starting a dictature with a visibly incompetent ruler is a strange way to do that. Maybe they hope a litteral angel will come from the sky to replace Trump ?
I don’t think they realize quite how incompetent he is. And as for the second part, a significant number of Christian conservatives appear to believe that Trump was sent by God, so they seem to think he is the angel.
Trump’s very quickly trying to crack through that floor.
Weird (and tired of trumplings) Eddie
4 years ago
@ Ohlmann:
Since you’re employed in the geek industry, maybe you can answer this:
Is it true that software engineers work under a 3-part business plan;
1) make it different than it was yesterday
2) get it on the market before noon today
3) don’t worry if it doesn’t work, we’ll have a new one out before anyone can complain
???
Ohlmann
4 years ago
Weird Eddie : some corporations definitely do any or all of thoses three, but not all of them.
1) is more a marketing strategy than an actual policy, even if it can happen. Typically, to attract investors and/or customers, the marketing will overhype a new addition or ask for the 12th reskin of the year, but the engineers themselves tend to prefer to stay tried and true.
2) is probably the most common, especially in videogames. That being said, “release date : when it’s done” is a real strategy of a lot of corporations, and not just the old Blizzard slogan. Also, sectors who use this a lot rely either on having a fresh stream of unexperienced rookies (videogames) or overpaying their engineers (financial tech), because software engineers can and will walk out of a job if there is no proper compensation for 12+h days.
3) is sometime an actually justified way to engineer a software, sometime a cynical way to pretend to have accomplished objectives, often not done at all.
Let me explain a bit. When your software is easy to replace and failure aren’t costly, pushing a version, getting informations from what work and not from the users, and fixing it is faster than properly test everything in house, because customers often have the wackier use case and also are a lot more numerous and less costly than testers. That’s typically the case for web sites ; Facebook don’t kill anybody when it malfunction and it has a ton of functionality, half of them I am sure even the engineer don’t know about.
The videogame industry have more problem pushing fix and getting feedbacks, but cynically often use its customers as testers because budgets are tights, and customers actually don’t walk out that much when the game is somewhat broken. It can backfire hard, but it often work. In addition to that, because of how games are made, managers have somewhat more incentives staying on budget and on schedule than actually doing good games, because being unsuccessful can be pinned on other factors, but being out of budget alway mean someone fucked up. (very probably the manager not reserving enough time to actually build everything).
(note that while the industry is fucked up, a lot of people in it are both very skilled and very cool ; you can safely assume most of them take pride in trying to publish good games… But they also have good reason to rush a project or squeeze employees hard)
A ton of software however is made to just work, and do work. There is a lot of practices and tool to make reliable softwares, and thoses softwares landed rovers on Mars, predict weather, route your favorite site to your computer, all of that silently. The Boeing 737 MAX shitshow was particulary shocking for software engineers like me because they should do very reliable software, have done some of the most reliable software in the past, and allowed themselves to slip out of their standard and kill hundred of people.
Voting tech is very badly seen because the guys doing thoses software have 0 accountability (nobody have ever been fired because they fucked up an election), their softwares are basically impossible to audit because they aren’t open sources, and the corporations doing that neither come from background that let one hope for reliability, nor advertise themselve as being super reliable. That is coupled to an election being sneakily a relatively hard thing to do mean I would stay very far away from any voting machine.
Surplus to Requirements, Observer of the Vast Blight-Wing Enstupidation
4 years ago
@Ohlmann:
That being said, “release date : when it’s done” is a real strategy of a lot of corporations, and not just the old Blizzard slogan.
It is possible to err too far in that direction, though *cough*Half-Life 2: Episode 3*cough*.
Moggie
4 years ago
@Ohlmann:
Voting tech is very badly seen because the guys doing thoses software have 0 accountability (nobody have ever been fired because they fucked up an election), their softwares are basically impossible to audit because they aren’t open sources, and the corporations doing that neither come from background that let one hope for reliability, nor advertise themselve as being super reliable. That is coupled to an election being sneakily a relatively hard thing to do mean I would stay very far away from any voting machine.
In the case of voting machines, I wouldn’t trust them even if the software were open source and audited. How do you know that the software running on the machine exactly matches the published source? And Ken Thompson’s paper Reflections on Trusting Trust is as applicable today as it was 35 years ago.
