We're too busy taking car companies' lines that it's reasonable they don't want people going to a local mechanic because "The data your car collected on you might get hacked and sent to China!" rather than asking "What are cars doing collecting this data, if it's such a risk?"
This is really cool, and I could see myself using this. Sometimes I need functionality like this, but can't be bothered to build up the infrastructure around it. This is perfect for that use case.
I imagine that would be one positive effect of endless entertainment being ubiquitous- since people can zone out in front of a screen with any entertainment medium of their choosing, they are less likely to opt for alcohol.
Though interestingly, drunk driving fatalities are rising despite lowering alcohol consumption[1], even discounting the pandemic contriting large jumps up. So of the people that do drink, they're more likely to drive after doing so? It seems like an interesting topic to study.
Anecdotally the drinking culture that existed for me in my youth has completely disappeared. House parties, a staple of my college years, are nonexistent. Talking to recent grads from the same school, their social life looks very different.
I think this is connected to social media but in a different way. Young people are very aware that any deviant behavior will be recorded and posted. Also they are deathly afraid of being "cringe."
Yes!! I was thinking of doing that myself but I don't have the chops to create an android app and no time to learn right now. But the protocol can't be that hard.
Based on another comment parent made, the Xbox was there before the kid, so presumably they don't want to handicap their own experience using it, just limit their kid's. Which should, by all respects, be a reasonable thing to want.
Personally, I have no issue watching things that are shot at 60fps (like YouTube videos, even live action) but the motion smoothing on TV shows makes it look off to me.
I dunno if it's just a me thing, but I wonder if a subconscious part of my brain is pegging the motion smoothed content as unnatural movement and dislikes it as a result.
The motion smoother also has to guess which parts of the picture to modify. Is the quarterback throwing the ball the important part? The team on the sidelines? The people in the stands? The camera on wires zooming around over the field to get bird’s eye views? When it guesses wrong and enhances the wrong thing, it looks weird.
Also imagine the hand of a clock rotating at 5 minutes’ worth of angle per frame, and 1 frame per second. If you watched that series of pictures, your brain might still fill in that the hand is moving in a circle every 12 seconds.
Now imagine smoothing synthesizing an extra 59 frames per second. If it’s only consider the change between 2 frames, it might show a bright spot moving in a straight line between the 12 and 1 position, then 1 and 2, and so on. Instead of a circle, the circle of the hand would be tracing a dodecagon. That’s fine, but it’s not how your brain knows clocks are supposed to move.
Motion smoothing tries to do its best to generate extra detail that doesn’t exist and we’re a long way from the tech existing for a TV to be able to do that well in realtime. Until then, it’s going to be weird and unnatural.
Film shot at 60FPS? Sure. Shot at 24 and slopped up to 60? Nah, I’ll pass.
Personal guess based on the impression I get from my parents' TV: You know how when you pause video while something is moving quickly, that object is blurred in the frame? Motion smoothing has that to work with, and causes the blur to persist longer than it should, which is why it looks bizarre - you're seeing motion blurs for larger movements than what's actually happening. Like the object should have moved twice the distance for the amount of blur, but it didn't. Something recorded and replayed at a high framerate wouldn't have this problem.
If you publish to GitHub, also mind that you grant them a separate license to your code[1] which grants them the ability to do things, including "[...] the right to do things like copy it to our database and make backups; show it to you and other users; parse it into a search index or otherwise analyze it on our servers [...]"
They don't mention training Copilot explicitly, they might throw training under "analyzing [code]" on their servers. And the Copilot FAQ calls out they do train on public repos specifically.[2]
So your license would likely be superceded by GitHub's license. (I am not a lawyer)
Maybe; I'm not even going to bother parsing all that tonight.
OTOH, if I create software and publish it on gitlab, and I'm not a github user, and someone else copies it to github, that doesn't scrub my license off or give github any rights at all to my software, no matter what their agreement with whoever uploaded the software was.
Another thing to add to the uBlock filters alongside the cookie banners. Can't wait to wade through a dozen regulatory warnings to visit a website if you don't block it all.
reply