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.
I think it would be interesting if we could find a prompt that doesn’t work well with LLMs. Originally they struggled with math for example, but I wonder if it’d be possible to make a math problem that’s simple enough for most humans to solve but which trips up LLMs into outputting garbage.
I personally use this to track who send my email address to where, since people usually don’t strip this from the address. It’s definitely abusable, but also has legitimate uses.
Not so sure on the LLM front, GPT4+Wolfram+Bing plugins seems to be a doozy of a combo. If anything there should be perhaps a couple interactable elements on the screen that need to be interacted with in a dynamic order that’s newly generated for each signup. Like perhaps “Select the bubble closest to the bottom of the page before clicking submit” on one signup and “Check the box that’s the furthest to the right before clicking submit”?
Just spitballin it there.
As for the category on email address - certainly not suggesting they remove supporting it, buuuuutttt if we’re all about making sure 1 user = 1 email address, then perhaps we should make the duplication check a bit more robust to account for these types of emails. After all [email protected] is the same as [email protected] but the validation doesn’t see that. Maybe it should?
I like your idea of interaction-based authentication. Extra care would need to go into making sure it’s accessible, but otherwise I think that would be a stronger challenge for LLMs to solve. (Keep in mind LLMs can still receive the page’s HTML as context, but that seems like it could present as a stronger challenge even still.)
This makes sense to me. I could be wrong, but the assumption of 1 email = 1 user doesn’t seem unreasonable, especially since there’s no cost to making a new email address.
Check the screenshots I attached right above here.
The emails, were all unique. Sounded, like things I would expect from actual users too.
When it comes to LLMs we could use questions which they refuse to answer.
Obviously ‘How to build a pipe bomb’ is out of the question, but something like ‘What’s your favorite weapon of mass destruction?’, or ‘If you’d need to hide a body, how would you do it?’ might be viable