It's going to be difficult for you to ignore Americans, the Internet is lousy with them.

I'm discussing association football. I like to think we're talking about the same thing, but there seems to be some disagreement.

It would be like saying "it's 36 degrees out" and not specifying Celsius or Fahrenheit, then getting tied up into knots if someone asked for clarification. Nice to know if I need to wear more layers or fewer.

The training set is itself presumably typically infringing whether the output is or not.

I personally don't think the output is infringing, but there is a plausible legal argument to be made that it is if the output is similar enough to copyrighted input.

I used to get pretty much all the cheap humble bundles unless I was completely uninterested in them, so I have a moderately sizable steam library, so I know there's plenty of games on steam which would run fine on a chromebook. They are all either old or indie (even vaguely modern versions of Bejeweled will cause the fans to spin up on your GPU at higher resolutions) but there's tons of games which would work fine.

Which then brings us to... there's way more than 99 of them. Probably most of them are visual novel type "games" by now, but those games wouldn't suffer at all for running on a very limited PC. Was someone (google or valve I guess) preventing those games from being used for fear some kid might see hentai on a google device or what?

That most likely was MacOS 6, unless it was an even older version. Few people installed 7 on B&W macintoshes, most software at the time would run on either or even ran better on 6 due to the lower resource use. I was familiar with other operating systems and found it to be the least impressive, but once you got used to it, it was pretty usable. My mother was an old school physical pasteup graphic artist, and she got a IIci and Pagemaker and was able to use the system with very little help from me. It came with System 6, and was upgraded to 7. Later I ran netbsd on it.

I like how we have cool engineering and cool computer science posts today instead of just legal battles, doomer gloom, and political intrigue.

> They now embark on a new task...what's wrong with Google these days?

That's pretty much status quo. They do some stuff, get users, get bored with it, kill the project.

See y'all in the reeducation camp, they're taking me sooner rather than later.

You can and should ban people in real life. Refusing to interact with people who are not interestes in a mutually beneficial relationship is normal. If someone car jacks me, I cross him off the invite list to my BBQ.

Unfortunetly the scalability of AI is not infinite. There is a point of diminising returns. Some of what AI researchers are up to is either doing more with less, or with making bigger clusters that are better connected.

I am a cruel master, I have never even let it download a software update.

Using a few simple truths to cover up bigger lies. Oldest trick in the book. The dude is a shit bag and you need to start questioning the motives behind every single thing.

My opinion, and this is simply an opinion, I don't have proof. Is that Trump is anti-capitalist. He does not want to see free enterprise in this nation. Instead he wishes to consolidate the power of industry under the Executive Branch. He essentially is taking steps to create a very American form of a fascism to supplant liberty, democratic, and capitalism. The Soviet Union could not beat us in the Cold War, but it turns out we were far more vulnerable to an enemy within than we realized.

I dunno....just plain sounds like this is a Chinese Communist govt. asset in charge of the US's one chip manufacturer....

If I were IBM that fucker would be out quick...if for not for the security risk, but the coming stock drop.

It could be a 3 * delta ~= 0 thing, except they say it's a bigger factor than smoking or obesity, which are known to be significant in absolute terms.

It is fascinating.

Since the summary does say "effective nightmare treatments are currently limited," I will go ahead and point out the correlation != causation thing. I would be very surprised if there weren't any other obvious causes of the nightmare that are known comorbidities, like poverty, or serious illness, or trauma, and so on.

For example, depression is associated with a 1.5x to 2.5x higher risk of early death, even when adjusting for other risk factors. The summary shows no sign of having adjusted for other risk factors, so 3x doesn't seem out of line.