Hacker News Reader

Comment by xyzzy123 | Hacker News Reader

I believe this gets to the crux of the argument. Broadly, I think there are two views on why the U.S. has laws regarding "unreasonable" search and seizure:

a) Being searched all is unreasonable because it inconveniences those being searched, takes up their time and involves a degree of forced submission.

b) Being searched all the time is unreasonable because it grants the government too much power over the individual. The government should not be able to snoop into the lives of private citizens without good reason. True freedom and perfect enforcement of crime are incompatible.

You can hold (a) or (b) or both. Digital searches happen quietly and invisibly, so is it OK to be constantly searched if it involves no personal inconvenience?

I believe you hold mostly position (a), whereas those who believe this program is bad (including me) hold (a) and (b).

View on HN

Point (a) isn't sufficient to understand the 4th amendment.

For example, the Supreme Court in Kyllo vs. U.S. decided that using a thermal imaging device to scan houses to find indoor marijuana farms is an unreasonable search, and reversed Kyllo's conviction for growing 100 marijuana plants.

You'll easily note that scanning a house from the street does not inconvenience the person being searched, take up their time, or involve a degree of forced submission.

For another example, the Supreme Court said that attaching a tracking device to a car without a warrant is also an illegal search. This also does not fit the criteria of (a). (While it immeasurably slightly worsens mileage, the court decided that it was a form of trespass. One does not need to be present in order for trespass to occur.)

For a more historical example, Ex Parte Jackson (1878) says "Letters and sealed packages subject to letter postage in the mail can be opened and examined only under like warrant, issued upon similar oath or affirmation, particularly describing the thing to be seized, as is required when papers are subjected to search in one's own household. The constitutional guaranty of the right of the people to be secure in their papers against unreasonable searches and seizures extends to their papers, thus closed against inspection, wherever they may be."

Again, 4th Amendment protection of letters exists even when the conditions of (a) are not held.

This is the same reason I find disclosure of geolocation data collected outside a call without a warrant to be an unreasonable search.

Could you be more explicit on which reason is this "same reason"? I listed several reasons.

Do you think law enforcement use of E911 pings without a warrant is trespass on your possessions, which was the reasoning behind the US vs Jones (2012) case I mentioned?

In any case, United States v. Graham disagrees with you. Geolocation data is "information voluntarily disclosed to a third party" and so "ceases to enjoy Fourth Amendment protection."

What do you find wrong with the reasoning in the Graham ruling?

In reviewing your other comments, it sounds like you also disagree with the conclusions of pages 47-56 of this most recent opinion by Judge Leon, where the judge describes how collection now is different than from a few decades ago, and in a manner that the Supreme Court has indicated other cases indicated may have substantial 4th amendment concerns.

Why specifically do you disagree with the reasoning behind his opinions?

Obviously US v. Graham (US District Court for Maryland) and Leon's opinion (US District Court for DC) are at odds. Leon points that out in his own opinion, on p66. It's going to take the Supreme Court (or a big change in the law) to sort this one out for the US.

I would argue that the e911 data is not a voluntary disclosure. The cellular companies were compelled to generate and provide this information, as well as generate and store hordes of other information to assist law enforcement. In short, I don't think the government should be able to compel a company to store a class of data about its users, then because now they are holding it (at the behest of the government), then strip the data of privacy and compel the cellular companies to provide it to law enforcement.

I don't find a substantial difference between the police planting a tracking device on your vehicle, and requesting geolocation data records for your mobile phone, they achieve largely same goals, thru only what I consider marginally different means - meaning, both should require a warrant.

The exception to that would be geolocation data that is generated as part of making a phone call. I want to hold the line on what can be monitored to the era before always on mobile communications devices. Just because the data is there, does not mean it should be able to be gotten without a warrant.

i think I find your reasoning flawed, as a matter of personal opinion at least.

> Who you called, who called you, how long you talked - this stuff, which is otherwise known as call detail records it available to nearly everyone who works for the telco and is not really what I would consider private.

