3 ms·
I'm guessing that sneak wasn't referring to HTTP Basic Auth, but rather the fact that when an HTTP client says "GET", the server effectively says "HOW MUCH?" (
by nitrogen 14y ago
I'm guessing that sneak wasn't referring to HTTP Basic Auth, but rather the fact that when an HTTP client says "GET", the server effectively says "HOW MUCH?" (A crude allusion to the classic "jump/how high?") I think sneak is arguing that receiving a response to an isolated GET request implies that the GET request was "authorized."
Edit: after reading selections of IRC logs and e-mails from the indictment, my sympathy with weev has been reduced. Granted, an indictment will select the most salacious excerpts to make its case in the strongest way possible, but the conversations in the indictment are far from white hat (unless the defendants manage to convince the court that their talk of phishing and spamming was merely juvenile humor).
- gknoy 14y agoI wonder what one could get an expert witness to testify to with regards to the meaning behind the HTTP protocol with respect to 'authorization'.
- sneak 14y agoSergey Bratus[1] (of Langsec[2] fame) testified as an expert witness for the defense in this very way within the last few days. [1] http://www.cs.dartmouth.edu/~sergey/ http://www.cs.dartmouth.edu/~sergey/ [2] http://www.cs.dartmouth.edu/~sergey/langsec/ http://www.cs.dartmouth.edu/~sergey/langsec/
- tptacek 14y agoDo you really buy this line of argument? How many banking applications configure themselves so that they rely on the intended meanings of HTTP verbs and authorization headers as their primary overt security mechanism? And of those, how many do so correctly? I get why Bratus would testify. The defendant here needs all the help he can get and is morally entitled to the best case he can possibly present. I respect and admire everyone who is trying to help him out. But presuming he's not guilty of a real conspiracy to defraud anyone, I have a hard time believing it's because AT&T's web application vulnerability entitled the public to their database.
- sneak 14y agoI do. If there is any crime here (I don't think there is even one, FYI) it's AT&T not taking adequate measures to safeguard their customers' PII. I don't think trafficking in any information should be a crime, though (unless it's the government - an asymmetry is necessary there), so I don't think a criminal trial is in any way justified.
- nitrogen 14y agoI think it's reasonable for them to argue that AT&T's server's willingness to give them the e-mail addresses means that obtaining the addresses was not illegal, and that despite mulling over the darker possibilities available to them, by choosing not to put the e-mail addresses to illegal use they committed no crime. They could argue that they should not be convicted of conspiracy because they ultimately decided not to abuse the list of addresses.
- tptacek 14y agoI would like it to be harder than it seems to be to prove conspiracy to commit fraud. I don't think I'd like it to be harder than it seems to be to prove unauthorized access. I know that's the opposite of what most nerds like me want, but I think we're well served by a very broad definition of unauthorized access, and we're poorly served by vague conspiracy laws in more places than just online. Note that under the US Code, you need both elements. Just plain unauthorized access isn't a federal crime; you need an intent to defraud.
- nitrogen 14y agoI know that's the opposite of what most nerds like me want, but I think we're well served by a very broad definition of unauthorized access, and we're poorly served by vague conspiracy laws in more places than just online. I do agree with you regarding conspiracy, but you are right that I would in principle prefer to have every Internet-facing system as robustly secured as if it had been independently reviewed by you, cpercival, and the people who wrote the space shuttle's software. A small part of the reason I want this is so that absolutely anybody can confidently write and deploy scraping software that collects and analyzes information in new ways (e.g. IBM Watson, better search engines, or some other as yet undiscovered idea).
- tptacek 14y agoI understand what he's saying, but a legal defense based on the response to an HTTP GET constituting de facto permission to access whatever that URL addresses is going to be disappointed. I'm not saying the defendant here is guilty; I'm just saying, he's not not guilty simply by reason of HTTP. As a matter of law, the onus is not on the operators of web applications to ensure that their defenses are effective. If you do something stupid in your web app, but it remains clear to me that I'm not meant to access whatever your stupid app failed to prevent me from accessing, I can still be charged with a felony for doing it. Now, if whatever it was that you tried and failed to protect wasn't obviously protected, so that I can argue in court that a reasonable person would have believed they did have permission to interact with it, I'll probably win in court. The prosecution will no doubt collect all my emails and any other piece of evidence it can find to build a case that I knew I wasn't supposed to hit your site that way. If you were especially dumb, and built an application that not only exposed sensitive information, but did so in a way that any normal person would just have assumed that sensitive information was theirs for the taking, the prosecution will have a hard time making their case.