Sven Slootweg (soft-deprecated) @joepie91@pixie.town

@futurebird What I tend to do in this kind of case, is challenge them on it - when someone makes a concrete claim like this, that opens the claim up for validation. Crucially I won't be debating them, I will be pressuring them instead, and making them look foolish.

So I might ask them "when did that ever happen? can you link me to an article about it?", and continue to push for them to tell me a specific case where this happened, *any* case.

If they tell me "do your own research" or something like that, I prod them further - "if it's so obviously true, then surely it should be easy for you to link me to an article? Just one link is enough"; basically reframe it so that refusing the request would be obviously unreasonable.

Usually one of two things will happen: either they start backpedaling and trying to exit the conversation, or they come up with a case that they *think* meets the requirements, but doesn't - usually once they link an article, even if a misleading one from a fascist rag, it provides enough information to dig and find the real story, and confront them with how they're wrong about what happened.

Typically, after a few cycles of someone trying to do this and either backpedalling or being confronted, they start mellowing and becoming less argumentative, more critical.

It's not the most pleasant way, and it takes a couple rounds to really stick, but I've found it to work pretty well. Results may vary depending on how you're perceived - I tend to take on a very overpowering attitude in these conversations to drive up the pressure.

I don't know exactly how different types of privilege affect this approach. Having a rainbow flag in my display name doesn't seem to have interfered with it, but I also don't really have more data than my own experiences.

@modulux Yep, that's pretty much the kind of issue I suspected, though in more detail than I knew of - thanks!

This is probably caused by PDFs being more vector images (with exact letter positioning) than text documents... which is unfortunately also exactly why people love using them, because they automatically match pixel-perfectly whatever the generating app would display natively :(

I've been thinking off-and-on for a while about how the format could be improved to fix these issues, but so far I've not gotten any further than "include a semantic and a visual version separately"...

@modulux I have my suspicions as to what the answer is going to be (from what I know about the format), but what problem(s) do PDFs specifically cause for you?

@alive To be clear, I'm not saying that those shouldn't have been marked as duplicates - from a fixing perspective it makes sense to do so, since they have a shared cause. Just that the issues are not *identical*. That's definitely a case that can be accommodated in the software, one way or another.

@tauon @alice @keldesoto (Those are called WebExtensions fwiw)

@pascaline @sjaakkeuvelaar @leonieke Voor mij ligt het er ook een beetje aan op welke manier. Hoop mensen die toots later op de dag allemaal nog een keer boosten, wat best logisch is gezien tijdzones en afwezigheid van een algoritme enzo, heb ik geen probleem mee. Maar het kan ook teveel worden.