I see what you're saying, as applied by the Smith v MD, that the information in a PEN register is fair game, but even in Smith, the collection of records had to be manual, which prevented its collection and use en masse, meaning that its application was self-limiting. Sure, the police could probably collect all the PEN registers, but as most of it would be useless, why would they?

Regarding your earlier statement though, for some perspective, had the British been able to keep tabs on the metadata of the populace, our founding fathers would have all been hanged for treason, and America would never have declared independence.

I don't mean to suggest that being subjects of Britain would be necessarily bad, but I think it's fair to suggest that our founding fathers considered their actions just, and as a matter of course, would have likely considered actions that would have resulted in their death equally unjust. If you agree, I think that should have some bearing on your interpretation of the fourth amendment.

The collection of records has not been manual since sometime in the 80's - the switch can generate a report of all this data. But I get the point.

I think metadata alone is useless - it gives you associations, not content, you could be calling someone not very nice to talk about the fact that they owe you money for work you did on their car - or about blowing something up. All this metadata is useful for is generating leads for traditional police work (aka shoe leather) to follow up on and dig up information the old fashioned way.

I'm confused as to why you'd be against the collection of GPS metadata then. GPS data is, effectively, only associative metadata. It can tell you where you were, and when you were there, but not what you were doing. That's why I'm having a hard time finding consistency in your position.

That isn't meant as judgement of you, by any means... the entire issue is a morass. We want law enforcement agencies to be effective (assuming that we're not acting criminally), but we don't want them to be overseers.

I personally think that the precedent is for us to err on the side of liberty over justice, and I think that in the spirit of presumption of innocence and all that, but that's influenced by confirmation bias, at the least.

It depends on the KIND of GPS metadata - the GPS that is generated when I call happens is attached to the call metadata. The Metadata that is generated just by having your phone on is not. I'm ok with things attached to calls, I'm not OK with blanket tracking of peoples locations.

The standard that I use is what information would you have gotten from a pen register in say 1973 or at least inferred from other sources of data, which is, calling party, called party, duration of conversation, location of parties in call, and really nothing more. Which leave out whole hosts of other kinds of data, like e911 pings, and tower data (showing all the users on the tower) and so on. Because much of this data is compelled to be kept by the government beyond what would be needed for engineering purposes (or in the case of e911 data generated by order of the government) it most certainly shouldn't be available without a warrant.

I find your argument, which is based on what the government compels others to do, to be oddly limited. You often focus on the phone call aspect of a phone, and compare it to a 1970s pen register. But that doesn't fit with modern phones.

An SMS message, or a smart phone check for new messages or to get map data, would under the 1970s technology parallel be implemented as a phone call. Under your logic, law enforcement would get all this information as well, no?

If so, most smart phones positively leak information. (Note that according to Judge Leon's opinion, it's not known if the metadata we're talking about includes data related to text messages. See p52, footnote 56.

If not, why not?

Also, the current system captures more information than a 1970s pen register. As footnote 57 points out, the pen register in the 1970s Smith decision did not keep track of if the phone call went through, nor the length of the call. It then points out the difference between building location from the 1970s, and caller location from mobile phones, with reference to New Jersey vs. Earls. 'the New Jersey high court found that "cell-phone location information, which users must provide to receive service, can reveal a great deal of personal information about an individual."'

So at least in New Jersey there is a distinction between the two.

The point in Judge Leon's opinion is that "This rapid and monumental shift towards a cellphone-centric culture means that the metadata from each person's phone "reflects a wealth of detail about her ... associations" ... that could not have been gleaned from a data collection in 1979." (page 54).

While you are entitled to your beliefs, I would enjoy it if you compared your beliefs to the actual opinion under discussion.

I should point out, the original post I meade that spawned this rather large thread, had no bearing on the decision - it was my opinion purely.