Jenora Feuer
4 years ago
@Moggie:
Considering that some people I do trust on matters of security photographed themselves standing next to a bunch of unguarded voting machines and had diagrams of the locks showing how easily they could be opened if you had the right equipment… the ‘how do you know that the software running on the machine’ part is definitely a thing, and doesn’t even require malfeasance by the company making the voting machines, just incompetence.
Of course, malfeasance is definitely still a possibility. In a less societally critical case, there was a big scandal in Quebec several years back where restaurants were deliberately faking their sales numbers to reduce the amount of tax they’d have to pay to the government. (A restaurant chain founded by Celine Dion was involved, making it even more of a media circus.) During the investigation, it was found that some of these restaurants had custom ‘zapper’ software on their POS systems which could help with that, where you could specify how much you wanted to skim off (in terms of dollar amount or percentage) and it would scan through the recorded data in the database to remove just the right amount. This rather obviously required significant inside information on how the data was stored, and strongly suggested that the companies making the POS software had somebody doing a nice little side business selling ways to get around that.
(This is why every food receipt in Quebec now has a 2D barcode printed at the bottom of it, to get some tracking information to make it a lot harder to blank some information out from the middle of the data.)
Naglfar
4 years ago
@Jenora Feuer
Considering that some people I do trust on matters of security photographed themselves standing next to a bunch of unguarded voting machines and had diagrams of the locks showing how easily they could be opened if you had the right equipment… the ‘how do you know that the software running on the machine’ part is definitely a thing
There’s a significant amount of evidence that there was voting machine tampering and/or malfunction in the 2018 Georgia gubernatorial election. In that election there were a few of machines in precincts that otherwise voted strongly for Abrams (D) but those machines had mostly votes for Kemp (R). It’s possible it was an accident, but seeing how that election was I suspect it could have been more deliberate.
Solution: use paper ballots
Hypatia's Daughter
4 years ago
Naglfar
I think it’s more that they actually want an end to democracy. They want authoritarian fascism. Maybe some are simply blind, but I think many actively want this.
And, yet, no matter how much they cheer on the inception of a fascist state, no “average” person living in one has liked it or fared well under it; and every fascist state has gone down in flames, usually within a lifetime or two.
It takes a few years, then the “average” person to realizes that the boots of their masters aren’t only on the necks of their “enemies”, but on their necks, too.
Do these people actually want to live in N Korea or Russia?
If Trump’s avoids conviction in the Senate and wins the election, he will consider himself above the need to cater to any voting bloc (He’ll either quit after his second term, or just remain in office past his term.) If that happens, does anyone seriously think he will care about the wishes of a whiny bunch of religious fanatics if he doesn’t need their votes?
Tohka
4 years ago
All the current trump news is rallying up misogynists and bigots in general to come out of the woodwork like insects to cheer and energize themselves. The whole thing is a mess…
One good news was that now that one place like reddit mgtow is almost gone, I hope other places do the same too like youtube. I noticed it’s always their same type of comments directed at women. The type of comments you’d expect by now from them. Then you check out the account that’s commenting and they’re always following some mgtow type channel. It’s expected that these can’t function without women’s attention and knowing that women exist.
kupo
4 years ago
I wouldn’t trust them even if the software were open source and audited
I can’t think of anything worse than open source software for voting. Who the hell would validate that the code being added isn’t malicious? Someone would need to be in charge of merging those changes. And what if polling station 134 is on v1.2.543 and polling station 328 hasn’t updated since v1.0.45? And from how badly the polling stations were run in 2016 that’s the *best* case. And what if someone pulls their code in protest, like with that ruby gem ICE protest? It doesn’t even have to be the polling software that gets pulled down, it could be literally any dependency.
…you’re giving me nightmares, Moggie! ?
kupo
4 years ago
OT:
Kirk Douglass died, so of course women are getting mobbed on Twitter for daring to point out that he allegedly raped Natalie Wood.
Jenora Feuer: while that high-tech trick is dead, there’s still the low-tech trick of taking cash and making change without involving the cash register. Then there’s no receipt to fake.
Switching to mostly credit cards is probably making a huge improvement for accuracy in taxation.
