qnleigh
a year ago
> I would have thought about privately disclosing these findings to Dotpe. But all the API requests are right there in plain sight...
There are pretty common ethical standards about disclosing vulnerabilities privately before disclosing them publicly. I don't see how the obviousness of the vulnerability changes the situation. By warning the company, you give them the opportunity to remedy the problem before announcing to the world that anyone with a laptop can exploit it. Probably they were just hoping that nobody would notice, which is stupid of course, but now they don't have the chance to build up a better wall before the flood of fake orders that could cause real harm to the small businesses whose financial information you disclosed online.
Perhaps I'm being too optimistic about how the company would respond, but I still think it's hard to justify not doing a private disclosure.
mainframed
a year ago
Adding to this, in some countries he is already past the gray-area to what constitutes as computer fraud.
Pissing off the company, whose systems you accessed without authorization, is one way of getting to experience the full force of the justice system.
moralestapia
a year ago
Curious.
How is this, specifically, fraud?
nkrisc
a year ago
The gist of some various laws around the world is that simply obtaining credentials does not authorize you to access the system, and accessing it without authorization is the illegal part.
drdec
a year ago
This principle is clear if you apply a real world analogy. Just because you happen to have keys to a building doesn't mean you can enter without authorization from the owner. (E.g. you may have kept copies after a lease expires or a sale, it maybe you found them, etc.)
whamlastxmas
a year ago
Considering it’s a API available without any authorization, the better comparison would be walking around on unfenced private land. There’s nothing to indicate they don’t want people on it but it’s also obvious it’s private land.
ryathal
a year ago
It doesn't matter. It's still just as illegal to get into an unlocked car or one with wide open doors without permission. The same premise applies to computers in a lot of places, access controls don't matter. If you access something on a computer not indented to be accessible, it's considered a crime.
paledot
a year ago
Is it illegal, in fact? If a cop saw you, you'd be arrested and prosecuted for attempted auto theft, and your "I just wanted to see how comfy the driver's seat was" defense would ring hollow in court. But sitting in an unoccupied car without authorization isn't trespassing unless it's parked on the owner's land, and I'm not sure what other laws would apply to that specific act.
hunter2_
a year ago
Walking around isn't usually a big deal until told to leave (verbally or by way of conspicuously posted signs), since that is a prerequisite to trespassing. Otherwise, delivery people would operate in a gray area which would be very problematic for them, since not all deliveries are requested by the recipient/owner.
However, although you are free to walk around in search of the front door, you can't start eating the fruit off the trees. Perhaps that's the better analogy: the trees are happy to serve up a delicious treat for anyone requesting something of it, but that doesn't mean the tree sets the rules. Just because fences preventing this are popular doesn't make them compulsory.
moralestapia
a year ago
I get the unauthorized access argument.
But how does it become fraud?
esrauch
a year ago
Defeating access control by using credentials that aren't yours is fraud.
Like, if you found a company badge laying around, go to that office and flash the badge to the security guard and go in. You've committed fraud by tricking the guard into thinking you're authorized to enter when you weren't.
moralestapia
a year ago
I see, thanks.
No credentials involved here, though.
hunter2_
a year ago
TFA mentioned sending requests with a table number that the sender was not at. That is hardly any different from the idea of showing a badge that wasn't issued to you. The ease of spoofing doesn't matter at all, in the eyes of such laws.
The same could be said about typing any URL that wasn't knowingly supplied to you by the owner, but a "reasonableness test" in court would sort those out from nefarious activity.
nkrisc
a year ago
The question a judge (or jury) would answer is: would a reasonable person think they had permission to access it?
API documented on the website under a section called “For Developers”? Probably, yes. API reverse engineered by intercepting requests? Probably not.
Note that the blog was taken down before I could read it myself.
tetha
a year ago
Interestingly enough, the very lawsuit-happy nature of a major german party has "backfired" quite a bit recently. A security researcher was found not guilty of circumventing security measures or accessing authorized computer systems or resources without authorization, because there were no security measures or authorization on the API to circumvent.
Though note that this would not help one if one started to use or abuse the API to get free food or cause financial damage to a restaurant through fake orders. For example, ordering the corn soup through the API could really backfire if someone wants to present it as good old fraud or theft, or if the recipient of the unexpected soup got into trouble and started to look for someone to hand the damages to.
cornel_io
a year ago
People have been convicted of hacking for merely editing URL strings, under the theory that were knowingly accessing systems in ways that they were not supposed to. This would be similar.
