Similar issues can arise with police body cameras. Seattle's initial attempt to use body cams had to be stopped because all of the footage was available under state open record laws, and someone was filing requests for all of it and putting it up on YouTube. This was a massive violation of the privacy of numerous people caught on those videos. Police go into people's home, often uninvited and on short notice, and can easily see intimate and private details of those people's lives.
At first Seattle tried editing the footage before release to redact parts that might violate the privacy of anyone other then the officers and alleged criminals. I don't remember how many officers were involved in the initial body cam deployment, but it was enough that they generated so much footage that they didn't have the budget to edit it. I remember at one point they were even asking hackers for help with automating some of this.
A couple years or so later, they resumed using body cams. I'm not sure how they dealt with the privacy issues.
I wish I had more concrete information, but I remember hearing about one police department treating bodycam recordings in public and in private places differently for this reason. I imagine the officer would tag it when they left their shift or offloaded the footage.
In general, I've heard reviewing the footage is a huge undertaking and an unexpected burden. I wouldn't be surprised if, like many backup systems, they're not often tested and found to be insufficient when they go to pull a video...but, then again, they also deal with evidence which has similar requirements.
A system with enough manual control that the officer has the task of tagging and offloading the footage sounds like a system where inconvenient footage will be mysteriously deleted.
I feel like that's being addressed as body cams get rolled out. I've heard about procedures to address faking damaged equipment and other excuses for "missing" footage.
Cruisers have had video for awhile now, so it's not completely new.
This strikes me as such a good opportunity for putting the cognitive surplus to work. I would totally volunteer to review and tag police footage for my local PD. Given the popularity of cop reality shows, I'd think plenty would.
Sure, but I think it's possible to set a system up like that so it's reasonably hard to misuse. E.g., I have to sign up, sign an NDA, and get trained in person. Then I'm not just given full access to untagged video, but I'm treated like a Mechanical Turk worker: I get assigned a video, tag it, appropriately, and move on. And if I leak stuff inappropriately or misuse my powers, I face criminal charges for privacy violation.
The privacy issue is that an agent of the public is intruding into private spaces. The body cam footage is just revealing in a more obvious and exploitable way that the problem already existed. Whatever the body cam sees is already being recorded by the mark I cop eye.
I imagine that the privacy issues would be less severe for the recordings where cops knock, present a warrant, and then enter to execute it, because that is the due process instituted for the public interest breaching into private spaces.
Agents of the public should be doing their duties in public.
As for the automating, I'd use glyphs easily recognized by machine vision to mark different areas in the cop shop. Stencil-paint a glyph on the walls of the bathrooms and locker rooms, and the camera can automatically tag that video as not immediately available for public release, and blur an area surrounding the glyph for automated redaction.
If people are concerned about body cam video of themselves being released automatically without redactions, they can put up glyph stickers in their homes, or wear glyphs on their clothing. A public-private toggle switch on the cam might help, but is susceptible to bad user input or insufficient usage training.
Without some kind of explicit marker, AI may never be able to determine the transition points between public and private spaces, and certainly could not deal with adversarial attacks against that classifier. At worst, over-glyphing or inappropriate wearer inputs could exhaust the video-review resources.
But I think that it is in the public's interest for all calls to the emergency lines to be public, regardless of the potential privacy issues. Instead of having the 911 call center employees worried about covering their asses in case they make a mistake, support them with good training, checklists for the most common issues, a second tier of higher-trained crisis managers to join in for non-routine calls, and maybe an on-demand pool of expert support, such as the psychiatrists, lawyers, or explosives disposal technicians.
Leaving the postmortem analysis of an emergency only to officially authorized people allows cover-ups or deceptive re-framing to happen.
My neighbors left their apartment door open and did not respond to my knocks or my voice. I called the cops, because I wasn’t going to enter the apartment. The cops entered the apartment and found my neighbors in bed. Whether they were asleep or in the throes of passion, I don’t know. But if the cops had had body cams, it would have been captured.
I’ve heard a story about the cops being called by neighbors because someone was screaming for help. Turned out the lady of the house was doing the screaming. She was tied up in bed and the husband was in costume and collapsed on the floor having a heart attack.
These were both legal entries into private places. I think most people would agree these are things the public doesn’t need access to if they are captured by body cams.
Yup. Cops see a lot of things that shouldn't be open to FOIA requests. There's also the little issue that cops are human--they use the bathroom on occasion. If the camera might be running (they're human, they might forget to turn it off) and the video could be FOIAed they're not going to be welcome.
I think a reasonable compromise on this would be to modify the FOIA with regard to things like bodycam video and 911 tapes. Anything which is to be released gets edited for stuff that shouldn't be released and the requestor pays a fixed hourly rate for this.
None of the body cam footage I have seen has ever shown the wearer's own crotch. But this is the perfect use case for manually or automatically tagging inside-bathroom footage as private.
In any case, I don't think that the possibility that someone might skeeze out on public-record video overrides the public interest in having unrestricted access to those records.
In the ancestor post example of the neighbors with the open door discovered in bed, if video existed, it would surely show the cops loudly announcing at the front door before entering the residence, and continually announcing their presence while inside, before encountering the occupants. That's an attempt to salvage privacy, as well as a reasonable thing to do to determine if anything is wrong inside the residence.
I think a reasonable expectation of privacy might necessarily include closing your front door. And an open front door in a dense residential area is reasonable grounds to conduct a warrantless search. Urban residents usually don't want strangers or neighbors just walking right in to their homes without an explicit invitation, so doors stay closed and locked. In some states, entering an open residence uninvited [during daylight hours] is not considered trespassing. That requires an explicit act by the residents to inform guests that they are not welcome inside, such as closing and locking the front door, or telling them to leave.
