4 ms·

I wonder, if the device is equipped with a microphone and/or a webcam, does it mean that the school has the right to remotely activate them for "monitoring" pur

by astrobe_ 11mo ago

I wonder, if the device is equipped with a microphone and/or a webcam, does it mean that the school has the right to remotely activate them for "monitoring" purposes? It not too far from what they did when the monitoring software sent the screenshots of an email that never existed. And what if he joked about stabbing his girlfriend/boyfriend? Would the school report him to the police? What the police would do in this case?

runjake 11mo ago
I have worked extensively with this technology and have witnessed many of its pros and cons firsthand. I have seen it misused, but I have also observed it saving students' lives and preventing mass violence events. A major point to consider in the public conversation is what happens after a tragic event occurs. The school district is often called out for ignoring the warning signs, not paying attention to things that could have prevented the event, etc. So the other side of abolishing this technology is that school districts no longer have those tools and public expectations should be adjusted accordingly. What really happens is that public opinion ebbs and flows between support and opposition, depending on what tragedies have happened near term. The policy and legal frameworks used by US schools clearly state that school district staff are not allowed to remotely activate the microphone or camera on a student's device. There's also legal precedence: https://en.wikipedia.org/wiki/Robbins_v._Lower_Merion_School_District https://en.wikipedia.org/wiki/Robbins_v._Lower_Merion_School... When the Robbins case occurred, school districts everywhere took notice. In the organizations I've worked with since then, we no longer activate the microphone or webcam, regardless of the Chromebook's location. However, I can't speak for every school district, whose morals and ethics may vary greatly. > It not too far from what they did when the monitoring software sent the screenshots of an email that never existed. It did exist but it wasn't never sent. The software runs as an agent on the student device and inspects the DOM tree for text phrases it considers alert-worthy: self-harm, threats, drug use, etc. > And what if he joked about stabbing his girlfriend/boyfriend? Would the school report him to the police? What the police would do in this case? This entirely depends on the school and police personalities involved, but the answer is "possibly" or "probably", depending on the jurisidiction. Regardless of the outcome, I think what's really important include the following: - There ARE bad actors employed in every school district! Many of them would love to spy on students, collect naked photos and share them. - School districts need STRICT AND ENFORCED use policy and minimal "need to know" access for TRAINED district staff. No hand slaps. Termination of employment, and legal and criminal consequences are in order. - Auditing flipped on for everything possible (for CYA, if anything). If school staff flips on a webcam, that should be logged somewhere that cannot be scrubbed. In the case of a webcam activation, I'd have it auto-notify key personnel and probably legal. Those audit logs should be reviewed often by multiple auditors -- preferable a third party. Audit events should be backed by extensive documentation, such as a help desk ticket, if anything. - If possible, students should obscure the webcam when not in use to protect themselves. If feasible, I also suggest they get a cheap dummy mic off of Amazon and keep that plugged in. If this type of product survives litigation, we need to move toward assurances of privacy (eg. verifiable Private Cloud Compute model), so this doesn't turn into another Flock situation where certain government entities may have a global/national single-pane-of-glass. I almost said "on-premises" there, but that would be a disaster because school districts don't patch their stuff.
- WalterBright 11mo ago
  > but I've also seen it save students' lives and stop mass violence events. The saving lives thing is always the excuse for total surveillance. Trading away your freedom for security gets you neither.
  - mikkupikku 11mo ago
    It also relies on knowledge of a counterfactual situation. Was the guy arrested for a threat genuinely going to hurt people, or was it a dumb joke that was taken seriously by somebody snooping in a conversation they lacked the context to even understand?
    - runjake 11mo ago
      From experience, this is generally easy to figure out. The ratio between dumb joke threats vs actual threats is something like >99.9%.
      - mikkupikku 11mo ago
        > The ratio between dumb joke threats vs actual threats is something like >99.9%. This makes the situation ripe for false positives. If 99.9% of threats are jokes and you correctly identify joke threats as jokes 99% of the time, that remaining 0.9% is several times larger than the real threats. And while the difference may seem obvious to you and I, being able to perceive the difference probably becomes harder across cultural divides (such as between teachers and a younger generation of students.) Furthermore, any bitter teacher with an axe to grind can leverage "zero tolerance" rules and strategic ignorance to deliberately construe a joke as a real threat to get rid of a kid they don't like (I've seen this sort of thing happen.) The more surveillance there is, the more opportunities there are to catch somebody making a edgy joke in what they thought was a private conversation.
  - runjake 11mo ago
    Touche. I get that and agree. It's certainly a polarizing conversation. I'm hoping the conversation and courts arrive at definitive guidance and regulations that preserves freedom, doesn't add to the surveillance state and provides some kind of answer to the half or more than half of the population that expects school districts to surveil everything kids do on their devices (self-harm, harm, bullying, etc). It's a really weird experience to hear the same powerful people argue both sides. How do you expect us to do one without the other? And again, it's... safe to assume there are a lot of bad actors in education where enforced safeguards are needed.
tehwebguy 11mo ago
This is a substantially more serious scenario now that transcribing and analyzing audio / video content is so much faster & cheaper today. Previously some freak had to watch or listen to everything in nearly 1:1 time to eavesdrop.

