Not sure what a thinly resourced piece like this is doing so high up on HN. There’s really very little evidence that this has anything to do with racial slurs. From a quick read: last June, the channel was banned for “harmful and dangerous” content. The ban was removed in under 24 hours. YouTube has not confirmed the reason. And yet: > Experts suspect that it was the usage of words like "black" and "white" that confu…
It’s so high because it fits general HN narrative of evil big tech.
YouTube blocked chess channel after mistaken detection of racial slurs
31–40 of 584 posts
Re: YouTube blocked chess channel after mistaken detection of racial slurs
#32“ The researchers now suggest that social-media platforms should incorporate chess language in their algorithms to avoid future incidents like this.” Or maybe don’t automate banning YouTube channels via nebulous black boxes? Or can advertisers stop being afraid of anything and everything?
> nebulous black boxes Probably a nitpick, I think you mean "opaque black boxes", you can see things through a nebula.
Re: YouTube blocked chess channel after mistaken detection of racial slurs
#33Not sure what a thinly resourced piece like this is doing so high up on HN. There’s really very little evidence that this has anything to do with racial slurs. From a quick read: last June, the channel was banned for “harmful and dangerous” content. The ban was removed in under 24 hours. YouTube has not confirmed the reason. And yet: > Experts suspect that it was the usage of words like "black" and "white" that confu…
Re: YouTube blocked chess channel after mistaken detection of racial slurs
#34Earlier quoted context omitted.
The block was for “harmful and dangerous” content. That’s incredibly vague. Even describing a game using aggressive language could fall foul of such a broad filter. How many times do non-chess channels get caught in this filter? That’s the first question I’d be asking. I have to assume more than zero times. How many chess channels have not been banned by this filter? Again, presumably more than zero. And yet the arti…
Have you watched adagmator? He doesn’t really use aggressive language
Re: YouTube blocked chess channel after mistaken detection of racial slurs
#35Re: YouTube blocked chess channel after mistaken detection of racial slurs
#36Delegating prosecution of possible hate speech to commercial entities was always going to be bad. Youtube is essentially a public forum with audience of entire countries, where the people who run it choose what gets amplified, what gets shunned and what get outright blocked. Who would be happy with a private entity deciding the future mindsets at this scale?
Re: YouTube blocked chess channel after mistaken detection of racial slurs
#37inb4 someone will suggest that FIDE should change the color names to be more inclusive + why does white have the first move advantage, why can't black be the first. Perfect breeding soil for just another day of pointless race drama on Twitter
But given that the only patch to chess in the last century was to prevent playing 0-0-0-0-0-0 with a promoted rook, I doubt I'll live to see that happen.
Re: YouTube blocked chess channel after mistaken detection of racial slurs
#38Their either needs to be competition for Youtube or they need to face regulation. You can't have it both ways. The system does not work without one or the other.
Re: YouTube blocked chess channel after mistaken detection of racial slurs
#39Not sure what a thinly resourced piece like this is doing so high up on HN. There’s really very little evidence that this has anything to do with racial slurs. From a quick read: last June, the channel was banned for “harmful and dangerous” content. The ban was removed in under 24 hours. YouTube has not confirmed the reason. And yet: > Experts suspect that it was the usage of words like "black" and "white" that confu…
And what do you do to force them to confirm the reason? That's right, get it to be so high up on HN.
Re: YouTube blocked chess channel after mistaken detection of racial slurs
#40Earlier quoted context omitted.
Do we really need to split hairs like this? The algorithms aren't completely opaque, as they have been trained on a set of representative inputs, and are continuously monitored. They are based on neural networks, presumably, with weights inspectable by the engineers deploying the model. So it is more "nebulous" and not completely opaque.
We do, because in terms of writing english black boxes are opaque by their very nature. OP could remove "black boxes" to make the sentence correct. You can't actually see inside a black box. That's why it's a "black box". Sure you can probe it, but that's not the same as being "nebulous", a thing you can actually see through sometimes. I expect better of HN.