Definitely not Steve
4 years ago
@Ohlmann and Catalpa,
Thanks for the suggestions. I’ve been having social anxiety through the roof this week, and it’s been tough for me to predict whether I’ll feel up to getting out of the apartment on any particular day, so I might have a slow time getting involved in anything. I do at least have a few hobbies I would like to start up (tabletop RPGs being one of them), and I’d expect there to be lots of hobby groups in this city, so maybe as my brain calms down I can get out more and start going to something regularly.
This is also a city with a lot of political activity, so I might get involved in that in some way, but my job is also somewhat political so I don’t know if that would be a bit too much for me.
@Numerobis,
I like the calendar idea! I might try that and see if it eases my anxiety about leaving my apartment.
numerobis
4 years ago
Total coincidence, news about receipts in Germany. Where a bakery that’s unhappy about needing to issue receipts is making receipts out of fondant, on a pastry.
In the German state of Thüringen, the Minister President has been chosen and has accepted the vote. He belongs to the FPD (Liberals) but was voted in largely by the AfD (the far-right “Alternative for Germany”).
How could he?!
Why did the Christian Democrats – Merkel’s party – put their voting weight behind the same candidate that AfD supported?
Accepting legitimation from the far right also legitimises the far right.
The alternative candidate was being supported by a left-green-social democrat coalition. He is an excellent politician who has succeeded in proving the left party “die Linke” to be votable.
The complete story is even more disgusting.
Ohlmann
4 years ago
@kupo : how is that any better with closed source software ? You have all the same problem (litteraly everything you describe also happen with closed software), *and* nobody can tell if someone tampered with the program.
Making it very hard to tamper a machine without that being detected is doable (even tho that mean people using the tampered machines will be ignored, and of course there is an aspect of technological arm race here). The fact there is basically 0 security on that end is one of the big proof the guy doing that aren’t serious. A standard telephone have better security against tampering than a voting machine from what I have seen.
Ooglyboggles
4 years ago
Days later and we still have no idea who won at Iowa Caucus. What a trainwreck. Seriously Nevada needs to switch to some paper voting and counting instead of relying on a different app.
Specialffrog
4 years ago
The type of voting machine used in Bolivia and Venezuela sounds reasonable. You select an entry on a machine, it prints a ballot, you verify it and then put it in the box. The paper ballots are counted as the official result but you have a quick unofficial count from the machines.
I believe this is the mechanism many election monitoring orgs recommend.
Though as Bolivia demonstrates bad actors can still claim that the results are fraudulent under this system.
Naglfar
4 years ago
@Daughter
It’s quite frustrating that the center has allied itself so much with the far right. This is a dangerous progression. In America a sort of inverse has happened, where many people who called themselves centrists and were registered as Democrat became hard right Trump supporters. I knew several people who did that.
Diego Duarte
4 years ago
@Specialfrog
We have the same mechanism down in Peru as well and it seems to work just fine. The problem is how the votes are reported though, because your candidate can win a precinct, but the person in charge of reporting can report otherwise. At least there is a paper trail if there is a need for recount though.
Weird (and tired of trumplings) Eddie
4 years ago
@ Ohlmann;
… for want of a “/s”, a discussion thread was lost…. 🙁
Although I DO have the worst attitude towards computer technology of anyone I know, I was being sarcastic.
Weird (and tired of trumplings) Eddie
4 years ago
… for want of a discussion thread, an idea was lost….
… for want of an idea, an employee was lost….
… for want of an employee a business was lost….
… for want of a business an economy was lost….
… for want of an economy a civilization was lost….
all for the want of a sarcasm font….
I’ll hafta remember that.
Ohlmann
4 years ago
At that point, I am not entirely sure to what you point, but generally speaking, when Poe’s law start to show its head, then something bad is happening.
And something bad is happening in computer engineering right now 🙂
(I am reasonably high on personal responsability for software engineers. Like, do not work for Amazon or Facebook ever, and think about the consequences of what you code, be it a facial recognition tool or a navigation help for a plane)
(also, when I am in doubt, I try to err on the side of people being non sarcastical, because I find it much less awkward to answer straightfully to a sarcasm than counter-sarcasting someone who actually wanted to help)
kupo
4 years ago
how is that any better with closed source software ? You have all the same problem (litteraly everything you describe also happen with closed software), *and* nobody can tell if someone tampered with the program.
