Look, we can debate the proper and private way to do Captchas all day, but if we remove the existing implementation we will be plunged into a world of hurt.
I run tucson.social - a tiny instance with barely any users and I find myself really ticked off at other Admin’s abdication of duty when it comes to engaging with the developers.
For all the Fediverse discussion on this, where are the github issue comments? Where is our attempt to convince the devs in this.
No, seriously WHERE ARE THEY?
Oh, you think that just because an “Issue” exists to bring back Captchas is the best you can do?
NO it is not the best we can do, we need to be applying some pressure to the developers here and that requires EVERYONE to do their part.
The Devs can’t make Lemmy an awesome place for us if us admins refuse to meaningfully engage with the project and provide feedback on crucial things like this.
So are you an admin? If so, we need more comments here: https://github.com/LemmyNet/lemmy/issues/3200
We need to make it VERY clear that Captcha is required before v0.18’s release. Not after when we’ll all be scrambling…
EDIT: To be clear I’m talking to all instance admins, not just Beehaw’s.
UPDATE: Our voices were heard! https://github.com/LemmyNet/lemmy/issues/3200#issuecomment-1600505757
The important part was that this was a decision to re-implement the old (if imperfect) solution in time for the upcoming release. mCaptcha and better techs are indeed the better solution, but at least we won’t make ourselves more vulnerable at this critical juncture.
Captchas pretty much worthless. They’re easily bypassed for basically free. You’re better off putting your instance behind Cloudflare with their captcha
Okay, so do you mind explaining why the servers onboarding the most spam users are the ones without Captchas?
If they are so ineffective, why are they effective now?
Invisible captchas are about as useful as graphical ones and are significantly less annoying to the end user
Sure, so implement them in v.0.18 rather than leaving that essential feature for a future release - that’s all I personally want.
I don’t care about the technical implementation of the Captcha, but given the current threat landscape of low effort bot attacks, removing the feature in the meantime just makes the fediverse worse off.
I’m mixed about this. When applied correctly, a graphical captcha will let zero bots in, at the expense of a false positives and frustrated users. On the other hand, invisible / proof-of-work captchas will let a fraction of the bots in, while providing better experience for legitimate users. Pick your poison basically.
Absolutely untrue. There are services that will solve captchas for you for hundredths of a penny. It’s essentially free.
Then it’s no longer only a bot, right? There are real humans working on those captcha farm. Those captcha farm also won’t solve the captcha instantly, but there will be some delays for a human to solve the captcha. You’re effectively turning graphical captcha into proof-of-work captcha this way, which will have the same effect as mCaptcha due to increased cost (in this case, captcha farm cost instead of computational cost) for the bot operator.
To be fair, “Captcha” can now mean those ai photo discrimination tests. Captcha: “Select the cats” - Me: “You call these cats?” Looks at the cartoon depictions of nightmare fuel “cats” as depicted by Picasso.
There are still graphical tests we perform that are much harder for computers to perform - at least without near-nation-state sized financial backing.
Yes, the ol’ scrambled captcha has been solved by multiple approaches these days, but Its not nation states I’m seeking to keep out (and I’ll be fucked if they ever did, I might add), I’m just looking to make it harder for some internet edgelord’s low effort spam attempts.
Sure but you can pay a company in India a few bucks for a few hundred captcha solves. It doesn’t matter what the captcha is, because a human is actually solving them, you’re just outsourcing it for literal pennies. It’s not difficult, either
Look, you keep returning back to a point I’m not making, and it seems like its in bad faith.
You keep saying how captcha’s aren’t perfect. They never needed to be and any sufficiently advanced attacker can bypass them. We’ve gone over that at length, you returning to this argument just shows how little else you have than “Mondays always suck” / “Evil shall persist” mindset.
Your entire position of chasing me on “oh, but captcha doesn’t solve ALLLLLL bots”. Yeah, and laws don’t deter ALLL crime either.
Shall we remove these pesky laws of civil society? I mean, after all why abide by rules that any one person can chose not to follow the laws? What good are they anyways?