Whether or not that seems reasonable to us is a different matter, but basically it boils down to the fact that "they left the door unlocked" doesn't make it legal to walk in.
ensignavenger
a year ago
I believe the conviction of which you are thinking was overturned on appeal, though.
rvnx
a year ago
He is in India:
If any person without permission of the owner or any other person who is incharge of a computer, computer system or computer network
- (a) accesses or secures access to such computer, computer system or computer network or computer resource;
- (b) downloads, copies or extracts any data, computer data base or information from such computer, computer system or computer network including information or data held or stored in any removable storage medium;
[...]
- (e) disrupts or causes disruption of any computer, computer system or computer network;
[...]
- (g) provides any assistance to any person to facilitate access to a computer, computer system or computer network in contravention of the provisions of this Act, rules or regulations made thereunder;
If any person, dishonestly or fraudulently, does any act referred, he shall be punishable with imprisonment for a term which may extend to three years or with fine which may extend to five lakh rupees or with both.
====
Though, I prefer a lot the poster of the blog post than the company...
ceejayoz
a year ago
By the sensible definition, it isn’t.
By the definition that killed Aaron Swartz, it probably is.
AdamN
a year ago
He wasn't prosecuted for logging in and looking around. He overtly did copyleft type things like finding ways to take copyrighted journal articles and release them into the public domain. Overzealous prosecution for sure regardless.
ceejayoz
a year ago
https://en.wikipedia.org/wiki/Aaron_Swartz#United_States_v._...
> On July 11, 2011, he was indicted by a federal grand jury on charges of wire fraud, computer fraud, unlawfully obtaining information from a protected computer, and recklessly damaging a protected computer.
> On November 17, 2011, Swartz was indicted by a Middlesex County Superior Court grand jury on state charges of breaking and entering with intent, grand larceny, and unauthorized access to a computer network.
> On September 12, 2012, federal prosecutors filed a superseding indictment adding nine more felony counts, increasing Swartz's maximum criminal exposure to 50 years of imprisonment and $1 million in fines.
The only civil copyright proceedings were JSTOR settling with him out of court.
mainframed
a year ago
I don't know Indian laws. But this wikipedia page [1] gives a list of types of computer fraud in the US under the CFAA:
Types of computer fraud include: * [...] * Accessing unauthorized computers * [...]
He accessed their computers to access purchase information of other people (e.g. his friend) and business data. I guess making it public, thereby damaging the companies reputation and potentially getting sued by their lawyers is one way to find out, whether he was "unauthorized" to do so.
ryukoposting
a year ago
Well, the link leads to a 404 so it seems like the author has been convinced.
ffsm8
a year ago
> 37,529 restaurants use Dotpe for QR codes.
At that scale, it would take years to get fixed without forcing it like this.
It's too small for them to care about the liability of security and too large to move quickly
loktarogar
a year ago
> At that scale, it would take years to get fixed without forcing it like this.
But it also might not take years. The point of responsible disclosure is to give them the opportunity. If they don't take it, fine - that's now on them.
Instead this guy is committing fraud with actual financial damages (wasted food) and then sharing how others can commit the same fraud on a massive scale, potentially causing more damage. This is now on him and Dotpe, not Dotpe alone.
hunter2_
a year ago
> fine - that's now on them
Is that legally true? The legal risk of having published this without responsible disclosure vanishes if the conventional period of opportunity is ignored? That smells fishy.
loktarogar
a year ago
I'm not a lawyer so I'm not speaking to the legality of this. Legality is not the only thing that matters.
If you publish an exploit without at least making an attempt to fix it, and someone follows you and exploits it, then there's a direct moral line between you and that exploitation. They more likely than not wouldn't have exploited it without you putting that published info in their path. It's now on you, them and the company, morally. Any damages that result from this interaction are because you and the company enabled them to happen.
That's different to someone else stumbling on it and exploiting it. That's purely on the company and the exploiter.
chippiewill
a year ago
Maybe, maybe not.
But in responsible disclosure you usually give a 90 day notice period before publicly disclosing and "forcing" them.
tomalaci
a year ago
In this case? Nope. This must be treated as willful design decision to open up API to entire public (including PII/phone-number leak as per design), even if they say they totally didn't meant that to happen. Government itself should then be notified to go after these guys for failing to do the most basic access controls.
I mean, come on! To treat this as a proper security vulnerability just gives too much leeway for these fast-and-loose businesses/systems. It will just encourage more such crap to proliferate.