Because usually closed source software has an organized team running it and not just anyone can contribute and there’s some level of accountability for the person accepting and merging changes (plus everyone else down the line). That person is also usually the most experienced or an architect or something to that effect. With open source it’s whoever created the repository or whoever they gave the rights to.
It’s basically professionals vs. hobbyists. Would you want to fly in an airplane built by hobbyists? I once had a guest lecturer in college and he was a software engineer who built cars as a hobby and he thought agile was the best method for building a car because he could build a whole car in a weekend. He shared photos. It had a styrofoam body and he had never safety tested it and who knows what the emissions were like. Software hobbyists take that same level of care with open source code.
I’m not saying there aren’t some great open source libraries out there. But those are the exception.most of the time it’s people putting a passion project out there and their passion is not secure or efficient code.
We Hunted The Mammoth
@Ohlmann
I don’t think they realize quite how incompetent he is. And as for the second part, a significant number of Christian conservatives appear to believe that Trump was sent by God, so they seem to think he is the angel.
Trump’s very quickly trying to crack through that floor.
@ Ohlmann:
Since you’re employed in the geek industry, maybe you can answer this:
Is it true that software engineers work under a 3-part business plan;
1) make it different than it was yesterday
2) get it on the market before noon today
3) don’t worry if it doesn’t work, we’ll have a new one out before anyone can complain
???
Weird Eddie : some corporations definitely do any or all of thoses three, but not all of them.
1) is more a marketing strategy than an actual policy, even if it can happen. Typically, to attract investors and/or customers, the marketing will overhype a new addition or ask for the 12th reskin of the year, but the engineers themselves tend to prefer to stay tried and true.
2) is probably the most common, especially in videogames. That being said, “release date : when it’s done” is a real strategy of a lot of corporations, and not just the old Blizzard slogan. Also, sectors who use this a lot rely either on having a fresh stream of unexperienced rookies (videogames) or overpaying their engineers (financial tech), because software engineers can and will walk out of a job if there is no proper compensation for 12+h days.
3) is sometime an actually justified way to engineer a software, sometime a cynical way to pretend to have accomplished objectives, often not done at all.
Let me explain a bit. When your software is easy to replace and failure aren’t costly, pushing a version, getting informations from what work and not from the users, and fixing it is faster than properly test everything in house, because customers often have the wackier use case and also are a lot more numerous and less costly than testers. That’s typically the case for web sites ; Facebook don’t kill anybody when it malfunction and it has a ton of functionality, half of them I am sure even the engineer don’t know about.
The videogame industry have more problem pushing fix and getting feedbacks, but cynically often use its customers as testers because budgets are tights, and customers actually don’t walk out that much when the game is somewhat broken. It can backfire hard, but it often work. In addition to that, because of how games are made, managers have somewhat more incentives staying on budget and on schedule than actually doing good games, because being unsuccessful can be pinned on other factors, but being out of budget alway mean someone fucked up. (very probably the manager not reserving enough time to actually build everything).
(note that while the industry is fucked up, a lot of people in it are both very skilled and very cool ; you can safely assume most of them take pride in trying to publish good games… But they also have good reason to rush a project or squeeze employees hard)
A ton of software however is made to just work, and do work. There is a lot of practices and tool to make reliable softwares, and thoses softwares landed rovers on Mars, predict weather, route your favorite site to your computer, all of that silently. The Boeing 737 MAX shitshow was particulary shocking for software engineers like me because they should do very reliable software, have done some of the most reliable software in the past, and allowed themselves to slip out of their standard and kill hundred of people.
Voting tech is very badly seen because the guys doing thoses software have 0 accountability (nobody have ever been fired because they fucked up an election), their softwares are basically impossible to audit because they aren’t open sources, and the corporations doing that neither come from background that let one hope for reliability, nor advertise themselve as being super reliable. That is coupled to an election being sneakily a relatively hard thing to do mean I would stay very far away from any voting machine.
@Ohlmann:
It is possible to err too far in that direction, though *cough*Half-Life 2: Episode 3*cough*.
@Ohlmann:
In the case of voting machines, I wouldn’t trust them even if the software were open source and audited. How do you know that the software running on the machine exactly matches the published source? And Ken Thompson’s paper Reflections on Trusting Trust is as applicable today as it was 35 years ago.
