Just an update on this front, server load is stable right now as I type this after rimu did some stuff on the server to help. It's pretty clear this was a DDOS and not a rogue AI scraper because it was hitting the same url many times a second instead of crawling tons of urls like a scraper bot would.

We'll keep an eye on things. Thanks for your patience.

I don't think the privacy issues here are too salient. Pretty much everything on the fediverse is public already and have likely federated outside any particular region like the EU, so GDPR doesn't really have any teeth. The exception to that would be if instance admins are using database access to also feed private messages to an LLM (especially a corporate LLM). I know that the "private" in private messages on the fediverse can be conditional...but it should at least be considered private from LLMs as an expectation since those messages are inaccessible to things like scraper bots or listening instances designed just to harvest data.

My biggest concerns here would be twofold:

1. False positives - LLM sycophancy is a thing. So, I worry that if you ask an LLM to dig through a big pile of text looking for a thing, that it will tell you that it found that thing...even if it is completely removed from context or completely made up. The false positive rate might be low (I have no idea), but I guess I just don't trust the LLM enough to let it take the wheel with stuff like this.
2. Outsourcing moderation - LLMs are not going to be up to the task of moderating everything, just ask digg. However, tools to help moderators effectively do their jobs are helpful as well. There is a balance to be struck here. I think, for me, something like asking an AI essentially, should I ban this person, just feels like you are outsourcing your decision making too much. It is too far on the automation side of the scale for my tastes.

All that said, people can run their instances how they want. I don't really have strong opinions on LLMs/AI in general, I just kinda hate big tech companies. That is my foundational belief in the work that I do for the fediverse - fuck big tech and the oligarchy they have built/funded in my country. That is really the only axe I have to grind in all this.

Also fun fact, you can probably upload it to most other piefed instances just fine.

rimu has pretty strong opinions on social media. This filter is optional and can be turned on/off by an admin. Some of my contributions to piefed have been to make filters or features that are strongly opinionated like this optional. For piefed.social specifically though, rimu has all of them on because that is his instance and he runs it the way he wants.

I don't necessarily disagree. I haven't really taken a close look at how this is implemented, but it also hasn't really been a high priority to revisit, at least not for me. There are still plenty of more fundamental features to get right first in my opinion. The big one I have worked on for the next piefed version is to get local sticky posts working for example.

My experience from working with rimu though is that he has been pretty receptive with contributions to make it less opinionated in these kinds of ways. I have removed or made optional tons of stuff that he spent time coding and I haven't really gotten any pushback from him over it. I know it kind of makes me sound like a douche to just say open a PR, but if somebody out there feels strongly about this filter, that is probably the fastest way to get it changed.

the whole þ fiasco

That has since been removed. Yeah, rimu is certainly opinionated and passionate about what he believes in, but has also been pretty receptive to feedback, both from users and from admins (like in the private voting case). Fortunately, there are alternative threadiverse platforms out there for people that want them. Both lemmy and mbin do some stuff better than PieFed, and that's ok. The different projects have maintained working relationships at the dev level to try to make sure interoperability outside the base activitypub spec doesn't completely break (the post-moving feature/FEP was a collaboration between PieFed, lemmy, and NodeBB for example).

You brought this up in the matrix room, so I will try to put down here a summary of what I responded in there so that it lives in a more permanent fashion.

Lots of profile changes are not federated out in realtime, extra fields being one of them. Instead, your profile, as seen on remote instances, is only refreshed periodically. In the case of PieFed, when an instance receives an activity from a user (post/vote/comment), and their profile information hasn't been refreshed within the past 24 hours, then the instance will query your home instance to make sure it is up to date.

What this means is that if you update your extra fields right now, it can be up to 24 hours before that has propagated out to remote instances depending on when your profile was last refreshed on their instance. It could be even longer if you don't federate any activities out in that time.

A very similar process also happens for lemmy as well as for other "actors" in the ActivityPub parlance. This includes communities and feeds.

Something like that could be a good place to utilize "trusted instances." So, instead of just choosing a random instance of all of the piefediverse, it chooses a random trusted instance.

The feedback in this thread overall has been helpful.

Yeah, I think I see the issue in the jinja template. PieFed handles some video hosting sites a bit specially to make the compatibility a bit nicer, but the blur setting was not being passed on to those special cases. I think this is a pretty simple fix.

In the meantime, if you switch to either of the compact view settings, then the videos won't be loaded in the feed.

Yeah, there was recently a case where somebody was doxxed in the modlog. At that moment, there really wasn't anything that you can do about it. So, being able to have some control over modlog visibility is a way to prevent abuses like that.

