there are quite a few ways to use more granular targeting. for example, we have specific url patterns that get challenged if certain headers have certain values and are missing others, while other urls won’t get challenged.

it would be trivial for them to not DDoS and yet they keep doing it.

if they were calling Lemmy APIs directly they would significantly reduce the load they bring to our service. if they were speaking ActivityPub they could even get the content delivered to their front door directly via federation.

they don’t care that they DDoS websites. they don’t care about optimizing for how certain types of websites are built, to reduce impact on third parties. the only language they speak is DDoS.

they intentionally spoof legitimate browser user agents and cycle through massive ranges of IP addresses. they have enormous pools of IPv4 addresses available that allow them to only use each IP for a couple of requests before cycling to the next one, which is yet another way they evade detection, as they are bypassing any rate limits we have configured that way.

we’re currently challenging visitors who aren’t logged in regardless of the browser due to extremely aggressive LLM scrapers. as long as you’re staying logged in you won’t get challenged.

Cloudflare actually stopped their service

what?

we are on the latest version now, there is no need to wait

doesn’t really look much different between the two instances. the theme differences are minor. the current default theme on piefed.world is even a bit more compact than the quokk.au default theme.

there are also user settings for more compact post listings, e.g. the variant with smaller thumbnails:

and after writing all this i realized you were referring to the bottom navigation. we updated several piefed versions in the last few days, if this is still the case for you i recommend raising this as a bug directly at https://codeberg.org/rimu/pyfedi/issues

in the past couple of days it seems like LLM crawlers have been ramping up again, with totally normal traffic patterns like these

we’re currently looking into trying to identify them better without having to serve everyone with a captcha or something like anubis to access the site.

maybe worth to name and shame those apps with lack of proper lemmy markdown support.

and ideally check if that has already been reported to their devs, and doing that if it hasn’t been done yet.

fwiw, LW/PFW require local users to be at least 18, or, if higher, whatever their local laws define. for other instances this is of course nothing we can control.

all content is generally owned by its creator and only the creator can change it. for lemmy 1.0 this will be changing for marking as nsfw and managing tags, so it seems that this is somewhat being relaxed.

at this point they’re long past the point where they’d be tolerated here to keep a single account. they’re constantly impersonating and harassing other people, they need the new accounts to be able to evade bans.

it just takes one mentally ill person to make it their hobby to harass people on the internet

only the person creating a post can mark their own account as bot or change the title

feel free to get in touch with us if you have issues with api limits. we can’t change them for individual users but we may be able to give you advice on more efficient usage.

your bot shouldn’t be logging in every time it checks what is going on; it’s best to persist the JWT you get from logging in and keep using it. the signup/login rate limit (which for whatever reason has the same counter) is relatively low to limit abuse, but the other endpoints should be more relaxed. you’d also have to hit pretty high request counts for us to even look at it. i recommend using a separate bot account (see bot rules).

I changed the password after leaking that screenshot

presumably in the time between leaking and changing it? once the password is changed the old password won’t work anymore, and neither will any of the previous login tokens.

there is currently nothing I can see that would be a strong bot indicator. there are various indicators of an actual human using the account.

of all the scrapers we see, the requests identified as originating from Meta seem to be well behaved overall. they appear to (mostly) be respecting robots.txt where present and their request volume to Lemmy.World is only averaging slightly above 5 requests per minute over the last 2 weeks. they also don’t spoof their user agents to pretend to be web browsers, or at least I have not seen credible accusations of this happening.

this is a bug in lemmy-ui. it works in alternative interfaces, e.g. at https://t.lemmy.world/c/إسلام.

the bug report about this was just recently closed due to a lack of external contributions to address the issue: https://github.com/LemmyNet/lemmy-ui/issues/2207

hello,

based on your other post, which has since been removed by moderators of that community, it seems that you are looking for the !perchance@lemmy.world community instead of this one.

you’re not going to have much luck here, as here people won’t even understand what you’re talking about.

please remove the phishing link from this post immediately. you can defuse it e.g. by putting it in a code snippet and replacing the dots with [.], e.g. netprocesse[.]com.