It's the reasonable expectation of privacy that's the key. And even bench judges sometimes struggle with it. There's no chance that an AI could make that determination at the current level of development, including as it does factors such as time of day, character of the neighborhood, and local law.
It could detect and redact certain body parts, especially faces. The auto-redacted videos could then be released immediately. If an unredacted version of a video is requested, the video review officer could independently remove each redaction box using custom software tools, based on a checklist or flowchart of guidelines. I'm not quite certain as to whether the requester should be required to pay the full costs of un-redacting the video or not. Certainly, only the first requester should pay it, if anyone must.
I'm not sure I fully understand the issue. Don't police body cameras serve most of their purpose if they are available as evidence in legal cases? As far as I know, the evidence in evidence rooms is not by default available to the public, though it may end up being disclosed as part of a legal case.
If the problem is that prosecutors are protecting police unfairly, then publishing body camera footage isn't really going to help unless it's enough to cause significant public outcry. After all, there are plenty of instances of publicly available videos of police doing things where the public doesn't believe the legal repercussions are sufficient, and there is public outrage, and it doesn't matter.
Right, I'm asking why all body camera footage would be subject to such requests, or why the requests wouldn't be subject to significant redaction for obvious reasons. I am all for police transparency, but I don't see much function in having all body camera footage public as long as the footage is easily obtainable for any legal case.
They defaulted to bing public. Government workers needed to follow the law, and there was no blanket exception given for this kind of footage, thus the issue.
By comparison 911 calls are public in many areas, so they likely made an accurate legal assessment.
Open records principles accommodate time delay much more often than they accommodate total opacity. You can probably FOIA the police files on old cases, including details of the evidence.
I generally support regulations that protect privacy, as I'm sure most of us do. But they can come at a very real human cost, as in this case. Especially when it comes to regulations that affect people who aren't able to express their preferences (because they have died, are incapacitated, are not able to think clearly for whatever reason, etc), any general rule is going to lead to some bad outcomes.
FWIW I tend to think 911 calls ought to be public because they are public records. There are probably ways of making them public but not completely easy to access (eg, you could only listen to a specific record on site in a records archive). That would at least guard against the problems of releasing everything on the internet, which would impose major reputational risks to everybody who ever made a 911 call.
This is one of the privacy conundrums that we mostly have never come to grips with. There are a lot of things that would seem to be legitimately public records.
But there's arguably a big difference between records that you have to go to some county clerk's office to access and possibly provide at least a cursory justification for doing so and being able to click a few keys on your computer and skim through records for the hell of it.
Both are public at some level but one is not like the other.
> But there's arguably a big difference between records that you have to go to some county clerk's office to access and possibly provide at least a cursory justification for doing so and being able to click a few keys on your computer and skim through records for the hell of it.
I think it's important to consider the context. Privacy is important, but in an environment where authorities happen to "disappear" records that might be used against them or could be used to hold them to account, and fight tooth and nail to prevent easy access to them as in the US (and are often able to weaponize the courts to prevent access), there's a case to be argued that there are clear benefits to keeping it public in a way where there are no barriers.
The converse is also true: in an environment where random strangers take perverse pleasure in using any record in which you may appear to provide the worse possible representation, where you have no way of controlling this potential invasion of your privacy, where the record will endure _forever_ (and other groups will go to great lengths to ensure access to these records by anyone on the planet who happens to be curious), and where your ability to contextualize an event in which you happen to appear will be drowned out by whatever outrage machine has the biggest megaphone and fastest meme team, there is a strong case to be argued that there should be significant and prohibitive barriers to public access.
They're 911 calls. Show me one example of a 911 call and its misrepresentation being used to really harm somebody. Control only means those in power will abuse that control and prohibit its use as a method of ensuring accountability and transparency in how 911 hotlines are run.
Further, I think privacy is worth preserving for privacy's sake. We all have a right to it. But there are times where we have to give up rights in order to serve the interests of the community. I know it's not a popular way of thinking in the US, but it's just a fact of living in a society with other people. We don't have the complete freedom to do anything and everything we want. Giving up the privacy of 911 calls has clear benefits, and I haven't yet seen sufficient evidence that enough harm is caused by them being public to consider otherwise.
Even further, these states aren't making these records private for privacy reasons. You're just playing into their narrative and defending them on morality when they're doing it to cover their asses.
Really harm seems like an arbitrary and unrealistic bar for this stuff.
Calls before the caller died often make the news, personally I don’t want to be remembered that way and I doubt most of them want those moments made public. Often, 911 callers are in situations they would like to keep private and outside of criminals cases it seems unlikely for significant public benefit to occur should they generally be made so.
Citing 9/11 is a great example. We have better information of what happened, how people became trapped, and how the response was handled. At the very least family members should be able to obtain access to recordings about their family and callers should be able to obtain their own calls. This isn't some wildly hard concept. Calls to a public entity about private individual happen. Medicine is another field that exists but family members can gain access if needed.
One way that I've often thought we should divide it is that part of the public record is who is accessing the record and why they're doing it.
E.g., I'd be happier with the government having data on me if I knew I'd get notified any time somebody looked at it. And I'd also feel safer if privacy-focused non-profits were watching the stats on who's viewing what to ask questions like, "Why has Greg looked up 14 addresses on women in his neighborhood over the last month?"
It's a problem of scale and similar to other issues of automated criminal justice. If having your issues heard in court were a few clicks away that would blow the system up, which is why cities swat away automated filings in any way possible.
It implies an uncomfortable truth that justice relies on underutilisation in order to service demand.
Could you articulate this “big difference” a bit more? So far the only one I can parse from your comment is ease of access, driving to a clerk versus walking to my desk.
That doesn’t seem particularly convincing. Or at least I don’t see how this feeds into the privacy “conundrum”
Take for the example Norway’s rules (cited below) that tax returns can be viewed by anyone, and the list of who viewed each record is also public. This requires limitations on how the data can be reproduced.