Is there a software solution on the app developer level that combines like posts together?

As mentioned in this thread already, piefed consolidates all the comments for crossposts when it detects them. As an example, you can look at this post on piefed.social. The link I shared is for the post on ![email protected], but below it you can see comments from the same article posted in ![email protected] as well as ![email protected] in their own sections as you keep scrolling. So, problem solved, right? Well...

One of the key phrases I used above is "when it detects them". So, how does piefed detect crossposts? The answer is pretty simple, it basically just looks for other posts that point to the same destination url. In the example I linked, that would be the Guardian article that is being discussed. This is the same way that lemmy detects crossposts. This approach is nice and easy and computationally cheap on the database (quick), however, there is a big shortfall of this method...posts that don't point to a url (discussion posts) can never be detected as crossposts. Lemmy offers the ability to hit the crosspost button on a discussion post and it will create a big block quote of the original post for you, but it isn't actually recognized as a crosspost in the software.

I don't have a good technical solution to be able to make discussion posts (and other non-url posts, like piefed events or polls) be crossposted properly. It likely would need to be tracked in the database somehow, but it would rely on users somehow indicating that the post they are making is meant to be a crosspost. I don't know really...

Anyway, that is the current state of crossposts. Thanks for coming to my TED Talk.

This was removed. Instead, users have the option to simply not federate out their votes if they want.

A PieFed admin can disable not-logged-in users from viewing NSFW content. IIRC, this was added after the recent UK laws about NSFW content.

piefed.world is on an older version of piefed where that wasn't added yet. Similarly, instances that are on older versions of piefed won't have links to the modlog on a community's sidebar or on a user profile's dropdown.

I'm not sure what is holding up piefed.world from updating. They are still on the 1.1 branch when 1.2 has been stable for some time now and we are preparing 1.3 for release very soon.

Alright, after another user pointed it out to me, I suspect the reason you aren't seeing links to the modlog is likely because you were on piefed.world. They have been slow to update their piefed instance and are still on the 1.1 version. That means that updates to piefed that have been introduced in 1.2 (current stable version) and 1.3 (current dev version) are not present. Some updates to the modlog since then:

• (1.2) searching and filtering the modlog if you are logged in
• (1.3) introducing a dedicated link to the modlog on the footer of every page
• (1.3) adding a link to the modlog on each community's sidebar
• (1.3) adding a link to the modlog on a user's profile page

Going to the communities page and clicking the "Joined" button gives you a big list of the communities you are subscribed to and a convenient button to allow you to unsubscribe from each.

direct link

Yep! Absolutely. Here is the relevant portion of the docs.

I am using R2 for my instance and I set up those environment variables in my .env file before building the docker container for the first time, and everything just worked. The specifics of different S3 providers can vary a bit. There is a special set of instructions for using Wasabi S3 for example (I think this was contributed by Jerry of feddit.online).

Yes, actually! There are two ways you could force a user to get fetched that I can think of:

1. @-mention them in a post or comment and then click that link after the post/comment is made...if piefed doesn't already know that user, it will go fetch that user.
2. Just input the lookup url directly. This is what is happening behind the scenes in #1, but the url structure looks like this (looking up myself using piefed.zip as an example): https://piefed.zip/user/lookup/wjs018/piefed.wjs018.xyz

tl;dr - Very specific way that lemmy.world broke means that there is a gap in activities that isn't going to federate

From a brief discussion on matrix with MrKaplan, it seems like the specific way in which lemmy broke for them (only really possible at their scale), the activities that happened during that period were marked as successfully sent even though they weren't. Going back and resending all activities from that point on could do really weird things and end up duplicating tons of posts/comments all across the fediverse, so they have opted to not do that. So, we will just have to live without those posts/comments in the wider fediverse unless you manually force federation for them.

Interstellar has a standalone desktop app and supports all three threadiverse ecosystems (lemmy/piefed/mbin).

I know the blocking stuff is actively being worked on as we aren't happy with it yet. However, that is an issue where it would be literally impossible for every user to be happy with whatever end state it results in. That is an area where the different software platforms can offer people different experiences so that they can seek out the type of experience they are looking for.

Almost everything else you listed here can be turned on/off by an admin. I didn't know about the 8-character username thing, probably worth revisiting that. The downvotes being disabled with low attitude is one I don't see being removed, but could be an admin-set threshold (or disabled) in a future version. In fact, I suspect that the formula was written this way to make the threshold more well bounded instead of a simple up - down calculation.