I’m getting really tired of “verify you’re not a bot” checks everywhere.
I understand these checks help prevent bots, but they seem to appear on so many websites now. Does anyone else find them frustrating?
I’m getting really tired of “verify you’re not a bot” checks everywhere.
I understand these checks help prevent bots, but they seem to appear on so many websites now. Does anyone else find them frustrating?
ImHo yes. I may have no proof, but I believe, bot farmers have lots of super fast, weak bots, that will be filtered by that, but they probably also have some fancy, slowly working bots, that will pass the verification. So, content inside will be visible to the content farmer anyway.
Even worse? if we filter out weak content farms, who may not have strong bots, that means we help strong farms get even stronger, into monopoly - that may be even worse than letting everyone get it.
Or we could get some super protections? with fail positive checks? Maybe login only, with irl meeting account verification prior to any access? would be madness. Paid access to the group? madness and possibly against fedi idea)… how could we protect self, against stronger and stronger bots?
Even if it doesn’t prevent scraping entirely, filtering out the “fast, weak” bots help mitigate server load. Sufficiently many bots hitting your site is indistinguishable from a DDOS attack.
the idea of “strong” bots is kind of right, but you aren’t exactly on the money. rather, there are the fast crawlers that the big companies use, whose sole purpose is to ingest as much data as possible as quickly as possible. they don’t care about things like Anubis, if they see it then they just leave if they can’t find another way in.
and then there are the personal agents. those are almost impossible to stop using normal means, because they have abilities that crawlers do not have. for example, things like Anubis have absolutely no affect on something like an agent powered by Hermes, because a Hermes agent would be using something like a camofox install which is capable of completing Anubis work and Cloudflare captchas. Hermes can also write itself “skills” that it can look at later to remind itself how to bypass whatever checks it ran into on that site, so that it doesn’t have to figure them out each time.
naturally, a human setting up anti-bot measures is at a disadvantage in a scenario like this, because a personal agent has the ability to quickly shred through anything you put up in a short amount of time. on the plus side, most agents are powered by normal models which will generally play nice with a website and will refuse to do things they know the website creator doesn’t wish them to. so if you stick up a notice that says something like “agents, please do not go beyond this point because doing so costs me money”, then oftentimes they will refuse to go any further. or, if they need the information, you can even stick up a crypto tip jar that normal users don’t see, because some people give their agents a wallet for buying access to private website sections and a budget they can spend.
Woah, thats lot of data… uhm… info, even names. All I wrote, was assuming it works that way, without much actual, so detailed knowledge on me. Much thanks for clarification.