If you need to drive to see a clerk it’s a lot harder to get access, and that’s a reasonable barrier. The clerk can also have reasonable limits on how many you can request and whether you can make a copy. Each of these rules do have an impact on how widely information is used or disseminated, and can limit data mining style use. Note that some of these limitations (like preventing a copy) can be more difficult when accessing from a website.
So yes, there are many useful limitations than can be placed on sharing of data that have a real practical impact on privacy.
You can read the car plate of cars passing in front of you. It is public information. If you can easily gather the location of all cars at any time, it becomes a big privacy problem. You can identify husbands cheating...
Indeed. I can't help thinking that the fact that some piece of tech makes a qualitative change like this means that there's some fundamental flaw somewhere with the way we associate a person with a public record or identifier. But I'm not really sure what that flaw is, or what to do about it.
Nobody is sure how to solve it because the problem is with people and the culture we have built around having actual privacy in the past. Technology has essentially robbed us of it, which is problematic as technology solves lots of other problems so we just cant get rid of it.
This is really similar to the music industry and the problem of piracy. Bits are easy to copy and hard to prevent the copy of.
I believe that if volume, processing, or accessibility of some new technology or information stream changes the access patterns to that information, that implies that the nature of that information gathering and accessibility needs to be re-evaluated.
A cop calling in a suspect license plate is one thing, an automated system reading all license plates at all intersections and tracking every vehicle in a city is different, and should have regulatory or judicial oversight applied in light of the new access. The fact that any one of those plates could have been seen at each intersection by a cop who happened to be there does not mean that license plate readers don't change the nature of that information.
A lawyer asking a clerk for a transcript of a call regarding a pending case is one thing. Transcribing all calls and getting a text message when someone called 911 yesterday and mentioned your name is a big difference.
I could be wrong but maybe one is recursively enumerable and the other is not? i.e at a clerk you need to know what you’re looking for in contrast to a list of records that you can just look through.
The records are no less public in either situation? They can be accessed by anyone, regardless if they’re actively looking for a specific record or stumbling across one while scrolling through their local clerk database on a Saturday afternoon.
I still don’t understand the conceit.
If one were to say checking documents out from a central clerking office acts as a check or something against unauthorized and potentially unethical use of public records, well then yes absolutely I can see some merit.
> The records are no less public in either situation? They can be accessed by anyone
No, that's not true. If you have to physically go to the county clerk's office then you have to physically be in the country. That is a significant barrier.
An example that is even more salient than 911 records: if you make any political contribution in the U.S. you are require by law to give your home address which then becomes part of an FEC database that can be accessed by anyone on earth with an internet connection.
You're correct, it is a barrier, it makes things more difficult for the requester.
How you request it or where you are when you make the request doesn't change the legal status of the record itself if a data set has been made public by your state legislature.
If your local laws say you have to make the request in person, so be it. It's still an open record that can be queried and retrieved by members of the public.
If your local law say you may make your request online at a specific web address, so be it. It is still an open record that can be queried and retrieved by members of the public.
The mechanism or methodology of requesting a public record does not change a public record factually and legally being public unless local law says so (or FOIA if we're dealing with a federal agency, but since the context here is 911 calls, this is much more likely going to fall to a how a state legislature decides to protect/expose public records).
Everything you say is true. Nonetheless, as someone who makes political contributions, it matters to me whether a Rumanian hacker can get my home address while sitting in their basement in their underwear, or if they have to get a U.S. visa and travel to my home town in order to do it. That is a salient difference to me, irrespective of the legal status of the information.
You're talking past each other. One of you is talking about whether or not it's public, the other about how accessible it is. They're distinct, but closely related, topics.
You have a valid concern. I do not intend to dismiss it but I will offer this is why I put forward the caveat previously that absent legislation that acts as a buffer against unauthroized or illegal use of public records (which almost every state has provisions in their record keeping laws over already), the mechanism doesn’t change or alleviate these concerns.
We need to weigh the purposes and drawbacks of policies and try to strike a balance in practice. In the case of public records, the purpose presumably is to fight corruption and maintain an informed electorate; the injury to personal privacy a major drawback.
Adding a cost to get the information is one way to adjust this balance— it ensures that the person making the request values the information for some purpose, at a level higher than the imposed cost. Then, when you pass regulations about appropriate and inappropriate motivations for using this information, there’s a standard against which the judge and jury can compare the proffered reasons for accessing the data.
We need to weigh the purposes and drawbacks of policies and try to strike a balance in practice. In the case of public records, the purpose presumably is to fight corruption and maintain an informed electorate; the injury to personal privacy a major drawback.
100% agreed. Monetary costs for records is a common approach to this, although many agencies allow for the costs to be waived at the request of the querying individual. It's entirely up to the queried agency to decide if they want to waive those costs or not, and it's a roll of the dice if that agency will bless you with waived fees.
Another way agencies have added "cost" to the requesting practice is through complicated and lengthy forms that are mailed to the local agency (and some agencies, I will offer, absolutely will NOT let you just walk in and hand over a form ABC-123, and require it to be mailed with a postmark). At least, complicated for the individual making the request. These forms though I think do a very good job of helping records divisions enumerate the type of requests that come in, the volume at which they are received and provides an excellent catalog for the types of data most commonly requested by the public to make searching and disseminating the requested records more expedient for clerks and court reporters.
But I absolutely agree, additional policy crafting is necessary to answer some key questions:
1) What should be released and how
2) What is the appropriate mechanism for making a request and what systems of work are engaged by jurisdictions in issuing a release
3) What are the conditions for a legally appropriate and publicly-interested release
I think this is as true for 911 calls as it is querying a comptroller for city budget items.
