It's swisscheese and fractional voting is a distraction. From what I'm seeing the adjudication process is far more serious as it may allow backfill. Regardless the audit needs to first verify the books match, that what's on paper matches what came out of the machine. If you get beyond that point the audit isn't finished.
This. You're focused on a specific detail. Personally I would turn that feature off for a market where it's not relevant or perhaps ship a build without it. For a presidential election you would pull out all the stops.
To put it simply, if the system as a whole were properly orchestrated the fractional voting feature would be completely immaterial and impossible to pervert the results. The process would involve doublechecking, cross referencing. You usually have redundancy.
It's similar to running a shop in some ways. You might have your inventory on a spreadsheet but that's backed by the real inventory and you do stock counts to keep the systems aligned. If the voting system were properly setup then it would not simply be a process of counting the votes but it would also be self verifying with processes double checking.
Dr Shiva got one thing right which you might be skirting over. The system is a blackbox on multiple levels. That in itself means you have to have faith in the magic that goes on out of sight. The system in its entirety is unverifiable and unreliable from what I've seen. I wouldn't approve it for production.
When you run a system like a shop or warehouse, etc you don't just focus on the computer system but the system as a whole. How everything works, the people, how they interact with everything, the rules they follow, the whole system. DVS is only a piece of the puzzle and if you only focus on that you may never figure out how they really cheated.
It's not always necessary to figure out how sometimes though. If you've ever worked in gambling you don't necessarily detect cheaters by looking up their sleeves but instead you track who is winning too much (improbably). Shiva has tried to do this but does a bad job sometimes. He is subordinate to me because I have worked with all these systems. He's more of the studying type or someone who buries themselves in books.
Like, keep your eye on the lady - the rest is smoke and mirrors.
TBH - the way you don't answer a straight question, straight - you have to be a politician.
Or greek, and I don't trust either.
Redundancy in voting?
All of this stuff is your bag - carry it yourself!
I shouldn't be rude. I should say that I don't have to trust any of this, and can write it off as eye-glazing details.
Further, I think I could get enough support to support the idea of manual voting, or
some crypto solution, removing the need to trust in black-boxes on many levels.
We could pay for such a system with the savings of not spending hundreds of millions on systems that run Windows 2000 (SP 2). Systems that don't have wifi, but do. Systems that cost millions but, can be undermined by a $1 memory stick.
But, don't leave it up to me - I would see all of these clever people, hung, and publicly.
It cuts through the eye-glazing small-talk, and restores trust, all round.
Redundancy in records and processes. There's a pipeline.
Basically if I have a spreadsheet of all the stock and there's also the stock on the shelves those are two records that can be compared. The master record is the stock or in this case what people actually voted for. Everything else after that is derivative.
Verifying, auditing and analysing a system like this can get pretty boring, complex, etc but if you do, if you notice something like there's a point where the count goes downstream and can't be verified back upstream then you may have a problem.
You need to connect all the dots. Actually they should have a dummy system as a honey pot. On of the reasons for using the counting machines is speed. It makes things so efficient you can actually have two sets of people counting with two sets of machines from two suppliers. You then compare results. You also have human counters doing random sampling to further verify the results. All three processes happening in unison then you compare the results of each.
We can argue until the cows come home and that's by design. We're kept in the dark, there's a lack of transparency, another case of priority rape as they will say secrecy is a security measure and we can speculate forever but the point is that we don't know and that's by design.
That's a unified problem. What we're seeing is a show and what really goes on backstage is anyone's guess. That is the weakness of the system. We should have to same demand, to blow the lid of this thing so we can find out.
Their obstruction is sufficient to effectively prove they cheated or proof of the same remedy as if they did. At this point I'm not sure it matter's how anymore. A lot of what I'm talking about was relevant before, during and shortly after the election.
At this point we shouldn't give up but the chances of finding anything are rapidly dwindling as there's plenty of time to get rid of the evidence, a lot of it is perishable and the system has been rigged to allow fraud potentially without leaving evidence.
In regards to DVS, what the CEO said collaborating that witness who suspected people might have been counting multiple times is far more damning than the fractional voting. They basically said if someone does cheat in an election and their machines are involved they wash their hands clean of it. It's on the operator and nothing to do with them.
There's no such thing as a truly secure system but you can make it a lot harder and narrow the scope. They appear to have broadly designed a system not only with DVS but the entire election to make it easier to cheat and facilitate fraud. They did this with the mail in voting. They also cheated in more subtle ways like making it easier for people to vote in democrat areas.
DVS doesn't exactly make voting machines to deliberately assist in a specific kind of election fraud because they don't want the liability. Instead they just don't focus much on security. The CEO confirmed what that witness said. Their company policy, almost a mantra is "we turn a blind eye". That's the only way to do business in a banana republic but it's not compatible with a first world approach to democracy.
If you knew anything about security then the choice of OS would be of little relevance though I would use Linux personally. Windows does have a large share of the embedded market though which people don't appreciate. It's really annoying with drivers. Sometimes they're done well for windows but not for Linux. I've had to work on porting some drivers for peripherals and reverse engineering a little.
This. You're focused on a specific detail. Personally I would turn that feature off for a market where it's not relevant or perhaps ship a build without it. For a presidential election you would pull out all the stops.
