I don't really know where to post this but this community felt relevant, apologies if this is not welcome here, feel free to moderate, ping, and redirect me if I should share it elsewhere (especially because I'm not a good fact checker, but I felt like it would be interesting to have a level headed discussion about this drama).
I originally posted to !tesseract@dubvee.org, but a moderator there banned me and deleted the post, I'm guesing the Tesseract dev feels harassed by the whole situation, and do not want to hear about it any more. The community is now locked.
My original post read as follows:
"Was curious about The List™ and took a look at what it contained. There's interesting stuff that tells us more about the intentions of whoever made it.
I'm not making this list to pile one the current drama, I don't plan on being added to The List™. I just thought that, when something like this arises, one thing the Tesseract dev could have done is explain why some individual keywords are included, this would have been more transparent. Because when you look closer, there's some absurd stuff in there, which can lead to confusion as to why some are filtered out.
I can't be bothered to go through content posted by each individual user and community that are filtered out, so I skipped those.
Community names
- ~25+ occurrences words referring almost exclusively to leftist political activism including but not limited to criticising american supremacy, the israeli-led genocide, racism, white supremacy, capitalism, paedophilia...
- 4 occurences of anime related content
- 3 occurences of right wing related content (Libertarianism, Conservativism, and a conspiracy theory)
- usgreenlandwar
- "Collapse" (discusses the environmental (and more) collapse of our civilization)
- any word referring to the green party of UK (their main platforms: anticapitalism, ecology, progressivism)
- "death to" (understandable, but depends on context)
- 4chan (and similar imageboard names) + greentext formats
- "billionnaire"
- "anonymous"
- 1 specific username and variants of their spam accounts / communities
- "Police" or "Cops"
Keywords
- ~30+ occurences of usernames, website urls, company names, sentences, to filter out spam / scam
- ~30+ occurrences of various sentences almost exclusively used by leftist circles, including but not limited to criticising american supremacy, the israeli-led genocide, racism, white supremacy, capitalism, paedophilia...
- 🥺?👉👈
- "What would (france|the french) do" (probably to prevent referring to the guillotine or the french revolution, still funny)
- "chatgip+ity+" used when someone criticises OpenAI's LLM
- "censoring me" (maybe because conspiracy theorists often claim to be censored)
- 4 Occurrences of sentences mentioning when an article is paywalled
- "pedophile" (can't talk about it?)
- "rap(e|ing)s? (kids?|child)" (can't describe the actions of the above?)
- "Perchance" (random LLM related website)
- 1 dead instance link, which was used by a computer science student that hasn't posted in 3 months
- "neoliberal" (can't even name a specific political and economic ideology?)
- "death to" (that one is understandable)
- "pray for nuke" (same as above)
- "remove from reality" (same as above but euphemism)
- "Kirk(e|'|’)d" (used to refer to the murder of Charlie Kirk)
- whenever any post is crossposted from a list of already filtered out instances
- whenever any user from a list of already filtered out instances posts or comments
Website URLs
I started this whole summary ~2,5 hours ago based on my search history. There about 300+ direct url entries and I can't be bothered to check every single one of them (I've been looking up their domain name on google, then directly entered their url in a separate browser, and searched the domain on lemmy). I went through about half of them before thinking I had better things to do, here's the pattern(s) I noticed so far:
- 4chan
- Any Blogspot blogs
- Any website hosted on Infinityfree
- Lemmy instances the Tesseract dev considers to be extremist (some are part of the free anarchist flotilla (btw quick fun fact, FAF is short for fascist in french, it's ironic for us french lemmy users))
- Any gumroad page
- Any itchio page
- Bunch of url shorteners
- Bunch of piracy related sites
- Bunch of cryptocurrency related sites
- Some are obvious scam / spam / typo squatting sites NOT shared on lemmy
- Left-wing related websites outweight Right-wing related websites 4 to 1
- A lot of links are just random companies over the world without them ever being shared on lemmy, or being scams, or spamming the web about their services
-
- there's a heavy bias against indian companies
-
- very few of them were probably filtered out because of their company name, for example a printing company from chicago has "freaks" in their name
-
- There's a french rodent control company that was filtered out, this one is interesting because its domain name is also the initials of the main left-wing french politician, but it could also have been filtered out because it obviously uses negative keywords related to their field. I feel like this perfectly highlights the flaws of this lists logic.