Something else I should point out - there were mobile (pre-cellular) phones in the 1970's and some geolocation data could be implied from their use - because in most cases they couldn't roam beyond their home area. For what its worth, some of this data could have been gathered as early as 1979 - when the first AMPS systems were being turned up.

Beyond that - I guess what I'm looking to say is, landline equivalence, I don't think you should be able to collect more information about a mobile phone user than a landline one, even if the information is there - at least not without a warrant.

My discussion here has really been about what is acceptable without probable cause and a warrant, and not just for the NSA - but for all Law Enforcement Agencies. I think Call Detail Records and Call Associated Geolocation Data is acceptable, but really, nothing else, not without probable cause and a warrant, I even think that using big data techniques on CDR's is novel and worthwhile - but it should be as a clue to find more information thru old fashioned police work, not as the end evidence.

Yes, I do believe I understand your argument. My question to you is three-fold:

1) the judge gives 10 pages of opinion, with 4 distinct points, on why land-line equivalency cannot be used as a guide. Can you speak to why you disagree with those 4 points?

2) Do SMS and app data count as "telephone calls"? Does an open TCP socket, which isn't in use, count as a "telephone call"? Under a landline equivalency doctrine, your answers must be yes. In that case, if most people's phones check for messages automatically every 5 minutes, then is there really that much of a difference than what we have now? Will your view change in the future when peoples' phones connect every minute, or have a continuous connection?

Also, since you mentioned the start of this rather large thread, there's another point. You said "I'm somewhat OK with big data being used in this case, these are cases that traditional police work cannot be used in"

Judge Leon pointed out that the government has provided no evidence that "big data" was used to solve any case, and that a government witness has stated that most of the times other methods could be used, only that metadata gathering is faster and cheaper.

3) Given that, how do you draw your opinion that traditional police work cannot be used in this case, when the government themselves cannot convince the judge?

> It depends on the KIND of GPS metadata - the GPS that is generated when I call happens is attached to the call metadata. The Metadata that is generated just by having your phone on is not. I'm ok with things attached to calls, I'm not OK with blanket tracking of peoples locations.

Maybe I'm fully mistaken here, but I think that's the exact same KIND of GPS metadata, which is that generated by tower triangulation. The way you phrase it, I interpret the latter type of GPS data as that which is generated by the internal GPS chip on a smart phone, but to be sure, the police can track the location of a dumb phone through triangulation when a phone is just 'on', without the requirement that calls be made, and the mechanism is the same.

Why is the former special, when the mechanics are exactly the same?

Because you have to do something beyond having your telephone ready to use, like making a call - and the call associated Metadata is a given, location is somewhat inherent to a landline.

> Because you have to do something beyond having your telephone ready to use, like making a call - and the call associated Metadata is a given, location is somewhat inherent to a landline.

Concerning cellphones, that is a distinction that only you make. Network-based location service doesn't make it, nor do the authorities, nor do the telcos, nor does your phone. Network-location can be done even on dumb phones, and if they can capture your location from making a call, they can capture your location without you making a call.

Regarding handset-based GPS technology also does not depend on whether or not you're actively making a call or not. Obviously, this excludes dumb phones that don't have GPS, but location tracking is done anytime your phone is on and the GPS radio is active. if the GPS radio is inactive, network-based location still works. If your phone is on, your GPS radio is active, if someone is tracking you based on that information, it makes no difference to them whatsoever whether your phone is making a call or not -- you are just as trackable either way.

Beyond this, again, I fail to see the distinction anywhere. If we're just talking about what's in Smith v MD, which was decided in 1979, then we can exclude all cell phone data whatsoever, as they weren't in use then. If we're talking about just landlines, then the content of the calls would be excluded anyway. Either way, none of it is necessarily conclusive without the content of the call itself, because anybody could have made the call.

I'm just confused as to where you chose to draw the line specifically, as there really isn't a line there. You're of course welcome to make the distinction however you like, but if you're basing it on some precedent, I'm afraid there isn't any. If you're basing it on technological limitations, then again, there isn't any. If it's just your personal preference on how trackable you think you ought to be, then I'm worried you're dramatically underestimating the implication of what you're saying.