Fundamental changes to these, matter. Even small ones. Especially big ones.
Moving the needle a few percent (or fractions of a percent) can be game-changers in business or politics through threshold effects.
Moving the needle by orders of magnitude is effectively an entire new physics regime. And that's what we're seeing here.
Just because Y is "the same as X" doesn't mean that Y is X, if the rate of change is 10x, or 100x, or 1000x, or 1,000,000x greater. A 10x change gets you from walking speed to a car. 100x is an aircraft. 1,000x is an ICBM. The Tsar Bomba is Greek Fire ... at scale.
Infotech advances change costs by factors of millions to billions, far outside the scales above: ICBMs compunded, several times.
And such changes have massive unintended consequences.
Information technology is the perception and feedback system of society as a whole. Changing properties of information perceptibility, parsibility, storage, retrieval, processing, and distribution of information fundamentally changes entire social orders, and often not for the better. Lowering costs makes marginally "profitable" -- often in the rents-extracting sense -- applications more tractable. In that context, religious and political propaganda, yellow journalism, advertising, noise pollution, popular demagogues, urban legends, email spam, robocalls, stalking, harassment, pseudoscientific myths (flat earth, antivax, truthers, birthers), media (social or otherwise) smear tactics, and automaded extortion are entirely predictable outcomes.
One can be used to automate extortion. For example, sites with mugshots popped up once they were online. If you got arrested for any reason and didn't want potential employers to see your mugshot in the first page of a Google search, you had to pay these sites an "administrative fee."
Similarly, you could post 911 records with addresses if they were available with a simple request, and I can see how that could bring property prices down, depending on the kind of incident.
Suppose someone used Microsoft Office to quickly template and print envelopes with the addresses of 10 different municipalities and stuffed them with boilerplate requests for public information, and then used the responses from those municipalities to carry out extortion.
Should the law act differently towards the individual who mails a lawful open records request than the one who queries for records via HTTP if they used the same records, provided by the same municipalities and jurisdictions to carry out unlawful activity post hoc?
I would have thought it obvious that friction is very much a part of the equation that has historically gone into deciding what information should be public, what should be formally sealed, and what should be actively expunged from the public record to the degree possible.
Suppose I hire an investigator to conduct a (legitimate) investigation into a political opponent, high-level hire, etc. including past addresses, any arrests, mentions in newspapers or online, anything they've written, etc. They'll dig up a lot of information, some of which are public records and some of which is simply publicly available information.
Are you seriously trying to argue that there is no difference between an in-depth targeted investigation of an individual and that same information being available about everyone to anyone on the planet with a press of a few buttons?
Sure, you can argue that there is no privacy and people should just get over it. But there is a significant practical different between information that there's some friction to access and information that's indexed and freely available to everyone with a computer.
Are you seriously trying to argue that there is no difference between an in-depth targeted investigation of an individual and that same information being available about everyone to anyone on the planet with a press of a few buttons?
This would be pretty close to it however
Sure, you can argue that there is no privacy and people should just get over it
on the opposite side, no, this is absolutely not what I'm arguing.
If querying a public record online gives rise to a sufficient argument that public harm is being caused, legislatures are not ignorant of this, and they are certainly not deaf to it. At the Federal level, SCOTUS recently reversed a decision of how to answer that very question with regards to Food Stamp data and how it can be queried, and even if it should be queryable at all: https://www.scotusblog.com/case-files/cases/food-marketing-i...
All of my comments here I am trying to make as objectively as possible without passing judgement on any concern or aspect of it, but simply remark on how open records laws are implemented and how laws affect those records and the ones who query them.
Any assertion that I am making an assessment or judgement about the people involved runs contrary to my motives in even bothering to comment here.
Why only consider automated extortion, and not automated data analysis for good? For example, looking for patterns of systemic discrimination or racism that could easily be explained away as coincidence when looking at only a single incident, but is blatant when taken in aggregate? Or automated searches for suspected corruption or bribery?
I think the bias should be in favor of transparency. Automated information gathering can be used to protect the public as well as for harm.
I suspect for every one bot combing records for evidence of corruption (run by a local concerned citizen) there will be thousands of bots run by disinherited Nigerian princes looking for easy marks and for dirt to use for social engineering.
On the balance you will get one corrupt official fired for a whole bunch of people scammed, extorted or otherwise taken advantage of.
Think back to history of spam for example, I recall it reached huge percentages back in the day, before effective measures were introduced. Even now spam traffic is 45% of all email traffic.
Depends on the jurisdictions’ open records laws I would imagine yes they more than likely would be. Presumably if they have such statutes, some jurisdictions have provisions to catch such types “burdensome requests” and either slap a heavy fee for processing such a request and hopefully deter the requestor, or they appeal to a higher court when denying the request entirely
But otherwise if a state has open records laws and you make a lawful request for a large body of publicly available records then I’m inclined to say by definition the clerk is required to provide them provided all requirements of the request are satisfied.
You could just ask the clerk for all the records, this is what public records databases do. They drive around to all the court houses and make copies of all the public records, put them into one big database and sell access to it.
Or if my city has a web application in front of this database I can not leave the house and do my research comfortably in my own environment.
Still not seeing the privacy implication of driving for records and clicking for them when discussing a public record that by definition, is in the public interest carried out by wards of public service (public service answering points).
You have been highly active in this thread and it doesn't seem like you are trying to understand the other viewpoint. Do you have a personal stake or other underlying assumption in the matter that might be biasing you to a particular conclusion?
If you don't see privacy implications to releasing recordings of e.g. people experiencing medical emergencies (medical records are otherwise clearly protected) then perhaps you need to reevaluate your understanding of what privacy means to different groups of people.
Do you have a personal stake or underlying viewpoint
Yes. I work in a directly advisory position with various municipalities across the united States implementing online open records systems of engagement, among other systems of public accountability and answer these types of questions every day.
