Guessing passwords is an attempt to access privileged information you have no right to access, and could not otherwise access without bypassing security measures.
Guessing a URL is an attempt to access (potentially) privileged information which was not secured or authenticated to begin with.
A password is a lock you have to break. An unlisted URL is a sticky note that says "private" on the front of a 40" screen. It's literally impossible for that information to stay private. Someone will see it eventually.
same as this famous case, in which a supreme court justice is asked "what is and is not pronographie" - of course he realizes if he defines "what is not" people are going to make all kinds of porn right on the boundary (see: japanese pronographies where they do the filthiest imaginable things yet censor the sensitive books, making it SFW in the eyes of their law). this judge avoided that.
Anyways, parallel to the fact that filthy pronographies can be made a gorillion different ways, a "hack" may be manifested also a gorillion different ways. Itemizing such ways would be pointless. And also in the same vein, strictly defining a black and white line "this is legal, this is not" would cause hackers to freely exploit and cheese the legal aspect as hard as possible.. businesses and data miners and all these people would also freely exploit it, at massive scale and with massive funding, since it is officially legal. Thusly it must be kept an ambiguous definition as with pronographies, as with many things
Do you think the current line, where it's based on you "knowingly" exceeding your access or deliberately damaging the operation of a computer system, is excessively vague?
The question can be easily inverted for the other side: if any user accidentally damages a service's functionality in any way, can they always be criminally liable? Can this be used by companies with no security or thought put into them whatsoever, where they just sue anyone who sees their unsecured data? Where should the line be drawn?
To me, this is subjective, but the URL situation has a different feel than something like SQL injection. URLs are just references to certain resources - if it's left unsecured, the default assumption should be that any URL is public, can be seen by anyone, and can be manipulated in any ways. The exception is websites that put keys and passwords into their URL parameters, but if we're talking solely about the address part, it seems "public" to me. On the other hand, something like wedging your way into an SQL database looks like an intrusion on something private, that wasn't meant to be seen. It's like picking up a $100 bill of the street vs. picking even the flimsiest, most symbolic of locks to get to a $100 bill you can see in a box.
>The question can be easily inverted for the other side: if any user accidentally damages a service's functionality in any way, can they always be criminally liable? Can this be used by companies with no security or thought put into them whatsoever, where they just sue anyone who sees their unsecured data? Where should the line be drawn?
I don't think the question can be inverted like that, not meaningfully anyway. The CFAA specifically requires one to act knowingly. Accidentally navigating to a page you're not supposed to access isn't criminal.
>To me, this is subjective, but the URL situation has a different feel than something like SQL injection.
I don't think the url below is necessarily that different.
> GET wordpress/wp-content/plugins/demo_vul/endpoint.php?user=-1+union+select+1,2,3,4,5,6,7,8,9,(SELECT+user_pass+FROM+wp_users+WHERE+ID=1)
> if it's left unsecured, the default assumption should be that any URL is public, can be seen by anyone, and can be manipulated in any ways
It can be, but not lawfully so. It's not possible to accidentally commit a crime here, for example in the IRC logs related to the ATT case the "hackers" clearly understood that what they were doing wasn't something that AT&T would be happy with and that they would likely end up in court. They explicitly knew that what they were doing was exceeding authorized access.
> On the other hand, something like wedging your way into an SQL database looks like an intrusion on something private, that wasn't meant to be seen
I think you've reached the essence of it. Now, let's say you just accidentally find an open folder on a bank's website exposing deeply personal KYC information of their customers. Or even better, medical records in the case of a clinic.
Lets say those files are discoverable by guessing some URL in your browser, but not accessible to normal users just clicking around the website. If you start scraping the files, I think it's pretty obvious that you're intruding on something private that wasn't meant to be seen. Any reasonable person would realize that, right?