You know it’s an inane point that has no logical conclusion, but I think you probably already know that and I’m done assuming good faith in your trolling.
Seems to me like you’re mis-framing what’s being said to fit your argument and claim the other person as a troll. No one has made a claim about stopping “ALLLLLL” bots or “perfection”. It’s about whether it stops enough to matter. And I think it’s safe to assume if someone had the interest and capability to write a bot, they can probably google “how to defeat captcha” and implement one of them. If there’s currently not a flood of bot accounts, I believe it’s from a lack of caring rather than the captchas doing anything.
There are solutions for bots, they should be implemented, but keeping the existing captcha isn’t worth it. Multiple things can be true, but I get the feeling you’re set enough in your opinion that you’re going to (continue to) attack the character of anyone who disagrees.
Because this spam-bot seems to be currently only targeting these instances.
So what you’re saying is that a poorly constructed door is better than none at all? Huh. That was my exact point.
No I am saying the this bot seems to specifically look for instances without captcha and doesn’t even try others. Low hanging fruits all that. If all admins enable captchas the bot would just switch to those and circumvent the cheap captcha that is currently implemented in Lemmy.
So the solution is to force everyone to be low hanging fruit in the meantime?
Look, I get where everyone is going in terms of improvements, but to remove an already working solution and leaving folks exposed in the meantime is not how we should be rolling improvements.
See my other comment. Lemmy already implements other ways to prevent this from happening that are much more effective.
Email validation works only until my domain get’s blacklisted…
Manual registration only works up until a certain size…
What other effective solution shall I consider? Those aren’t very effective to me.
If you reach a size where manual approval doesn’t work anymore you should seriously consider closing registration completely or increase the size of your admin team.
Also, can we not advocate putting more of the internet behind Cloudflare? They’re building an analytics platform that, unlike Google and the like where you can somewhat avoid tracking by blocking certain network requests, stuff served from Cloudflare directly means tracking can not be avoided. To block the tracking, you have to block access to Cloudflare. And when a huge portion of the internet, especially the top sites, are behind Cloudflare, you’re just cut off from a lot of the internet. Also putting everything behind Cloudflare is centralization, which seems counter to the goals of federated services.
I’ll also quote myself from elsewhere:
Whether it’s Cloudflare, or google, a bespoke system, or some other third party, you are going to be tracked. It makes no difference. At least Cloudflare offers a supremely good service for essentially free. You cannot escape tracking, other than using browser extensions, so just use that.
I’ve not been hit by Cloudflare’s bot detection as someone using brave and other tracker blocks, so I’m not concerned
You’re misunderstanding what’s happening, and it seems you think they’re operating the same way. With google, you can block network requests with browser plugins, disable javascript entirely, dns-blackhole google domains, etc. At the end of the day, you can still access the content being served without sending tracking data to google. With cloudflare acting as the first-party and directly serving you content, you can not block anything. It would be like if everything came through google’s servers (which, if you’ll remember, is part of the complaint about google’s push for AMP).
Yeah, you definitely don’t understand how this works. The reason browser extensions work is because they can separate out specific resources and specific network requests by domain; if Cloudflare is the one serving you the content directly (as they do), you can’t do anything with a browser plugin.
Nice to know that since you haven’t been affected it’s not an issue. Very solid.
I’m an engineer. I know how it works. I know it also makes no difference.
The company I work for uses Google, Cloudflare, multiple third party advertisers, and our own internal analytics system. Every user is being tracked in some way, and it’s confirmed through our server hits. You genuinely cannot escape tracking and it is useless to try.
Even if you do block all JavaScript tracking, you’re still hitting someone’s server and that’s being tracked. You literally are incapable of blocking this form of tracking.
Depending on your configuration, Cloudflare doesn’t have access to page content, only the url you’ve hit. The rest of their tracking, including what you’ve explicitly mentioned, is done through JavaScript shims that you can absolutely block through browser plugins.
While I agree in the practical sense (I use CloudFlare myself), it kind of goes against the spirit of the fediverse as it centralises around a single corporation.
I don’t fully understand your argument. You’re using a centralized caching layer, sure, but the actual application that matters is still federated?