fenomas

55 ms·

271.
by fenomas 3y ago
Yes, but that doesn't mean it's "illegal" to create a derivative work. It just means that the original author has rights they can exert over what you make.
272.
by fenomas 3y ago
I think you're confused, or at least talking past GP. It's not copyright infringement to create a derivative work - e.g. by adding illustrations to a book. What's infringement is if you copy and distribute the derivative
273.
by fenomas 3y ago
Reading this felt like a huge waste of my time. TFA picks out 4-5 realistically unavoidable features of AI training - like that a certain quality threshold value was chosen arbitrarily, or that there's more training data for English th
274.
by fenomas 3y ago
Not mentioned in TFA, but here's a fun Houdini fact I learned recently: shortly before he died, he commissioned H.P. Lovecraft and another writer to collaborate on a book debunking religious miracles. It was to be called The Cancer of
275.
by fenomas 3y ago
There are several variations of Hepburn, but none of them uses an apostrophe for small-tsu consonants. https://en.wikipedia.org/wiki/Hepburn_romanization
276.
by fenomas 3y ago
This is basically a case of the government removing a rule that almost nobody, including them, was really following. I guess they had it written down somewhere to use Kunrei, but like TFA says Hepburn is what's used virtually everywher
277.
by fenomas 3y ago
Surely it's more likely that Google is just appending random keywords to incoming prompts, the same way DALLE used to do (or still does)?
278.
by fenomas 3y ago
> And no part of the original source code is in the binary output. It's not about whether the binary includes the raw text of the source, but whether it copies the expressive content. Anything expressive (i.e. copyrightable) in a co
279.
by fenomas 3y ago
If a trained model doesn't meet the definition of being a derivative work, it doesn't matter whether the data it's not a derivative work of was curated.
280.
by fenomas 3y ago
Wikipedia: > In copyright law, a derivative work is an expressive creation that includes major copyrightable elements of ... the underlying work A trained model fails that on two counts, doesn't it? Both the "includes" par
281.
by fenomas 3y ago
Where this argument falls down for me is that "use" w.r.t. copyright means copying, and neither AI models nor their outputs include any material copied from the training data, in any usual sense. (Of course the inputs are copied d
282.
by fenomas 3y ago
This was fascinating to dive into for a little while. I'm not sure I'd enjoy it, but now I'm curious to read one of Keeler's absurdly improbable mysteries.
283.
by fenomas 3y ago
> That's precisely the kind of thing that turns people off codes of conduct. I don't know if you've ever moderated a large community, but if you do it by the letter of any law, people will endlessly probe for loopholes a
284.
by fenomas 3y ago
Here's a maybe more mobile friendly comparison: https://i.imgur.com/zUVn8rt.png But now that I double-check, I was comparing with the images zoomed to 200%. On desktop the artifacts are also noticeable at 100%, but not
285.
by fenomas 3y ago
At full size it's super obvious - I made a side-by-side: https://i.imgur.com/I6EQ05g.png
286.
by fenomas 3y ago
Here is a comment that must be helpful because it got upvoted, but it's impossible to understand because the sibling comment it replies to is hidden somewhere at the bottom of the comment list.
287.
by fenomas 3y ago
Another random app store anecdote: way back when (2010?) Adobe made a feature where you could publish flash content as an iOS app. Like you build a flash game, hit publish, and an .ipa file comes out. So the feature goes into open beta, and
288.
by fenomas 3y ago
Set app selling aside - I only mentioned that because it's a counterexample to your original post as stated. What TFA is about is selling in-app content from external stores. Apple used to ban that entirely, and now wants a revenue s
289.
by fenomas 3y ago
You might have meant something else, but what you wrote is: > they require apps on the platform to use their payment processing so that they can take that cut. Which is not true. Both steam apps and in-game content for steam apps can be
290.
by fenomas 3y ago
TFA is about Apple's policy on purchases that happen outside the app. It used to ban even linking to them; now it allows that but it wants a cut. Steam doesn't do anything similar - it has no rules about purchases outside the ap
291.
by fenomas 3y ago
I believe the rule you're talking about only applies to literal in-game transactions - i.e. the binary you put on steam cannot itself implement a non-steam wallet. But there's no business rule against selling in-game content els
292.
by fenomas 3y ago
> Steam. ... they require apps on the platform to use their payment processing That's not true at all. Steam literally lets you sell steam keys for your game from other stores, and takes no cut from those sales. https:/&#x2F
293.
by fenomas 3y ago
> I am no longer sorry for not assuming good faith from someone arguing that.. Sorry, but assuming good faith is like supporting free speech - if you only do it for arguments you agree with, you're not doing it at all. ;)
294.
by fenomas 3y ago
> Don't lie. We're not in youtube comments here - assume good faith and be civil.
295.
by fenomas 3y ago
That's my issue with it, really - I think it was widely presented and received as having "demonstrated" something, but if you look closely, by reasonable standards it didn't. Actually if you dig up the authors' init
296.
by fenomas 3y ago
That grievance studies thing was idiotic, and really ought to be dismissed and forgotten. Reasons: 1. The hoax authors didn't submit gibberish or anything, like with Sokal - they submitted coherent papers that they considered absurd,
297.
by fenomas 3y ago
I've been doing my part! https://news.ycombinator.com/item?id=35565212#35567418
298.
by fenomas 3y ago
> One of these days in your travels, a guy is going to show you a brand-new deck of cards on which the seal is not yet broken. Then this guy is going to offer to bet you that he can make the jack of spades jump out of this brand-new deck
299.
by fenomas 3y ago
No, by semantic information I mean information that's meaningful to humans but not to the JS engine. E.g.: // before prettier var matrix = [ 1, 0, 0, 0, 1, 0, 0, 0, 1, ] var result =
300.
by fenomas 3y ago
Sure, but the issue with prettier is that it goes beyond style and also mangles semantic details. That's their explicit design choice - they feel that conforming formatting in all cases is more important than preserving semantic inform

More ›