I am with the author on this one, I am fairly certain the issue of this was raised internally already, probably multiple times. Fortunately for the business, their management did the right decision - focus on quick and easy features, security is a non-issue, we will just blame the hackers and have legal channels deal with them. I mean, you even have people here berating someone uncovering gross negligence for Google-backed company. Why would businesses bother with basic security when they can play the victim so damn easy?
mpeg
a year ago
Author is in India, I would be very careful because it's much more likely the government will prosecute them for unauthorised access and irresponsible disclosure than do anything to the company.
Truth is even in the west this kind of irresponsible disclosure could land you in jail, much more so in a developing country where these laws are all relatively new.
krsdcbl
a year ago
Fully agree with you!
The API being unauthd is clearly a core design choice, and finding out any customer or service data is openly accessible with consecutive numbers through that API is not a zero day or something.
There is no "responsible disclosure" to be made here, going to the company and explaining what's the issue with all of this amounts to "handing out free consulting" if anything
mpeg
a year ago
Unfortunately that is not how the law works, at least in most countries. As soon as you enumerate ids regardless of whether there is any security in place it is unauthorised access and it's illegal.
komali2
a year ago
Right, I believe they're posting as if the ethical standpoint is normalized, to further highlight the absurdity and injustice of the current legal framework.
tsimionescu
a year ago
Why is it unjust to prosecute people who harm a business and unsuspecting customers of that business by disclosing 0-day vulnerabilities publicly without giving them even a chance to patch?
The poster here has no proof that the vulnerability was already being exploited. For all we know, as obvious as this was, no one else had yet thought to look.
This is like going around people's house doors, testing to see if they are unlocked, and if they are, posting a big sign saying "unlocked door" on each one. It's obviously an anti-social act masquerading as benevolent, and it should be punished. Of course, the company running such highly vulnerable code should also be punished, but that doesn't absolve anyone.
komali2
a year ago
Your metaphor is a big stretch. We're talking about a business and the expectations we should have for businesses.
Noticing an overhead pipe at McDonald's is dripping onto the griddle and pointing it out to people isn't harming the business, it's pointing out the business' gross negligence.
tsimionescu
a year ago
I don't agree with your alternate metaphor. In your example, publicly pointing out the leaking pipe can't cause any damage to the existing clients. In this case, publicly pointing out an exploitable vulnerability that gives access to personal information does bring extra harm to the customers.
If you want, a more apt comparison might be going around a business park and sticking big signs on every unlocked archive door you find. The companies not properly locking the doors are at fault, and customer data may already have leaked; but, you are virtually guaranteeing that even more customer data will leak by doing this. It should absolutely be illegal.
pnt12
a year ago
Seems outlandish. Citation needed? I'm aware of a couple of cases in the US, but not all over the world.
Secondly: can consumers be blamed for gross negligence? It's not reasonable for a bank to post account balances in public billboard and ask people not to look at others. We should contest when private data is available publically, hidden only by small obfuscations, not professional security practices.
mpeg
a year ago
So for example in the UK with the computer misuse act, intent matters. If you intentionally change an id because you expect you will be able to access other data it becomes a crime.
Your example is flawed because in this case the private data was not made available publicly at all – you need to intentionally exploit a software flaw to access it.
Of course, it also matters how you handle it. If you do enough to just discover the flaw, try to adhere to the bug bounty program scope (if any), use your own accounts in testing and responsibly disclose any findings as soon as you have a poc then you'll probably be ok.
In this case the author went way beyond just finding the flaws, and then disclosed it publicly in a completely irresponsible way without even trying to contact the company or any of the clients affected by it (some of which will certainly have a security contact that can liaise with the vendor)
pnt12
a year ago
I concede that intent matters.
Maybe a better analogy is a bank with open lockers and no vigilance: if someone enters and steals money, the police will look for them, because "the coffers were open" is not a valid defense. But customers will also demand answers from the bank - why were they so negligent and incompetent that someone can just enter and get their money?
We should hold similar values for digital systems.
Was the author's intent on stealing private data and causing harm? Did he gain from this abuse? Did the company take enough measures to safeguard their data?
Companies have been mostly not held responsible for their fuck ups, and no matter the law, that's wrong to me.
tsimionescu
a year ago
There are exactly two activities you can be participating in if you are exploring someone else's undocumented API: (1) free consulting, or (2) illegal hacking. Disclosing vulnerabilities you found in someone else's product, regardless of how obvious, is free consulting. If you're not responsibly disclosing them, then you were illegally hacking their systems.
thinkingemote
a year ago
Just because someone or something is unethical doesn't mean we should be unethical as a response.
We shouldn't limit ourselves to only be responsible and disclose properly when the vulnerability suits us.
That is both unfair and irrational.