Yeah we'd been discussing this but nobody had a chance to make this thread about it yet. It feels dishonest for the app to have these hidden biases, so I'd rather just kill it.
I've pulled tesseract down now.
Yeah we'd been discussing this but nobody had a chance to make this thread about it yet. It feels dishonest for the app to have these hidden biases, so I'd rather just kill it.
I've pulled tesseract down now.
The highest fucking honor I’ve ever received: being blocked by a wimpy little fascist:

Aw, I didn't make it. I'm not trying hard enough.
Good lord, I've tried to read the post, but Dev is such an edge lord I'm beginning to suspect Elmo himself vibe coded it.
At least I'm part of the list, being filtered out by this asshole is an honor.
Oh shit, I am on the block list. 😂
I didn't even know what Tesseract was before this, and I'm on the blacklist.
I didn't even know what Tesseract was
I still don't know.
My searching tells me Tesseract is open source OCR software, but that doesn't make sense in the context I'm seeing in these threads.
This page might help, it explains what alternative UIs are and which ones are available. Some of them can also be used in the mobile web browser.
https://fedecan.ca/en/guide/threadiverse/alternative-uis
(It says tesseract is available on lemmy.ca, but that is out of date now)
This issue with Tesseract affects anyone that was using it. So it is possible that someone did not see a comment or post you made, without knowing why they could not see it. Past that, it likely hasn't affected you since you said you don't use it :)
Thanks so much for sharing this, I couldn't find an actual screenshot of Tesseract anywhere what with Dubvee redirecting everything to a "goaway" picture.
It sounds like I might not have known if I was using it or not, depending on how I accessed Lemmy.
It's extremely unlikely that you were using it without knowing. Though if you were you'd know now because reloading the page would load Lemmy-UI instead of Tesseract. They don't look at all alike.
edit:
https://photon.lemmy.ca/ looks very similar to tesseract, if you want to see the difference.
It's a semi-popular Lemmy web frontend that a lot of Lemmy sites use instead of Lemmy itself. It seems like the original repo by asimons04 has been taken down, but here's a random fork of it for example.
So if i use mobile exclusively via voyager app. Does that mean im not implicated in this mess?
I have no idea what any of this means, technically. I understand and agree that some shadow ban black list blocking users is bad, and i dont want that, but i dont even know what people are being blocked from
I have no objection to stuff like the nsfw blocklist by default, I can even think of justification for a lot of filters. But the way it was done is shady and weird.
This whole thing is crazy for current users. I'd been trying to figure out why Tesseract was behaving so strangely when I hadn't changed anything. Digging through settings and messing around with filters. There was no notice at all. And this only makes sense if you only use Tesseract. I use Voyager on my phone and I was receiving notifications there that weren't appearing in my browser. I'd post comments from my phone and not be able to see them in Tesseract because I guess someone else in the comment chain was on this list. Usability nightmare.
And then you can't disable it without having a big ever-present content warning telling you you're in Toxic Mode. Yeesh. You know who else is a level-headed guy who's been wading through the sewage for 3(+) years? THIS GUY. I know my way around this glorious sewer, my dude.
Toxic mode does not disable the secret blacklist.
Yea I think some of this would work better as an automod or reporting tool.
I've played with Reddit's automod, and even well designed automod regex tends to make lots of mistakes, and you need to slowly make adjustments to refine it.
Other parts of it should have been opt in, and ideally in a way that you can import any list. That would also help with the gaming issue, since a bad actor can't look at one list and game it when people can be using any combination of blocklists. Similar to how ublock origin does it.
Hell yeah, made the list. You know us Canadians and how abrasive we can be.
Probably a good idea. Too much shady shit.
E: Am apparently on the blocklist! 🥳
To disable them (referred to as enabling Toxic Mode in the settings), it's as simple as:
Settings -> Advanced -> Click the "I feel like some content is missing. What's up?" text at the bottom -> Enable Toxic Mode
Hahaha
Line 2459, I'm on the list baby!
If you've got a blacklist I want to be on it
Waiting for the Great Leap Forwards
It's nice, and I know the https://lemdro.id/ instance has it as the default
However since it is a third party frontend, there is a chance that a future Lemmy update will come out without having full Photon support right away. So it's usually safer to keep them separate, and encourage people to try out the frontends to see what they like
Photon's Lemmy V1 compatibility is actually pretty much complete with a lot of technical improvements as well, I'm kinda hoping the Lemmy devs release Lemmy v1 soon as I don't really want to make a lemmy v0.x adapter
I agree with what you're saying, and some of this is also against our drafted new guidelines. While people have suggestions on the wording and implementation, those guideline proposals have been received positively
However this implementation doesn't really solve the problem effectively.
your comment would also have been hidden, along with any comment that wants to discuss the issue or make the same point you are making (or a completely unrelated topic, see Scunthorpe problem)
mods and admins are less likely to notice the obviously bad actors. So instead of those accounts getting banned and purged, they might still be posting harmful content or coded language that the filters don't pick up on and can't keep up with
Some words SHOULD get blanket banned, but we already have the admin level keyword blocklist for that.
Otherwise I think that a frontend is the wrong place for this functionality, and It should be handled by a proper automod or reporting tool/plugin
lemmy.ml also loves overreaching:

As an Australian I find this censorship offensive. I can't express myself without that word, it's like tying an Italian's hands!
Well that's a funny direct example
We did also adjust the default filters for our Canadian instances, to account for words that are harmless in French but harmful in English. For example: https://www.collinsdictionary.com/dictionary/english-french/late
I have no tolerance for violence and hate. None. If you want to see and/or spout that shit, again, use another client. If you’re as sick of seeing that shit everywhere as I am, well, maybe give this a try because that’s the vibe I’m going for.
I agree with this in principle. But I have no doubt I ended up on the blacklist for calling out his own daily violence against vulnerable individuals. I've never advocated for a violent solution to the many problems we discuss.
He also ... said a lot of stuff. Wow. The further I get into that post, the less meaningful it seem. Buddy's falling apart, and externalizes the blame on us.
I read the dev post. Oh boy, does this person has got issues. Not to imply I or anyone has no issues but some of those decisions.. 😔
Lots of bad assumptions, conclusions and misunderstandings too. Like this whole thing abt people banned from reddit being the bad ones on Lemmy. Buddy, my account on reddit is alive and well with fairly high karma. I left both in protest as well as a conscious decision to contribute whatever time (and actually more) to a non-corporate social media. I am on the blocklist. Bravo dev. 👏
top 58 comments