> GET wordpress/wp-content/plugins/demo_vul/endpoint.php?user=-1+union+select+1,2,3,4,5,6,7,8,9,(SELECT+user_pass+FROM+wp_users+WHERE+ID=1)
This is why I tried to make the clarification that I was referring to the address part of the URLs only, not the parametrized part. In my mind, something like /users?key=00726fca8123a710d78bb7781a11927e is quite different from /logins-and-passwords.txt. Although, parameters can also be baked into the URL body, so there's some vagueness to this.
> I think you've reached the essence of it. Now, let's say you just accidentally find an open folder on a bank's website exposing deeply personal KYC information of their customers. Or even better, medical records in the case of a clinic.
I guess if I try to distill my thoughts down, what I really mean is that there should be a minimum standard of care for private data. At some point, if being able to read restricted data is so frictionless, the fault should lie with the entity that has no regard for its information, rather than the person who found out about it. If a hospital leaves a box full of sensitive patient data in the director's office, and getting to it requires even the minimal amount of trespassing, the fault is on whoever did so. But if they leave that box tucked away in the corner of a parking lot, can you really fault some curious passer-by that looked around the corner, saw it and picked it up? Of course, there's a lot of fuzziness between the two, but in my mind, stumbling into private data by finding an undocumented address doesn't clear the same bar as bruteforcing or using a security vulnerability to gain access to something that's normally inaccessible.
So if I deliberately exploit a bug on your website and download your customer database by typing things in my browsers URL bar, I should not be prosecuted?
Cyber attacks are consentual, digital engineering is the only discipline where we have complete mastery of the media. If you make a system (or authorize it) what someone does with it is your fault.
Passwords are different from URLs because URLs are basically public, whereas passwords aren't supposed to be. Furthermore, this is not 1995. Everyone who is in the industry providing IT services is supposed to know that basic security measures are necessary. The physical analogy would be, walking through an unlocked and unmarked door that faces the street in a busy city, versus picking a lock on that door and then walking through it.
> Everyone who is in the industry providing IT services is supposed to know that basic security measures are necessary.
And everyone who doesn't have wool for brains knows to not carry large rolls of cash around in a bad part of town, but we can still hold the mugger at fault.
Nevertheless, URLs are as public as door knobs. If someone is merely observing that a door is unlocked and they have not stolen anything, they have done nothing wrong. People being prosecuted over discovery and disclosure of horrible design flaws based on URLs should never be prosecuted. If they use the information to actually cause damage, we can be in agreement that they are responsible for the damage.
>People being prosecuted over discovery and disclosure of horrible design flaws based on URLs should never be prosecuted. If they use the information to actually cause damage, we can be in agreement that they are responsible for the damage.
Are you sure? I seem to remember people getting burned for publicly disclosing security vulnerabilities after stubborn agencies refused to fix them for years. Stuff like, exposing thousands of SSNs through a public gateway... We are literally having this discussion on URLs because of famous cases where people DID face unfair treatment. I don't recall any actual fix for this legal chicanery either. If you do, I would be very interested.
> I seem to remember people getting burned for publicly disclosing security vulnerabilities after stubborn agencies refused to fix them for years. Stuff like, exposing thousands of SSNs through a public gateway..
This has never happened in the US on the federal level. Unless your definition of "getting burned" is a nasty email from a clueless non-LE government worker.
> We are literally having this discussion on URLs because of famous cases where people DID face unfair treatment
I don't think any reasonable person can read through the court filings in those (Auerheimer, Swartz) cases and agree with the claim that there was unfair treatment wrt the application of the CFAA, or that the CFAA was unfair because it covers those cases.
I totally understand how someone who has not spent time familiarizing themselves with the actual details of the cases might be under the opposite impression; they are frequently misrepresented by people with agendas and nerds who mistakenly understand judicial process as a "Captain Kirk vs Computer" scenario.
There's a trend in communities like HN to claim that the CFAA is bad because Swartz deliberately broke the law while he engaged in some pretty cool civil disobedience. That's not reasonable. Two things can be true at once: what Swartz did was in fact cool and laudable, it still shouldn't be legal. Similarly, a reasonable person might consider it cool and laudable to punch a nazi, doesn't mean it should be legal.