-
-
- edit: that one was actually shared twice on lemmy by a user asking relevant questions, my guess is the dev added them preemptively to prevent spam
-
- Some are links to website shared on lemmy, of those:
-
- a lot were obvious scams / spams
-
- some were ai related websites
-
- few were just devs asking ONCE for feedback on their project
Anyway if anyone is motivated enough to take a deeper look and share their findings, or correct me if I missed/mislabeled something, feel free to tag me!"
My take on it is they wanted to tone down Lemmy, likely to appeal to a broader audience. Not saying it’s right, just my impression.
Guessing they were filtering political content as it can wear people down being hit with it constantly.
The reason I’d guess there’s more leftist keywords on the filter list is probably due to Lemmy just being more left. You don’t see a lot of right type content or communities here (outside of very specific instances usually already defederated).
Lemmy can also get pretty extreme at times which can be off putting to really anyone. Things like calls for violence and the like. Noticed he had “death to x” filtered, “guillotine” etc. so probably trying to filter out those type of extreme comments.
Not defending their imposition of their filter list on unsuspecting users, just my attempt to make sense of their reasoning behind it. Like, they could have offered this as an optional setting to people with what’s on it listed upfront. People already create their own block lists and filter lists, so having done the work for them, why not offer it as an optional feature? Lots of users hate seeing political talk around election time. Boom, already an option in the client, turn it on and off at will. It’s like having ad block list options.
I’m sure the dev is over it all with the slew of “I’m on the list”, “Why aren’t I on the list”, “Can I be on the list” type of posts beating down on them as they’re banning people for that. Gives the poster YPTB content tho, so win-win? But yeah they sorta brought the ridicule on themselves with this. Can’t be blamed when you break people’s trust and ruin your credibility like that.
Sure but, if that really is the case the dev could just go ahead and... say... that? It could have been a transparent declaration or they could have made this at least "opt out". But instead it was done in the back, not documented or justified in any way.
That’s what I’m saying. Dude 100% could have had a completely different outcome had he been upfront, made it opt-in, and been transparent. He could have had a great feature of filter lists, instead he hid it and forced it on people without their consent.
Would have had a wildly different outcome.
Still would have pissed some people off who felt targeted by the filters, but again, opt in.
I think what's problematic is some of the bans seem to target the LGBTQ community (blahaj and wholesomeyuri being notable examples), as well as smaller news comms.
Another bit is how broad and sporadic the user ban list is. Sure some of the users appear to be of the abrasive type (like me for example, I get the feeling I could be nicer at times). But a lot of the names on it simply aren't problematic and have clean comment histories, empty of any political comments even.
I'm fairly certain Admiral Patrick was running a custom fork of his front end with a "ban from tesseract" button he'd use whenever someone seemed to mildly go against his world views.
It was slipped into his front end as well in a way that qualifies as malicious imo. He is an admin of his own instance, he could have simply blocked everyone at an instance level with the normal Lemmy modding tools. This was a way to circumvent the modlog so it wouldn't show and it was hidden inside a front end solution he was actively trying to distribute. It's incredibly shady. I left Reddit in part because of the shadow banning and hiding content, and this guy is trying to recreate it and spread it like some kind malware.
It feels like it goes against every principal the federation system stands for. I trust my instance admins to make the banning decisions, but he wants to add an extra hidden layer without informing anyone, hide it and then distribute it? It's highly unethical and frankly arrogant.
Well you see, people being different is political. Will drive normal people(TM) away.
He is also hiding comments with word lists. If you have "Zionist" or "Epstein class" in your comment it won't show. A lot of the user bans use regex so at least one other user caught a ban because they have "grimy" in my case (asgrimyr). It's quite insane tbh.
Yeah the user bans and the like, I got nothing. Your explanation would make sense. Idk how you end up with that many users on a filter list.
I did get a chuckle seeing beep on there after all the drama
There's so realm where any of this is ok. There's no reasonable thought behind mass banning anyone you might disagree with, or to make that choice for the rest of us. He didn't "ban politics", he banned left wing politics.
I agree, I also would like sometimes to not check Lemmy/mastodon and immediately get angry, but dev could've worked in a very different way to let people who don't want too much of that content. For example if a user wanted none of cat videos and pics, a simple customizable exclusion filter would take care of that, using a list of that sort (and hidden on top of that) makes me think the aim was another, and more personal.