[–] 26561862? 0 points 1 point 1 point (+1|-0) ago (edited ago)
From the crap on here, I'd guess Java is their language of choice. I had looked at some of their devs' linkedin accounts but I forgot what I saw there already. Bunch of Serbians. Stuff done under contract may or may not specify access to source code for review and audit, I'd guess no access given the mentality of the states that bought it. CM does that sort of thing for a living, and there's plenty of other people that could deconstruct the mechanics of the system. From descriptions it seems very enterprisey, shipping data into "cloud" and sending to Germany to turn into sausage. A voting system should be a closed loop, this looked very open with many points of attack.
Edit: lol, digging around on github, I found a psychic. Four years ago...
[–] 26560437? ago (edited ago)
The issue here is that floating point arithmetic is not perfectly reversible. The exponent and mantissa are both variable and there are numbers without perfectly round representations which lead to an increase in Epsilon as the number of operations increases. Yeah, we can create a "probable" reversal but with the number of adjustments I've seen it's likely raised Epsilon really high.
The question is how much tolerance for error does an election need? If we cannot accept a single vote error then Epsilon is a serious factor. If not, we'd have to make the estimation so that Epsilon was minimized.
[–] 26572441? 0 points 1 point 1 point (+1|-0) ago
It would be hilarious if they never thought to develop their own compiler and just got the idea from reading your post lmao