In any case, there's also a trend of misrepresenting the potential penalties involved. On HN, you'll see people posting about how Swartz was facing 30 years in prison, which is an outright lie. Swartz had, in fact, behaved as described in the indictment; he had two plea deals on the table. One for 6 months with the opportunity to argue for further leniency from the judge, and another for 4 months outright. Lawyers familiar with the case have stated that it was very likely that he wouldn't have gone to prison at all.
Swartz killed himself, so the CFAA must be bad, but it's probably realistic to assume that Swartz did not kill himself because he was scared of spending a few months in prison. He was likely seriously mentally ill, and a victim of the poor state of the US healthcare system, not of the CFAA or the DOJ.
Have a look at this: https://m.slashdot.org/story/180969 Clearly there is precedent in the US and elsewhere for prosecution based on accessing URLs. Politicians have even argued that you should not be able to tamper with website source code in your browser, or otherwise use websites in any way not anticipated by the owners.
>Two things can be true at once: what Swartz did was in fact cool and laudable, it still shouldn't be legal. Similarly, a reasonable person might consider it cool and laudable to punch a nazi, doesn't mean it should be legal. We live in an age where people call other people Nazis as if uttering the accusation gives them a free pass to infringe on the rights of those people. Even if it could be proven to be true, the facts would not grant them any such right.
Theft and unprovoked assault are neither cool nor legal. I don't care if we're talking about absolute assholes, either. For each and every one of us, there is probably someone in the world who thinks that they should have our property and have the right to attack us.
>Swartz killed himself, so the CFAA must be bad, but it's probably realistic to assume that Swartz did not kill himself because he was scared of spending a few months in prison. He was likely seriously mentally ill, and a victim of the poor state of the US healthcare system, not of the CFAA or the DOJ.
Idk that much about this case. It seems to me that Swartz was in a hell of a lot of trouble, that could have gotten him a prison term along with financial and career ruination. He clearly should not have killed himself but that kind of stress can make people lose sight of the future.
> Have a look at this: https://m.slashdot.org/story/180969 Clearly there is precedent in the US and elsewhere for prosecution based on accessing URLs
"prosecution based on accessing URLs" is a dishonest way of describing this case. It's a prosecution based on accessing URLs with malicious intent, while the persons responsible knew they were not intended to access said URLs.
That's like saying "prosecution based on walking through a doorway". Well yeah, except it was the middle of the night and the door to someone else's house had been accidentally left unlocked.
>Idk that much about this case. It seems to me that Swartz was in a hell of a lot of trouble, that could have gotten him a prison term along with financial and career ruination. He clearly should not have killed himself but that kind of stress can make people lose sight of the future.
Swartz had good lawyers; he was certainly aware that he wasn't in big trouble. He was facing neither financial nor career ruination. The damages he had caused were far from enough to result in financial ruination. The charges had made him even more of a celebrity and would've been a boost to his career.
Orin Kerr, a top subject matter expert addressed this extensively. In the second part he also offers the best criticism of CFAA, which is that it's almost entirely redundant given the existence of very broad wire fraud statute.
And for what it's worth, Swartz had been spending a lot of time thinking about suicide for years before the whole JSTOR debacle http://www.aaronsw.com/weblog/dying
I can't say I've ever struggled to make this determination, but I don't make a habit of trying random ports, endpoints, car doors, or brute-force guessing URLs.
But it was very tempting when i saw that my national exam results were sent to us in a mail as nationalexam.com/results/2024/my-roll-number. Why would i not try different values in the last part.
Try it once to see if it works, you'll probably be fine.
Find out that it works, and then proceed to look up various other people? Whether you're fine depends entirely on whether or not you genuinely believe that you're supposed to be accessing that stuff.
But when it's just a sequential-ish ID number, you have to accept that people will change the ID number. If you want security, do something else. No prosecuting.
What about guessing passwords? Should someone be prosecuted for just trying to bruteforce them until one works?