runjake 11mo ago

I have worked extensively with this technology and have witnessed many of its pros and cons firsthand. I have seen it misused, but I have also observed it saving students' lives and preventing mass violence events. A major point to consider in the public conversation is what happens after a tragic event occurs. The school district is often called out for ignoring the warning signs, not paying attention to things that could have prevented the event, etc. So the other side of abolishing this technology is that school districts no longer have those tools and public expectations should be adjusted accordingly. What really happens is that public opinion ebbs and flows between support and opposition, depending on what tragedies have happened near term. The policy and legal frameworks used by US schools clearly state that school district staff are not allowed to remotely activate the microphone or camera on a student's device. There's also legal precedence: https://en.wikipedia.org/wiki/Robbins_v._Lower_Merion_School_District https://en.wikipedia.org/wiki/Robbins_v._Lower_Merion_School... When the Robbins case occurred, school districts everywhere took notice. In the organizations I've worked with since then, we no longer activate the microphone or webcam, regardless of the Chromebook's location. However, I can't speak for every school district, whose morals and ethics may vary greatly. > It not too far from what they did when the monitoring software sent the screenshots of an email that never existed. It did exist but it wasn't never sent. The software runs as an agent on the student device and inspects the DOM tree for text phrases it considers alert-worthy: self-harm, threats, drug use, etc. > And what if he joked about stabbing his girlfriend/boyfriend? Would the school report him to the police? What the police would do in this case? This entirely depends on the school and police personalities involved, but the answer is "possibly" or "probably", depending on the jurisidiction. Regardless of the outcome, I think what's really important include the following: - There ARE bad actors employed in every school district! Many of them would love to spy on students, collect naked photos and share them. - School districts need STRICT AND ENFORCED use policy and minimal "need to know" access for TRAINED district staff. No hand slaps. Termination of employment, and legal and criminal consequences are in order. - Auditing flipped on for everything possible (for CYA, if anything). If school staff flips on a webcam, that should be logged somewhere that cannot be scrubbed. In the case of a webcam activation, I'd have it auto-notify key personnel and probably legal. Those audit logs should be reviewed often by multiple auditors -- preferable a third party. Audit events should be backed by extensive documentation, such as a help desk ticket, if anything. - If possible, students should obscure the webcam when not in use to protect themselves. If feasible, I also suggest they get a cheap dummy mic off of Amazon and keep that plugged in. If this type of product survives litigation, we need to move toward assurances of privacy (eg. verifiable Private Cloud Compute model), so this doesn't turn into another Flock situation where certain government entities may have a global/national single-pane-of-glass. I almost said "on-premises" there, but that would be a disaster because school districts don't patch their stuff.

WalterBright 11mo ago
> but I've also seen it save students' lives and stop mass violence events. The saving lives thing is always the excuse for total surveillance. Trading away your freedom for security gets you neither.
- mikkupikku 11mo ago
  It also relies on knowledge of a counterfactual situation. Was the guy arrested for a threat genuinely going to hurt people, or was it a dumb joke that was taken seriously by somebody snooping in a conversation they lacked the context to even understand?
  - runjake 11mo ago
    From experience, this is generally easy to figure out. The ratio between dumb joke threats vs actual threats is something like >99.9%.
    - mikkupikku 11mo ago
      > The ratio between dumb joke threats vs actual threats is something like >99.9%. This makes the situation ripe for false positives. If 99.9% of threats are jokes and you correctly identify joke threats as jokes 99% of the time, that remaining 0.9% is several times larger than the real threats. And while the difference may seem obvious to you and I, being able to perceive the difference probably becomes harder across cultural divides (such as between teachers and a younger generation of students.) Furthermore, any bitter teacher with an axe to grind can leverage "zero tolerance" rules and strategic ignorance to deliberately construe a joke as a real threat to get rid of a kid they don't like (I've seen this sort of thing happen.) The more surveillance there is, the more opportunities there are to catch somebody making a edgy joke in what they thought was a private conversation.
- runjake 11mo ago
  Touche. I get that and agree. It's certainly a polarizing conversation. I'm hoping the conversation and courts arrive at definitive guidance and regulations that preserves freedom, doesn't add to the surveillance state and provides some kind of answer to the half or more than half of the population that expects school districts to surveil everything kids do on their devices (self-harm, harm, bullying, etc). It's a really weird experience to hear the same powerful people argue both sides. How do you expect us to do one without the other? And again, it's... safe to assume there are a lot of bad actors in education where enforced safeguards are needed.

tehwebguy 11mo ago

This is a substantially more serious scenario now that transcribing and analyzing audio / video content is so much faster & cheaper today. Previously some freak had to watch or listen to everything in nearly 1:1 time to eavesdrop.