@Moggie:
Considering that some people I do trust on matters of security photographed themselves standing next to a bunch of unguarded voting machines and had diagrams of the locks showing how easily they could be opened if you had the right equipment… the ‘how do you know that the software running on the machine’ part is definitely a thing, and doesn’t even require malfeasance by the company making the voting machines, just incompetence.
Of course, malfeasance is definitely still a possibility. In a less societally critical case, there was a big scandal in Quebec several years back where restaurants were deliberately faking their sales numbers to reduce the amount of tax they’d have to pay to the government. (A restaurant chain founded by Celine Dion was involved, making it even more of a media circus.) During the investigation, it was found that some of these restaurants had custom ‘zapper’ software on their POS systems which could help with that, where you could specify how much you wanted to skim off (in terms of dollar amount or percentage) and it would scan through the recorded data in the database to remove just the right amount. This rather obviously required significant inside information on how the data was stored, and strongly suggested that the companies making the POS software had somebody doing a nice little side business selling ways to get around that.
(This is why every food receipt in Quebec now has a 2D barcode printed at the bottom of it, to get some tracking information to make it a lot harder to blank some information out from the middle of the data.)
@Jenora Feuer
There’s a significant amount of evidence that there was voting machine tampering and/or malfunction in the 2018 Georgia gubernatorial election. In that election there were a few of machines in precincts that otherwise voted strongly for Abrams (D) but those machines had mostly votes for Kemp (R). It’s possible it was an accident, but seeing how that election was I suspect it could have been more deliberate.
Solution: use paper ballots
Naglfar
And, yet, no matter how much they cheer on the inception of a fascist state, no “average” person living in one has liked it or fared well under it; and every fascist state has gone down in flames, usually within a lifetime or two.
It takes a few years, then the “average” person to realizes that the boots of their masters aren’t only on the necks of their “enemies”, but on their necks, too.
Do these people actually want to live in N Korea or Russia?
If Trump’s avoids conviction in the Senate and wins the election, he will consider himself above the need to cater to any voting bloc (He’ll either quit after his second term, or just remain in office past his term.) If that happens, does anyone seriously think he will care about the wishes of a whiny bunch of religious fanatics if he doesn’t need their votes?
All the current trump news is rallying up misogynists and bigots in general to come out of the woodwork like insects to cheer and energize themselves. The whole thing is a mess…
One good news was that now that one place like reddit mgtow is almost gone, I hope other places do the same too like youtube. I noticed it’s always their same type of comments directed at women. The type of comments you’d expect by now from them. Then you check out the account that’s commenting and they’re always following some mgtow type channel. It’s expected that these
can’t function without women’s attention and knowing that women exist.
I can’t think of anything worse than open source software for voting. Who the hell would validate that the code being added isn’t malicious? Someone would need to be in charge of merging those changes. And what if polling station 134 is on v1.2.543 and polling station 328 hasn’t updated since v1.0.45? And from how badly the polling stations were run in 2016 that’s the *best* case. And what if someone pulls their code in protest, like with that ruby gem ICE protest? It doesn’t even have to be the polling software that gets pulled down, it could be literally any dependency.
…you’re giving me nightmares, Moggie! ?
OT:
Kirk Douglass died, so of course women are getting mobbed on Twitter for daring to point out that he allegedly raped Natalie Wood.
https://twitter.com/JanieGotHerGun/status/1225205866405777408?s=19
This world is too cruel to continue existing.
Jenora Feuer: while that high-tech trick is dead, there’s still the low-tech trick of taking cash and making change without involving the cash register. Then there’s no receipt to fake.
Switching to mostly credit cards is probably making a huge improvement for accuracy in taxation.
@Ohlmann and Catalpa,
Thanks for the suggestions. I’ve been having social anxiety through the roof this week, and it’s been tough for me to predict whether I’ll feel up to getting out of the apartment on any particular day, so I might have a slow time getting involved in anything. I do at least have a few hobbies I would like to start up (tabletop RPGs being one of them), and I’d expect there to be lots of hobby groups in this city, so maybe as my brain calms down I can get out more and start going to something regularly.