I had been intending to avoid appealing to job titles in discussing the matter but this is my profession and the question of legal access and barriers to access weighed against public interest is a topic that happens in every meeting I sit in.
Let me just suggest that you may be too narrowly focused on the legal status of the information. This is indeed important. But, as I and others have noted and given examples of, as a practical matter there's a big difference between information that anyone in the world can mass download to a computer and information that requires someone to show up in person at a clerk's office and pay for a paper copy.
>We're ostensibly talking about records that were made public by law so I think that focus is 100% warranted.
One last comment. Yes. But I'd argue (and I imagine you agree) that many of the tradeoffs considered when those laws were passed reflect, among other things, that "public record" didn't originally often equate to big publicly accessible database.
In addition to this, laws can and should be changed, if we have a discussion about whether they're fulfilling their purpose -- like in this thread -- and realize that they're not.
this comment chain seems to be more focused on the moral aspects of privacy rather than the legal ones.
Sure. Fair. Valid.
Except it's really more nuanced than just saying "legality != morality", especially when the rubber meets the road, and when it comes time to actually implementing systems of recording (and potentially disseminating) the public's interaction with their local jurisdictions and municipalities, and I think those nuances get lost when sitting on the periphery of installing policy and reading about statecraft compared to being inside, operating the levers and pulleys and-as the phrase goes-"watching the sausage get made"..
That's why I'm being so willing to challenge some of the notions shared here today: I'm trying to be the voice of the insider and expose some of the reasons why certain things may be the way they are with regards to public data and open records.
You're right legal does not necessarily mean moral, but when it's time for measurable outcomes and results of human actions, and interactions between citizens and their local governments the legal framework does a much better job of delivering those outcomes and results on a consistent basis than most others, and is the framework we use of keeping the operators and participants equally consistently accountable (most of the time, anyway).
There are definitely many public records that should be as widely available as possible, so good work helping municipalities with those. I don't think people are saying that all searchable records are problematic. They are saying that some uses and some specific types of records are problematic -- those that do more harm than good.
As an example, if hypothetically I were a victim of some embarrassing emergency, I wouldn't mind being part of anonymous aggregated statistics, but I definitely would object to anybody but myself and essential personnel having open access to the 911 recording, and I definitely wouldn't want my employer or neighbors to know any details or anything linking me to them without my permission or a substantiated court order, released under seal.
* If I were receiving domestic violence and in fear of my life, how would I feel calling emergency services knowing my screams for help in one of my most vulnerable states in life would later be public?
* If I suspected I could be vulnerable to domestic violence, how inspired might I be to make the same type of call once hearing another person's similar emergency call?
It is recommended (IANAL, read this over the internet, but this seems legally sound to me) that if you are involved in an incident of self-defense, one would say, “I want to report a ____” and go on to explain the /outcome/ of events to the operator.
To play devil's advocate, 911 calls are information about yourself that you are giving to the government. There isn't anything more than that which makes it a "public record". I can think of other types of information about yourself that one gives to the government, such as your tax return. Should that be public record, and how is it different to a 911 recording?
That dies provide an interesting example. Requiring the accessing party disclose who they are (and who they may be acting for), and putting strict access limits on the requests based on that (once per 10 minutes or hour possibly, unless you petition for a slightly higher access rate because of need, such as your profession) and all under penalty of prosecution of the U.S. government neatly solves the issue for me I think.
If you have to supply your SSN, name and an email for each request, and the name and SSN are shown to later requesters, that easily allows the government to track abuse of an SSN to access records, allows for preventing third parties from accumulating all the records to circumvent the controls, and allows public view of records that might be problematic to have negative privacy consequences if data mined completely.
When I worked on an electronic medical record, the system had a concept called "break the glass". Any record could be accessed, but if that access was not whitelisted as normal for a particular time, place, person, and purpose, the user would be presented with a warning that their access would be recorded and flagged for subsequent auditing, to determine if it was appropriate or not. They could then click through to access the record.
If a care team needed to know right now whether a patient was allergic to medication X, the computer nurse would click through. If someone was just trying to look at the inpatient dietary orders of a celebrity, the warning would likely scare them off and protect patient privacy.
But the important difference there is that the person accessing was not given a digital copy of the record they were accessing, unless they photographed the screen with their phone.
If you release one thing over the web to one person, you permanently lose all access control over that thing. If the system records that one person looked up all the records, well now you know who to go to when you want to look at the records anonymously.
In the UK, we have a system called a "summary care record", which contains just the kind of information likely to be useful to care teams doing urgent work on you. Things like blood type, medication allergies. Then all the detailed stuff is held separately and is presumably harder to access.
Now that you mention it, medication allergy information probably wouldn't have triggered the warning, but I couldn't recall any specific examples of a legitimate access triggering the warning (as you would expect), other than maybe for VIP patients, where some restricted-access flag has been set. I.e. if the ambassador to Elbonia is allergic to mud crabs, you don't want that to be easily accessed information.
First, Federal law prohibits the use of an SSN as a "National ID number". This type of usage is explicitly forbidden. The fact everyone seems to ignore this notwithstanding, Federal systems have been specifically moving away from SSN's as a natural key specifically because of being prohibited, and because they are reused.
Second, a Corporation mechanical turking 1000's of people to datamine public records would not necessarily be easily caught by your system, especially since any body after collecting the data can disincorporate, and take the fruits of the operation with them to be reincorporated under another name; building larger and larger datasets each time.
To the mechanical turk point, that's where the enforcement arm comes in. It's rare that companies so blatantly break the law as a business model. Ubers happen, but not break-into-you-neighbor's-house-as-a-service companies. As soon as the latter gained any traction it would be shut down and people would go to prison. Nobody would start that company. Hopefully a law could be written to make the company you describe more of that kind of company and less of an Uber.