To put it simply, if the system as a whole were properly orchestrated the fractional voting feature would be completely immaterial and impossible to pervert the results. The process would involve doublechecking, cross referencing. You usually have redundancy.
It's similar to running a shop in some ways. You might have your inventory on a spreadsheet but that's backed by the real inventory and you do stock counts to keep the systems aligned. If the voting system were properly setup then it would not simply be a process of counting the votes but it would also be self verifying with processes double checking.
Dr Shiva got one thing right which you might be skirting over. The system is a blackbox on multiple levels. That in itself means you have to have faith in the magic that goes on out of sight. The system in its entirety is unverifiable and unreliable from what I've seen. I wouldn't approve it for production.
When you run a system like a shop or warehouse, etc you don't just focus on the computer system but the system as a whole. How everything works, the people, how they interact with everything, the rules they follow, the whole system. DVS is only a piece of the puzzle and if you only focus on that you may never figure out how they really cheated.
It's not always necessary to figure out how sometimes though. If you've ever worked in gambling you don't necessarily detect cheaters by looking up their sleeves but instead you track who is winning too much (improbably). Shiva has tried to do this but does a bad job sometimes. He is subordinate to me because I have worked with all these systems. He's more of the studying type or someone who buries themselves in books.
I am, with a laser-sight!
Like, keep your eye on the lady - the rest is smoke and mirrors.
TBH - the way you don't answer a straight question, straight - you have to be a politician.
Or greek, and I don't trust either.
Redundancy in voting?
All of this stuff is your bag - carry it yourself!
I shouldn't be rude. I should say that I don't have to trust any of this, and can write it off as eye-glazing details.
Further, I think I could get enough support to support the idea of manual voting, or some crypto solution, removing the need to trust in black-boxes on many levels.
We could pay for such a system with the savings of not spending hundreds of millions on systems that run Windows 2000 (SP 2). Systems that don't have wifi, but do. Systems that cost millions but, can be undermined by a $1 memory stick.
But, don't leave it up to me - I would see all of these clever people, hung, and publicly.
It cuts through the eye-glazing small-talk, and restores trust, all round.
Redundancy in records and processes. There's a pipeline. Basically if I have a spreadsheet of all the stock and there's also the stock on the shelves those are two records that can be compared. The master record is the stock or in this case what people actually voted for. Everything else after that is derivative.
Verifying, auditing and analysing a system like this can get pretty boring, complex, etc but if you do, if you notice something like there's a point where the count goes downstream and can't be verified back upstream then you may have a problem.
You need to connect all the dots. Actually they should have a dummy system as a honey pot. On of the reasons for using the counting machines is speed. It makes things so efficient you can actually have two sets of people counting with two sets of machines from two suppliers. You then compare results. You also have human counters doing random sampling to further verify the results. All three processes happening in unison then you compare the results of each.
We can argue until the cows come home and that's by design. We're kept in the dark, there's a lack of transparency, another case of priority rape as they will say secrecy is a security measure and we can speculate forever but the point is that we don't know and that's by design.
That's a unified problem. What we're seeing is a show and what really goes on backstage is anyone's guess. That is the weakness of the system. We should have to same demand, to blow the lid of this thing so we can find out.
Their obstruction is sufficient to effectively prove they cheated or proof of the same remedy as if they did. At this point I'm not sure it matter's how anymore. A lot of what I'm talking about was relevant before, during and shortly after the election.
At this point we shouldn't give up but the chances of finding anything are rapidly dwindling as there's plenty of time to get rid of the evidence, a lot of it is perishable and the system has been rigged to allow fraud potentially without leaving evidence.
I'm not arguing - not at these rates!
I'm saying that, as sophistication grows, trust diminishes.
I could deliver a secure system, for half the price, and that's bona-fide. Non-citizens, dead voters, maiden-name voting, oversees, all considered.
But I don't get invited to those meetings, and I don't specify the systems, and I don't get told the real requirement, just how it should present.
Secure voting, isn't what they ordered - they got what they ordered and now, we must all live by it!
In regards to DVS, what the CEO said collaborating that witness who suspected people might have been counting multiple times is far more damning than the fractional voting. They basically said if someone does cheat in an election and their machines are involved they wash their hands clean of it. It's on the operator and nothing to do with them.
There's no such thing as a truly secure system but you can make it a lot harder and narrow the scope. They appear to have broadly designed a system not only with DVS but the entire election to make it easier to cheat and facilitate fraud. They did this with the mail in voting. They also cheated in more subtle ways like making it easier for people to vote in democrat areas.
DVS doesn't exactly make voting machines to deliberately assist in a specific kind of election fraud because they don't want the liability. Instead they just don't focus much on security. The CEO confirmed what that witness said. Their company policy, almost a mantra is "we turn a blind eye". That's the only way to do business in a banana republic but it's not compatible with a first world approach to democracy.
If you knew anything about security then the choice of OS would be of little relevance though I would use Linux personally. Windows does have a large share of the embedded market though which people don't appreciate. It's really annoying with drivers. Sometimes they're done well for windows but not for Linux. I've had to work on porting some drivers for peripherals and reverse engineering a little.