This is also a city with a lot of political activity, so I might get involved in that in some way, but my job is also somewhat political so I don’t know if that would be a bit too much for me.
@Numerobis,
I like the calendar idea! I might try that and see if it eases my anxiety about leaving my apartment.
Total coincidence, news about receipts in Germany. Where a bakery that’s unhappy about needing to issue receipts is making receipts out of fondant, on a pastry.
https://www.npr.org/2020/02/04/802669731/germany-has-a-new-receipt-law-and-bakeries-are-getting-sweet-revenge
In the German state of Thüringen, the Minister President has been chosen and has accepted the vote. He belongs to the FPD (Liberals) but was voted in largely by the AfD (the far-right “Alternative for Germany”).
How could he?!
Why did the Christian Democrats – Merkel’s party – put their voting weight behind the same candidate that AfD supported?
Accepting legitimation from the far right also legitimises the far right.
The alternative candidate was being supported by a left-green-social democrat coalition. He is an excellent politician who has succeeded in proving the left party “die Linke” to be votable.
The complete story is even more disgusting.
@kupo : how is that any better with closed source software ? You have all the same problem (litteraly everything you describe also happen with closed software), *and* nobody can tell if someone tampered with the program.
Making it very hard to tamper a machine without that being detected is doable (even tho that mean people using the tampered machines will be ignored, and of course there is an aspect of technological arm race here). The fact there is basically 0 security on that end is one of the big proof the guy doing that aren’t serious. A standard telephone have better security against tampering than a voting machine from what I have seen.
Days later and we still have no idea who won at Iowa Caucus. What a trainwreck. Seriously Nevada needs to switch to some paper voting and counting instead of relying on a different app.
The type of voting machine used in Bolivia and Venezuela sounds reasonable. You select an entry on a machine, it prints a ballot, you verify it and then put it in the box. The paper ballots are counted as the official result but you have a quick unofficial count from the machines.
I believe this is the mechanism many election monitoring orgs recommend.
Though as Bolivia demonstrates bad actors can still claim that the results are fraudulent under this system.
@Daughter
It’s quite frustrating that the center has allied itself so much with the far right. This is a dangerous progression. In America a sort of inverse has happened, where many people who called themselves centrists and were registered as Democrat became hard right Trump supporters. I knew several people who did that.
@Specialfrog
We have the same mechanism down in Peru as well and it seems to work just fine. The problem is how the votes are reported though, because your candidate can win a precinct, but the person in charge of reporting can report otherwise. At least there is a paper trail if there is a need for recount though.
@ Ohlmann;
… for want of a “/s”, a discussion thread was lost…. 🙁
Although I DO have the worst attitude towards computer technology of anyone I know, I was being sarcastic.
… for want of a discussion thread, an idea was lost….
… for want of an idea, an employee was lost….
… for want of an employee a business was lost….
… for want of a business an economy was lost….
… for want of an economy a civilization was lost….
all for the want of a sarcasm font….
I’ll hafta remember that.
At that point, I am not entirely sure to what you point, but generally speaking, when Poe’s law start to show its head, then something bad is happening.
And something bad is happening in computer engineering right now 🙂
(I am reasonably high on personal responsability for software engineers. Like, do not work for Amazon or Facebook ever, and think about the consequences of what you code, be it a facial recognition tool or a navigation help for a plane)
(also, when I am in doubt, I try to err on the side of people being non sarcastical, because I find it much less awkward to answer straightfully to a sarcasm than counter-sarcasting someone who actually wanted to help)
Because usually closed source software has an organized team running it and not just anyone can contribute and there’s some level of accountability for the person accepting and merging changes (plus everyone else down the line). That person is also usually the most experienced or an architect or something to that effect. With open source it’s whoever created the repository or whoever they gave the rights to.
It’s basically professionals vs. hobbyists. Would you want to fly in an airplane built by hobbyists? I once had a guest lecturer in college and he was a software engineer who built cars as a hobby and he thought agile was the best method for building a car because he could build a whole car in a weekend. He shared photos. It had a styrofoam body and he had never safety tested it and who knows what the emissions were like. Software hobbyists take that same level of care with open source code.
I’m not saying there aren’t some great open source libraries out there. But those are the exception.most of the time it’s people putting a passion project out there and their passion is not secure or efficient code.