wjs018

From Sal on matrix:

From what I can find, the managing director of the company that runs serverspace.io (ITGLOBAL.COM NL), which is the company that rents me the VPS space, was arrested by Dutch authorities. The infrastructure in Amsterdam was disconnected without warning when this happened. Since my latest database backup is unfortunately not very recent, and because there is still hope that the severs will come back up, I am waiting and hoping that I regain access to the data. They originally published some somewhat optimistic message, but as of today their website is inaccessible.

https://www.comnews.ru/content/247454/2026-09-21/2026-w39/1007/globus-dlya-itglobal-umenshilsya

https://theins.ru/news/297340

I could not find any statements from Dutch police or really any source other than IT Global themselves. But, this event may be related to “Dmitry Gachko” (owner of ITGlobalcom Rus) being added to this sanctions list: https://eur-lex.europa.eu/eli/dec/2024/1484/2026-07-13

I recieved no warning, and I did not know about the ownership structure of serverspace and did not consider that the infrastructure was exposed to this risk. I figured “VPS in Amsterdam from a Dutch company” was rather safe. But “Dutch company operating in NL” does not mean the same as Dutch owned…

FYI, next version of PieFed has an “Unban All” button for clearing all domain bans to try to make this easier.

Screenshot from my instance:

PieFed has a default list of domains/instances that are blocked when you do a fresh install for a brand new instance. A critical difference between PieFed and the Tesseract blocks is that any of the blocks in PieFed are completely visible to, and under the control of the admins of the instance. So, ultimately, it is just an out-of-the-box default, but the instance admins always have ultimate control over moderation/federation choices. This Tesseract stuff is taking those powers away from the admins.

there is a work-in-progress page for Piefed that a contributor is working on

Yeah…sorry for the delay on this one. I have been absolutely slammed with irl work and haven’t had time to do much of anything recently. This is also why my actual contributions to PieFed have dried up the past couple weeks. I swear I haven’t forgotten about this, and I think the new structure will make this contribution a lot easier as I don’t have to replicate a lot of what existed for lemmy already.

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.

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.

I’m sure you will be pleased to know that the filter this post is about and most of the others mentioned in this thread are, in fact, off by default.

https://codeberg.org/rimu/pyfedi/src/commit/b168820a089ff6e835059f0d806f81b612987a79/app/models.py#L3513

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.