5 ms·
I guess another question is why we really need captchas. What are we trying to protect against that can't be accomplished with rate limits, voting systems, or o
by dheera 7y ago
I guess another question is why we really need captchas. What are we trying to protect against that can't be accomplished with rate limits, voting systems, or other ways to regulate meaningful use of a website?
Ultimately why does it matter if the user is a human or bot, as long as they are being a valuable user? What's wrong if a bot buys some of your inventory, pays for it and everything? What's wrong if an NLP bot responds to discussion threads with scientific facts and citations?
- lucb1e 7y agoIndeed, if they need to pay, there is no need for a CAPTCHA. As for responding to discussions, sure, you'd ban any human that posts spam the same as any bot. However, bots can spam your site faster than your human moderators can keep up with, so by using a CAPTCHA, only humans can post (ideally, of course), and thus moderators can keep up. As a security consultant, it is not uncommon to recommend a CAPTCHA for things like successive, three failed login attempts from a single IP address within a certain time period. But I do agree that CAPTCHAs are used too frequently, and some security people recommend them for just about everything. As someone who blocks a lot of tracking and feels the pain of these tracking monsters (that's what CAPTCHAs are these days, more than the Completely Automatic Public Turing test they're supposed to be), I always think very carefully whether a CAPTCHA is the only option, and I'm sure to recommend CAPTCHAs that fit the situation but are less invasive than a third-party one. Edit: oh, right, credit cards are common in many countries and banks set chargeback limits. It's still crazy to me that your 'public key' is also the only thing needed to withdraw money from your account, thereby necessitating a chargeback system. I guess for credit cards a CAPTCHA might be useful too.
- atomicUpdate 7y ago> if they need to pay, there is no need for a CAPTCHA > As a security consultant Hm? How does a security consultant not know that credit card numbers can be stolen and used by bots?
- lucb1e 7y agoSure, I know credit cards exist and that the numbers are stolen because they're so trivially easy to abuse, but payment systems are not very related to my work. I very rarely come across a product where I have to test payment features, and when I do, it's out of scope. Either the payment is handled by some third party (the usual case), or it has already been tested years ago and they're now asking to test some new feature (all other cases). More typical projects are testing traffic filtering solutions (firewall-like), blockchain startups (those are the worst), back-end (no payment) or b2b (contract-based payment, not online) applications... Even in the months that I was consulting at a bank, I never touched any sort of money system, there were a thousand other applications, services, websites, infrastructure things, and mobile apps to test. To give you a random example, they wanted to give people advice when buying a house through an app (where the final screen goes "and that's where $bank comes in: we can finance all this!"), so they had some external company prototype an app that was riddled with bugs in the login system. Or some internal service that POSTs data from one system into another. Or some API endpoint used for statistics. Etc. So the cases that I see are as a consumer, where I pay either by bank transfer (logging into my own bank's website), via iDeal (which also redirects you to your own bank's website to complete the transaction, but that one is instant instead of having to wait a working day), or sometimes via PayPal if that is the only option (I guess paypal do their own bot detection? No idea). So from my perspective, when I paid for something, the money is in the hands of the merchant and only customer support or a lawsuit would get it back.
- dangrossman 7y ago> What's wrong if a bot buys some of your inventory, pays for it and everything? 100% of the time, a bot buying things from a store is doing so to test a database of stolen credit cards the bot's owner has purchased/stolen. Accepting those sales means you'll get hit with chargebacks a few weeks later as the real owners of those cards see their statements. Then your store gets shut down for exceeding the maximum 1% chargeback ratio mandated by Visa and MasterCard. So preventing this scenario matters a lot, and when someone targets one of my stores for testing like this, enabling a CAPTCHA on the payment page is one of several, often-essential mitigations. Blocking IPs, blocking whole countries, including a nonce in the form, etc are on their own insufficient most of the time: the readily-available tools for this kind of attack already handle rotating IPs, retrieving a new form nonce on each try, spoofing the proper referrer, etc.
- oarsinsync 7y agoDo you have any stats on cart abandonment rate, and/or how that changed after you enabled recaptcha?
- dheera 7y ago"Cart abandonment" Why is this a bot-specific thing? I abandon shopping carts as a human all the time, including especially: * If you require a registration and login to checkout * If your UI is too hard or clunky to use * If shipping fees are higher than what I think is fair * If there are additional non-upfront fees
- luckylion 7y agoI think that was the parent's point: humans will likely abandon carts more often if they need to solve a captcha to advance to the check out.
- mrguyorama 7y agoOur company has industry leading fraud rejection rates and we don't use captcha at all
- founderling 7y agoHere is an answer from the hncynic bot: I had problems with my spinner at first. However, this is one of the things that is really annoying about using captchas instead of passwords. It is possible to create a secure account for something useless without using the system and you probably won't get spammers anymore. What a dumb idea, you would want to implement it yourself, because the people working and maintaining the system(s) will all have some way of doing that already. I don't know how much the government can take away from a site like this as well. But it's a bit like trying to ban a kid because the kid got an old friend on their facebook because his parents were "bad". So, even though you have a big idea about voting systems, it would be a better option to require a system that only exists to be able to be used for good reasons.
- SamuelAdams 7y ago> What's wrong if a bot buys some of your inventory, pays for it and everything? In the book "Spam Nation" (Brian Krebs), a group of students try to fight fake online pharmacies. To do this, they created an army of bots and placed thousands of fake orders every day. The goal was to create so many fake orders that the human processors (many fake pharma stores were not fully automated) had to spend a significant amount of time clearing out the fake orders before getting to the real orders. Now imagine this on a legitimate website. Not every website is automated, not every organization has the same resources that Amazon does. CAPTCHA's are a great way to ensure orders are coming from real people. They could still be faked, but the bar is a bit higher.
- aidenn0 7y agoDon't think bot; think botnet. Ratelimits do not work against botnets since they appear to be independent actors. e.g. if you think it's fine for everyone to do something 1-3 times, then you are letting a botnet of 10k hosts do something 10k-30k times. [edit] Also, NAT means that there could be hundreds or thousands of individual users on the same IP address (many dorms at smaller colleges are setup this way), so you don't want to rate limit by IP address either.
- dredmorbius 7y agoShouldn't open aggregate listing and blocking of botnets be effective? How many 10k node botnets are there? Also: egress hygiene should be a thing. Block subnets and ASNs if toxic behaviour is detected.
- dmoy 7y ago> How many 10k node botnets are there? About 10 years ago it was somewhere around 10 million bot computers. Obviously the distribution isn't uniform there, but that gives you an idea of the order of magnitude. Also that was 10 years ago, before smart fridges and tvs. So, possibly more now? I don't think "just block all the bots" is a feasible solution here.
- dredmorbius 7y agoNot how many bots, but how many botnets. The nets themselves take resources (time, effort) to set up and maintain. Presumably they're engaged (at least for the purposes of generalised website defence, though specific niches such as targeted commerce fraud may not apply) in systematic behaviour, which leaves signatures. Systemic cross-site collaborative detection. -- essentially what Google's CAPTCHA systems are, though there are others, such as CZ.NIC's Turris project -- could identify these, and via network-based domains of authority (ASN and CIDR assignments) assign reputations and target anti-fraud or anti-abuse countermeasures. Durable, privacy-respecting reputation tokens might be another approach. Present systems are far more primitive and reflect an earlier world-state.
- 7y ago