First off, ugh. I hate noticing typos in my comments after the edit period is over. I even emphasized this typo! :/
> First, Federal law prohibits the use of an SSN
Good point. It's always confused me how that's true and how it's used in taxes though. Is it sufficient if it's only part of an identifier (e.g. SSN + last name)?
> Second, a Corporation mechanical turking 1000's of people to datamine public records would not necessarily be easily caught by your system, especially since any body after collecting the data can disincorporate, and take the fruits of the operation with them to be reincorporated under another name; building larger and larger datasets each time.
That's why there would be disclaimers on the access page (Federal crime to access this data on the part of a third party without disclosing third party, so people don't do it) and a license included with the data as to how it is allowed to be used, which should survive any company change.
So, if turks think it's worth bypassing a clear message indicating they would be committing a crime, that's easily handled by the legal system. If a company thinks they can unincorporate and reincorporate to bypass a federal crime, I think they don't understand the legal system, and as long as the justice department pursues it, they'll likely have a hard time explaining where they got their data from (if they can even get people to supply their own SSNs to get it given clear indicators they would be committing a crime to do so).
Really, the thing I would be more worried about it identify theft profiles being used, but as long as there is clear indication in some manner (a letter once a week/month to the tax address on record of the SSN, if that's legal) then it either wouldn't be lucrative to use those identities in that manner or people might get a fairly early indication that their identity has been stolen.
It's an interesting thought experiment. If this were about anything other than pure information, then I think that would be entirely sufficient. Given we're talking about information that we're trying to control the flow of, and once it's entirely out (and in bulk) there's not much you can do, is it sufficient? I don't know. I think it's not really worse than just making it freely available though, some it's an alternative option given that we already see both ends of the spectrum in play right now in different places (e.g. nations that allow tax returns to be publicly seen with apparently little control other than logging, and nations that do not allow it).
It's instructive to look at why the ACLU went from supporting to not supporting the statute in RI:
Supporters of the bill included the Rhode Island chapter of the American Civil Liberties Union.
Steven Brown, the ACLU chapter’s executive director, said the organization’s support was based on the “privacy values in not having these often very intimate types of calls just available to everybody in the public.”
But over the years, the interpretation of the law by the courts has been “problematic,” Brown said, because judges have denied access to 911 recordings even when they would serve the public interest. He said the ACLU would support adding a “good cause mechanism” in the law to allow for the release of 911 calls if it can be shown that there is an “important public service in knowing what’s in those calls.”
Making 911 calls available because "your family wants to know" seems to be too far of a violation of privacy. While I feel sorry for the person who lost their brother, he can go to court and ask for the records to be released - which seems about the correct amount of privacy protection for something like this.
I understand the privacy side of things, but I'm suspicious of the motive. We've seen with police body cameras that the cops are quick to release exonerating ones, but "we don't release evidence during a pending investigation, but trust us, the video is exonerating" in cases where the footage is damning.
Police have massive short term gains from going against the rules. It makes their jobs easier (short term), safer (short term), more apparently successful (short term). Plus plenty of law enforcement exclusive privileges ready for abuse. Taken together this creates a massive demand for oversight and accountability, and big incentives to undermine those.
None of this exists in the field of EMTs. When they "hide behind data protection", it would be to protect against frivolous litigation after honest mistakes, which I find not only understandable but also preferable. If I ever need one of them, I'll surely want them to give their best medically rather than spending cycles on court-proofing their process.
Assuming that's true, couldn't that maybe be because exonerating evidence precludes the need for a pending investigation, whereas unclear or damning evidence requires further investigation to build a proper case before the whole thing goes to trial?
Our legal system is intentionally biased towards "innocent until proven guilty", so it generally makes sense to not release evidence of criminal activity prior to the case being brought to trial. The same obviously doesn't apply to evidence of innocence, especially if that evidence is strong enough that it likely means the case will never make it to trial in the first place.
> However, according to the Chicago Sun-Times, CPD officials explained that the lack of audio is due to the fact that the “sound doesn’t turn on until 30 seconds after the recording button is activated.” Chicago’s local ABC affiliate also similarly reported, “The body cam video that was released does not have audio as there is a 30-second delay when the officer turns the device on.” This is simply not true.
The "not tainting an ongoing investigation" excuse is particularly silly as in many cases the potential perpetrator is specifically granted access to the footage:
> A new report by the Leadership Conference on Civil and Human Rights and Upturn Research to be released Tuesday shows that the vast majority of the nation's largest police departments allow officers to view the footage before writing a police report or being questioned by investigators during use-of-force cases. The group believes that policy undermines police credibility and runs the risk of influencing how the officer describes what happened.
The first article explains that the footage was not edited; the audio data simply wasn't being recorded until after the officer pressed the record button (the first 30 seconds were from a buffer with no audio).
And no, it's not a silly excuse. Just because current policy in some departments allows one potential witness's testimony (the officer's) to be tainted for procedural reasons (ensuring accuracy of the police report), doesn't mean we need to risk the same thing happening with _all_ witnesses. That's a ridiculous argument.
I think a reasonable compromise would be a sort of "public-by-default after X number of days, but private if requested so by caller".
I understand the utility of them being public. On the other hand, I have made 911 calls that I definitely don't want to be publicly available, especially online where voice-recognition tools could be used to find a specific person's calls.
I'm fine with that. We want people to feel free to call 911 when they feel threatened. Some people will feel threatened for stupid reasons, and some of those stupid reasons are racist. But it's better to let people call without risk of being shamed and then let the operator decide how to react to a call.
> Protect them from what? Public shaming by people who look them up in databases? Being racist isn't illegal.
Retribution, that was obvious to me.
