0200

Public voting records allow replicability of voting results in a simple format which in turn allows everyone to verify the vote count for themselves (and that their vote is what they did cast and hasn’t been altered)… This is simpe, straightforward and embodies the ideal of Transparent, Open and Public. Echarp I agree wholeheartedly with your approach.

Best regards,

Serge

[top-politics] Re: Software initiatives

echarp — 2006-10-23 13:03:04+0200

> You are misunderstanding.

Maybe. But I think I do understand your point. Just I disagree with it.

> I’m not against different initiatives and developments.

Cool.

[top-politics] Re: Software initiatives

Serge — 2006-10-23 02:48:30+0200

Why the added complexity of trustees, tokens, anonymity and whatever else? Not to mention linking one’s public statements to their actual vote doesn’t seem possible anymore in this setup (goodbye transparency).

Most agree that having someone’s real ID being linked to their voting record and posts etc would be a real danger to the openness of any debate. If everyone first weighs their contribution on the scale of how much trouble they can land themselves in, you can kiss whistleblowers and advocates of controversial causes goodbye. Which would be a terrible thing for any meaningful democratic debate to happen.

The only apparent alternative to secret voting imo is moving the secrecy onto one’s real ID. Going to full anonymity brings right back the problem of having a trail for trust-building so the logical solution would be in my opinion a unique and authenticated online ID. This way there is a clear trail of votes and opinions for everyone, which actually allows one to choose and change delegates – this is not possible with secret voting if we are talking of large scale system, in which it is assumed a large majority of participants will have the votes of others delegated to them for some issues so that saying that delegates’ records have to be public and not voters is saying that maybe 20% of all voters might enjoy a measure of anonymity (and these 20% will be the ones probably delegating everything and participate very little in the debates in the first place).

This seems far simpler and actually is transparent (which anonymity is not – by definition it’s a cloak). Also this discussion has been active in 3 threads (among which leparlement and security) so apologies if I’ve repeated some previously stated things.

Serge

[top-politics] Re: Software initiatives

ketty . — 2006-10-20 17:27:59+0200

On 10/20/06, echarp <emmanuel.charpentier@free.fr> wrote:

> Let me try to summarize: > * real life person is public, many secret actions > * real life person is secret, many public actions > > > And I don’t se why it would be easier. > > In the first case, there are many secrets for one person. In the second > there is only one secret per person. This is why I consider it simpler. > > Plus it allows to develop an online persona with a history and a > reputation.

I agree with this.

And i also agree that using only one software, even if the source of the software is public and tested by many, is a bad idea. Some bugs will not be found untill the have made their share of harm. The best way to protect against bugs is to simultanly use independent software.

[top-politics] Re: Software initiatives

ketty . — 2006-10-15 17:39:32+0200

On 10/15/06, MG <magnusgus@gmail.com> wrote:

> > > Sure, but you have to differ between individual votes in an particular > issue and representative voting in an particular issue. > On the individual level there can be threats stopping you from voting > at all, or on what you want. > This is seen all over the world in traditional elections so it’s a sad > fact.

And threats would be no issue for representatives? Since representatives can have so many more votes than a mere individual and there votes are not anonymous, it seems to me that it would be very effective to threat and otherwise trying to influence the votes of representatives. The only way i see around it is to have the identity of representatives anonymous, to refer to a recent discussion in the swedish forum =)

[top-politics] Re: Software initiatives

Magnus — 2006-10-24 12:18:10+0200

[top-politics] Re: Software initiatives

echarp — 2006-10-24 10:29:42+0200

> > > You could use a selected board for that if you want to :) > > See?

Of course you could, but it is bad engineering.

> > Or try to come to a consensus. > > > > As already happens in all voting offices. > > Is it? And what even if so, if we call a discussion of a badly written > or torn ballot a “software problem”, isn’t the people in that voting > office a trusted board?

They are not a trusted board.