4 ms·

The incentive structure is bad. However much money you make for organizing unions, I'm positive the company they're attempting to unionize will pay you several

The incentive structure is bad. However much money you make for organizing unions, I'm positive the company they're attempting to unionize will pay you several times that to give them the list of workers who want to unionize. It's the same problem review sites have; companies that want to game the reviews will pay more than users who want honest reviews. You'll also have issues with companies paying someone to join the unionization effort to spy on it. That's not insurmountable with proper designs, but it makes it a lot harder to assist with organizing when you have to silo users from one another to prevent espionage.

brundolf 6y ago
Could the first problem be prevented at a legal level? The app company itself files some sort of legal document preventing itself from falling prey to that sort of corruption (although, really, the first time they did that the company would go under because nobody would use it ever again) The second problem is harder... maybe the tech could be designed such that nobody, including the makers of the app, could see the list until/unless a quorum is reached? I can't believe I'm saying this, but could the blockchain be useful here?
- brundolf 6y ago
  Edit: a simpler solution would be to give each person voting a unique, nonpersonal identifier, and then end-to-end encrypt their attached personal details, so after the voting period is done they would have exclusive authority to de-anonymize their information (so it can be verified as part of the final vote tally)
  - rtkwe 6y ago
    That solves the actual vote but that's not the big issue. At the start you have to get enough people interested and talk to other people to get them interested and the company can use that to do some 'totally not retaliatory firings/performance reviews of key organizers.' If it were as simple as just needing to make an up or down vote there wouldn't be so many issues but there's a lot of anti-union propaganda out there you might need to convince people out of before trying to vote.
tablespoon 6y ago
> The incentive structure is bad. However much money you make for organizing unions, I'm positive the company they're attempting to unionize will pay you several times that to give them the list of workers who want to unionize. It's the same problem review sites have; companies that want to game the reviews will pay more than users who want honest reviews. That's probably solved by having a union or a nonprofit develop and run the app, instead of a for-profit company. Is seems like the default assumption is for-profit companies are the best vehicle to accomplish any task, but there's a lot they just can't do or can't do well. > You'll also have issues with companies paying someone to join the unionization effort to spy on it. That's not insurmountable with proper designs, but it makes it a lot harder to assist with organizing when you have to silo users from one another to prevent espionage. Yeah, that's harder problem, but it's an issue that needs to be solved for any such effort. Maybe you could have some kind of validation/referral type feature, e.g. refer your coworkers anonymously, and part of on-boarding they verify a list of people and their roles, including the initiator. You could probably structure it so every person gets vetted, but it's unclear who started the process [1]. [1] e.g. initiator refers the effort to 10 coworkers. You have a list of 11 potential employees. Send random 3-worker subsets that together cover all 11 to each referral for vetting of employment location and title. At the end of it, you should have some confidence that you can filter out bogus people and management. Though you still have the problem of snitches and spies who actually work the job.
robbmorganf 6y ago
Maybe we finally found an application for Blockchain...

brundolf 6y ago

Could the first problem be prevented at a legal level? The app company itself files some sort of legal document preventing itself from falling prey to that sort of corruption (although, really, the first time they did that the company would go under because nobody would use it ever again) The second problem is harder... maybe the tech could be designed such that nobody, including the makers of the app, could see the list until/unless a quorum is reached? I can't believe I'm saying this, but could the blockchain be useful here?

brundolf 6y ago
Edit: a simpler solution would be to give each person voting a unique, nonpersonal identifier, and then end-to-end encrypt their attached personal details, so after the voting period is done they would have exclusive authority to de-anonymize their information (so it can be verified as part of the final vote tally)
- rtkwe 6y ago
  That solves the actual vote but that's not the big issue. At the start you have to get enough people interested and talk to other people to get them interested and the company can use that to do some 'totally not retaliatory firings/performance reviews of key organizers.' If it were as simple as just needing to make an up or down vote there wouldn't be so many issues but there's a lot of anti-union propaganda out there you might need to convince people out of before trying to vote.

tablespoon 6y ago

> The incentive structure is bad. However much money you make for organizing unions, I'm positive the company they're attempting to unionize will pay you several times that to give them the list of workers who want to unionize. It's the same problem review sites have; companies that want to game the reviews will pay more than users who want honest reviews. That's probably solved by having a union or a nonprofit develop and run the app, instead of a for-profit company. Is seems like the default assumption is for-profit companies are the best vehicle to accomplish any task, but there's a lot they just can't do or can't do well. > You'll also have issues with companies paying someone to join the unionization effort to spy on it. That's not insurmountable with proper designs, but it makes it a lot harder to assist with organizing when you have to silo users from one another to prevent espionage. Yeah, that's harder problem, but it's an issue that needs to be solved for any such effort. Maybe you could have some kind of validation/referral type feature, e.g. refer your coworkers anonymously, and part of on-boarding they verify a list of people and their roles, including the initiator. You could probably structure it so every person gets vetted, but it's unclear who started the process [1]. [1] e.g. initiator refers the effort to 10 coworkers. You have a list of 11 potential employees. Send random 3-worker subsets that together cover all 11 to each referral for vetting of employment location and title. At the end of it, you should have some confidence that you can filter out bogus people and management. Though you still have the problem of snitches and spies who actually work the job.

robbmorganf 6y ago

Maybe we finally found an application for Blockchain...