Examples: Physical violence as retribution is illegal, while being kicked off of board seats and other areas of influence isn't. Being 86'd (banned) from private establishments isn't illegal.
This should probably be handled the same way as donating your dead body for science (or profit): make this an individuals own choice. This improves societal cohesion: if the individual allows the release of this private data when it concerns him- or herself, any designated people are free to request this data, and each one of them should be free to share this freely, it's up to the individual to choose discreet people if so desired, avoiding anger between the mourning individuals with respect to the state. It also improves cohesion between individuals and the state in aother way: if the deceased chose to not release such private data, then the mourning individuals will be less angry at the state and more angry with the "poor" decision of the deceased...
As with all such systems you get the endless meta-fight of opt-in versus opt-out: as a stoichiocrat, my preferred solution is to provably flip a random coin toss, such that anyone who cares about this setting must explicitly opt-in or explicitly opt-out, those who act careless simply do not care about their setting, and the conundrum of systemic bias in the sense of all opt-in versus all opt-out is avoided. The probability of the coin toss can even be set identically to the fraction of people who explicitly opt-in vs explicitly opt-out (in example if 90% of the populace ignores the setting, and 7% opt for release and 3% opt for privacy after death, then for the 90% without explicitly stated preferencee a random 7 out of 10 people will have the details released and 3 out of 10 will have the details withheld)
I haven't heard the label stoichiocrat[1] before, and Google search only finds this HN page. Do you know if many others share these views, and if they've ever been implemented?
It's easiest to think of rules in a black and white, opt-in or opt-out way, and I hadn't considered the probabilistic option.
[1] Without knowing the words, I think you mean "stochiocrat", which would be related to "stochastic" and probabilistic. Google found one result for this, but it didn't provide good context; I think it was referring to how Venetian Doges were elected [2]. "Stoichiocrat" would be more related to "stoichiometery", coming from a Greek word that can mean "element".
You are correct, "stoiciocrat" cf "stoichiocracy" is apparently not the correct term. When I learned about this concept about 8 years ago [1], I seem to misremember this to be the word on wikipedia. Apparently the correct spelling is "stochocracy" although the results are very few if I try to look them up today. I was referring to concepts that seem to be more conventionally called sortition (apparently the better term at least in academia according to Google Scholar result counts), demarchy, random ballot...
[1] The history page of "stochocracy" at wikipedia suggests this wasn't my imagination:
While I still consider myself a "stochocrat", this is but one of my many ideologies.
I believe many different forms of governance are the best depending on the problem class or issue class, or properties of the problems themselves. For example a the problem "choose or design a setup to isolate heavy water from naturally abundant water" will not fare well under sortition, since it requires expertise. On the other hand, problems that have been broken down to such simplicity that anyone could execute it, but that typically suffer from corruption by a minority fare better under sortition: perhaps the selection or patrol routes and the assignment of cops to groups that run the patrol routes, would make it harder to initiate corruption, if you do not know ahead of time who will be your colleague, nor when or where.
Most decisions are not made by governments, but by individuals, and nowadays also computers (or this could perhaps more properly be seen as amplified decisions by programmers).
So if you ask "Do you know if many others share these views, and if they've ever been implemented?" I can not really answer, at least in the past systems have existed where people were raised with the concept of sortition, and all systems indoctrinate the next generation with the importance of itself. If you ask a person on the street what they think about sortition, they will probably not know the word nor the concept, and after explaining mechanically how to do it, they would probably consider it crazy and inapplicable to any problem. I think the number of people wanting to learn more about sortition would increase if not only explained mechanically, but also the properties (good and bad) were explained.
Have they ever been implemented? I consider it an important part of Athenian democracy. Also sortition is alive and well in many systems and places, think of sortition of a jury of peers, think of implicit sortition in mining a block in the Bitcoin system, think of explicit sortition in Algorand, ...
Note the last 2 examples are digital systems: a requirement for sortition to work is that "nearly anyone can satisfy the task" (such as verifying the signatures of transactions, and checking for sufficiency of funds, which any computer could do in principle) as opposed to "design a lock-in amplifier".
Our litigious society has driven us in two directions: limiting liability and secrecy. Limiting liability by outsourcing important functions to contractors, who can absorb a lawsuit and simply go under. A government has effectively unlimited liability since it can raise taxes, and can't easily go under, which leaves it broadly exposed. And secrecy like this article.
Something will have to give or we'll see more and more of these trends.
What happens when you adjust those figures for something like differences in private gun ownership rates between populations? It seems obvious that America has the most guns, and that having so many more guns than anybody else probably causes a dramatic increase in the number of justifiable homicides police are involved in.
I don't know about other countries, but it's probably an extension of the overall American feeling of government in general:
"Okay, I put you in office and will allow you to run certain parts of my life. But remember that I pay your salary, and that everything you have in that fancy office belongs to me."
Recordings held on hard drives fall under that "everything in that fancy office belongs to me" line of reasoning.
In almost every company, the person you can talk to when you feel threatened or are for some other reason panicking will very explicitly not automatically pass on this information to whoever is paying his salary. This is a weird reasoning (but I realise you might just be iterating the reasoning rather than agreeing to it).
Charge the requesters a review fee, per minute of footage?
That way you can categorize the video on demand, and release it as a public good after. The categorization would need an appeals process.
Categorization would be akin to: release now, release in a year (for embarassment), release in 30 years (for crimes) and perhaps never release (murder scenes etc., but let it bitrot in a cabinet for 30 years first).
If you're poor you can ask a privacy and open governance charity to buy a block of footage encompassing the footage you need.
Perhaps someone could write an RFC for this and send it around police departments. It should be framed as a minor source of income and that it may be a fun occasional distraction to some agents - if the tooling has a sufficiently powerful playback and fast-forward function.