For what its worth - I didnt read the decision because when I started this thread, I was commenting on an NBCNews article.

That said, I largely agree with the judge - I think I didn't know the full details - I didn't know the government kept historical data for 5 years. Based on the early reports, I presumed they kept the last 90-180 days on hand in an indexed database and if not needed it was aged out. So I made some assumptions of what was going on, and they were much more limited than reality.

I think that moment has hit all of us at some point. I remember seeing the first of the Snowden documents roll out, and thinking "Well, duh, of course they are."

It wasn't until probably months later that I realized how much greater the scope was in reality than even the worst of what I'd assumed... that this[1] had been built, basically, to accommodate all the data they were acquiring... that they had hooks not just into the telcos, but into the ISPs, and into service providers like Google, Facebook, et al.. It wasn't until well later that I realized that, now that the public has heard about this and exclaimed how against it they are, that nothing would be done without fierce political pressure.

Also, I'm betting the 5 years is just a target figure, or, more qualified, I bet what they actually mean is "at least" 5 years. Of course, having seen everybody involved with this lie at least once, and brazenly too. This is the first step in making all this go away, or at least paring it back to something reasonable. I'm not comfortable with every communication being tapped that the government just happens to be able to get their hands on.

[1] - http://en.wikipedia.org/wiki/Utah_Data_Center

I take it from your answer that you haven't read the court decisions I mentioned?

That's okay, I haven't read all of them either.

E911 data is a subset of geodata. US. v Jones concerns tower data, which the business captures as part of its normal business. This is not under mandate from the government.

"A majority of courts" (quote from p9) "have concluded that the acquisition of historical cell site location data pursuant to the Stored Communications Act‘s specific and articulable facts standard does not implicate the Fourth Amendment, regardless of the time period involved."

By your argument, since the phone companies are not compelled to store this historical geodata, then it's okay for the police to access these records.

"I don't find a substantial difference..."

Congratulations. But there is a substantial difference anyway. Trespassing on someone's car, in order to plant a tracking device, is quite different than, say, tailing someone to get the same information. The courts have said that the first is illegal. The second is definitely not illegal, even though it provides the same information.

I understand that you want a limited, very technical distinction on what information law enforcement cannot get from cell phone data. You base your argument on similarity to what land lines provide.

In doing so, you reject the 10 pages from this most recent legal opinion of Judge Leon which characterize why this similarity is not appropriate.

In doing so, you ignore other technical ways that law enforcement can and does get geodata from mobile phones even without using government mandated methods like E911. The government can bypass the phone company entirely and use IMSI catchers. By your logic, which only looks to see if the government is compelling a person or company to do something, then IMSI catchers are acceptable.

Would you kindly read some of the relevant court judgements, hopefully starting with this most recent one by Judge Leon, and comment with regards to them?

Correct. I largely hold position (a). I believe in position (b) as well, but there is a bar there, if something is above a certain level of invasion in invokes my belief in position (b) otherwise I consider it to be generally acceptable. Geolocation data is one of those things that invokes position (b) because I consider it to be an unreasonable invasion of privacy.

I just don't get it.

Based on parent comments, you're ok with the government knowing WHO/WHEN you call someone. But you have a beef with the government knowing WHERE you called from?

And your basis for that is that they've always been able to collect WHO/WHEN, but WHERE is different because it's new?

The geolocation data is not tied to placing a call, the cell carriers keep and have a record wherever my mobile terminal is whenever its turned on and registered. Its tantamount to putting a GPS tracker on my person. This to me is a totally unreasonable encroachment on my right to privacy.

The where attached to a placed call is a grey area for me. It's might be acceptable, it might not, because where is a given when using a landline.

Cool arbitrary line in the sand you're drawing there.

It's not really arbitrary - its based on what kind of monitoring you could do on a traditional telephone.

Reply on news.ycombinator.com