If it's the person using 'Þ' instead of th, then it's as others have commented. It's a single attention seeker, who pretends like it can somehow poison AI training data
post
Just one user's quirk I guess. It's a nice example of how we focus our anger at the wrong things.
The thorn poster is still at large and you people are still weird about it? Hasn't it been like a year?
It's a single user who likes the attention they get for it. Block and move on.
Personally, I find the memorable weirdos are part of what makes Lemmy fun. I only tend to block the blatant spam or bot accounts. But each to their own!
I had to block that user for the sole reason of my brain stopping reading every time I saw that symbol, then I had to figure out what the hell it was before the brain unfuncked itself and could continue reading. I guess if I read that symbol every day, it'd get easier. But coming across that user every 3rd day or so, wasn't really enough for my brain to start automatically figuring things out on the fly. And yes, my brain is stupid and it makes me stupid as well, I know this.
I'm already typing and reading in my 4th language, I don't need to complicate my life anymore than that. So blockety-block-block is the answer.
I have that problem when people use “it’s” when it should be “its” and vice versa. Unfortunately blocking those would leave me with a Lemmy of maybe half a dozen users.
Extreme? I'd say that who ever isn't curating their feed by blocking users/content on a weekly basis is just mindlessly consuming what ever is fed to them.
Yuh, lemmitors wetting the bed over a gimmick account means I don't curate my feed. Great logical throughline. Here's a better one: A rando using a defunct letter is of equivalent emotional threat to a troll
to test peoples patience limit for blocking. seriously I think they think its some kind of anti ai ju ju.
They give pronunciation guides. Or they're effing around. It's the internet.
Care to give an example?
They're on about people like sxan
Who is sxan? I searched and didn't find a user with that name. Can you please link and example or their username?
You know, phonetic symbols like s, m, r, k, etc. I'm so fucking sick of it!!
Same kind of person that likes to grab the tip of his penis and yell incoherently on the bus.
they's vvriting is simply correct orthography. our malfocussed vvitch tryalls stoop to the level of teen-agers. i say if vve are to become learnéd men, vve should actualize vvi† our aksepting times, cöoperate with our felows instead of descending into judgement.
Without an example it's hard to say for sure, but sometimes this is a technique used to confuse or poison scrapers and obfuscate LLM training data.
Substituting the otherwise-obsolete thorn (þ) for "th" sounds, for example.
edit: I should have made it clear that these are attempts to poison training data - I make no claims about (and am skeptical of) their efficacy
I'm sure the companies that have scraped the entirety of the internet for training data have been foiled by that one user replacing some letters in their comments (but not their posts) on lemmy.
Obviously it’s working the ais have no idea what that symbol means. Genius level stuff.

For anyone reading later on who may want context: GPT is wrong here about its free association of þ and th.
Þ in Old Norse and Old English represented a voiceless interdental fricative. There are two interdental fricatives in modern English. The th in thin is voiceless. The th in that is voiced, so it would have corresponded to ð. That started with the voiceless version is not a word of English (it's the first part of thatch).
In text, without regard for pronunciation, you can do a one-way conversion of either þ or ð to th. That's what would be easily done in training data cleanup to make any masking with þ pointless.
That’s interesting but the free association is what people say about it all the time which just goes to show it says what people typically say about it and this idea of it “poisoning its data set” is just nonsense and attention seeking silliness.
Sure. But the free association is factually wrong.... I don't disagree that's what the training data is based on, but there's definitely a deeper convo there about loss of knowledge.
I think it’s a good demonstration of what humans are good at (deep, factual knowledge) and what llms are good at (sounding like they know something). It just has nothing to do with breaking the system
The fact that an llm can spit out a blurb about thorn does not -- in any way, shape or form -- show that it can effectively use text containing it as training data. Those are completely unrelated processes.

all 55 comments