Seems like the caller's interest should trump that of the victims' families, and doing so could make the system safer.
If you are taking drugs with someone and they look like they are dying it should definitely be safe for you to call 911, even if you are participating, or are somewhere you shouldn't be, or don't have a visa, or whatever.
If you break into a building, perhaps to steal, and your companion falls and breaks a limb you should definitely be able to call 911 (bad luck for your companion who will surely be arrested though).
A person's heart stops, you call 911, the try CPR and they die anyway (well, they're already "dead" if you're doing CPR). You shouldn't have any liability.
Grand juries and review boards could still confidentially review 911 calls for cases like this, where the family fears malpractice by the first responders.
==
This is no different for why TSA should not screen for drugs or anything but weapons that could threaten a flight; why people not already in the court system should not be arrested in courthouses (or nearby). Why people who are here illegally should get vaccines and drivers' licenses. What seems "unfair" can actually promote the public welfare.
> A person's heart stops, you call 911, the try CPR and they die anyway (well, they're already "dead" if you're doing CPR). You shouldn't have any liability.
There's already some kind of provision for this in many places; it's known as a "Good Samaritan law".
> If you are taking drugs with someone and they look like they are dying it should definitely be safe for you to call 911
In Washington state, this is how it is, and there are education programs, and statutes to this effect, telling kids/anyone "If you call 911 because you think your or your friends life is at risk because of alcohol or drug use, neither you nor they will be arrested for that".
This is sad to read. I understand the will to protect the privacy of the people calling in, but wouldn't the public interest prevail in some cases?
What I mean by that is, when weighing between keeping 911 records sealed for the sake of privacy; against releasing them to the public for the sake of public service (for example for performance review, reconstruction of the chain of events ...), shouldn't the public's interest prevail (as in many versus one)?
You're failing to consider the possibility that there is a public interest in maintaining privacy for 911 calls. It could be the case that people are more hesitant to call in something if they know the recording will become public. This could conceivably do more harm to the public at large than keeping the records private would.
911 recordings are records generated by public servants, and have implications for the wider public. This is similar to court records, which are also default-public.
Unless you're contagious, your health has no wider public implications. And if you are, your wishes are also going to be ignored.
> This is similar to court records, which are also default-public
It's hard for me to see how the trial of a rapist is similar to a rape victim calling for help.
The trial needs to be public so we can monitor the application of justice in the legal system.
The details of somebody's worst moments of their life needs to be public because...?
If I ever need to call 911, I want to only focus on getting help fast. But as it is, I also need to think about how I'll sound if this call is being played on the news or Twitter.
It is important conceptually because it involves essentially a petition of government power and a check on abuse on both sides.
If they provides absolutely horrible public service as in callously calling the rape victim a slut who deserves it and then later denied the misconduct and circles the wagons that is important.
Or if the abuser of the calls isn't held accountable.
Transparency vs privacy isn't a straightforward matter and there is ample room for debate and differences in opinion.
Using them to monitor abuse does not require making them public. It just requires making them available to those who claim they were abused, or if they did not survive their estate or whoever would have had medical power of attorney for them or someone similar.
> But as it is, I also need to think about how I'll sound if this call is being played on the news or Twitter.
If that matters, perhaps it wasn't an emergency that required the use of 911?
I imagine if you had a short-lived mental issue that caused you to go to court over it but were eventually found innocent, that might also be viewed as highly private, but the records would likely be public.
There will always be edges around any law or policy where positive outcomes are not includes and negative outcomes are. It's trivial to come up with negative outcomes for almost anything, the question is whether frequency of magnitude of the positive and negative outcomes result in a net gain of enough to make the law useful overall.
The question is not whether 911 calls include private information. They do and invariably will by their nature. The qeustion is whether the gain by making them public outweighs the loss, and whether we can incentivize the positives while mitigating the negatives.
One possible way to do this might be to make them gates through providing your SSN and name (which is then attached with a timestamp to future requests as an access log), severely rate limiting based on requester's SSN and name (with the ability to petition higher limits based on profession/need, automatically expiring over time), and licensed by the governing institution to only be used/copied in specific ways.
I mean it wouldn’t be right to publish a list of all the people with AIDS, right? That could have some implications for the public but it still probably shouldn’t be available.
Thankfully, that particular one is hopefully dealt with by the fact that deliberately infecting someone with AIDS is punishable and hopefully deterred. The whole world doesn't need to know that a person has AIDS for that person to take necessary steps to avoid infecting people.
They already are. At least a dozen third parties have access to most material aspects of your care.
Public 911 recordings are essential as they are the a key way to evaluate the performance of the public safety services. Dispatch is public as well and should remain that way.
I’m a public employee. My salary is listed on a website. My investments, debts, certain assets, outside activities, and the same for my spouse are a FOIA away.
If you don't want a source used in civil courts you can just tell the courts they are't allowed to use it.
That's what is done for accident investigation. Accident investigators need to understand what actually happened, not just a bunch of useless backside covering as everyone tries to avoid getting sued. So, their evidence just isn't admissible in civil courts. An investigation by, say, the NTSB or analogous organisations in many developed countries produces a public report, but you can't take that report into a civil court because it says in the law and on the report that it can't be used that way.
It would probably take more effort to ensure this was effective in the revenge-hungry United States of America than in countries where the mainstream zeitgeist incorporates the understanding that hurting somebody else doesn't fix anything, but it could be done.
At first Seattle tried editing the footage before release to redact parts that might violate the privacy of anyone other then the officers and alleged criminals. I don't remember how many officers were involved in the initial body cam deployment, but it was enough that they generated so much footage that they didn't have the budget to edit it. I remember at one point they were even asking hackers for help with automating some of this.
A couple years or so later, they resumed using body cams. I'm not sure how they dealt with the privacy issues.