treesprite82

7 ms·

1.
by treesprite82 4y ago
> A code block is not the same as a paragraph. [...] Right, but for Markdown they're a difference in output despite the input \n\n being just whitespace, and for the final text they have semantic effect despite being just a gap be
2.
by treesprite82 4y ago
I think diminishing returns applies regardless of whether it's improving itself through optimization or breakthrough new ideas. There's only so much firepower you can squeeze out if everything else were to remain stagnant. Like if
3.
by treesprite82 4y ago
I think the original article's chosen image plays well with unsettling/uncanny nature of current AI generated images, faces in particular. Definitely disagree with claims of "it's never gonna look right" in the twee
4.
by treesprite82 4y ago
> capable of constructing a slightly better version of itself With just self-improvement I think you hit diminishing returns, rather than an exponential explosion. Say on the first pass it cleans up a bunch of low-hanging inefficiencies
5.
by treesprite82 4y ago
> a couple of newlines A couple of newlines to create a new paragraph is semantic whitespace, as one newline or a space would not do so. Markdown (although not HN's Markdownesque syntax) even has significant trailing whitespace, w
6.
by treesprite82 4y ago
> Don’t force a kid to walk in front of your 4,000-pound metal box traveling at god knows what speed Some consolation, according to the tweet chain, is that it'll be at 5MPH and the child wouldn't be in front of the car (just o
7.
by treesprite82 4y ago
Could you upload a screenshot from your phone? Searching images online I see plenty that roughly match the colors they're using on the video/website, but none as dark as your iPhone. Reference: https://i.imgur.com/
8.
by treesprite82 4y ago
The distinction they're drawing is between ignorance of the law (e.g: you take someone's phone because you didn't know stealing was illegal) and lack of knowledge/criminal intent relating to the act you committed (e.g: y
9.
by treesprite82 4y ago
No knowledge/criminal intent relating to the act can be a defense (other than for strict liability crimes like statutory rape), ignorance of the law is not.
10.
by treesprite82 4y ago
Cut and dry if the open-source contributors are "blocked persons", but are they? They'd no longer be allowed to contribute to the project, but does someone having contributed in the past mean the sanction on the project exten
11.
by treesprite82 4y ago
> ...your response to which is to liken me to a dog while crying bully Dog whistles are blown by humans. Arguably it implies the people you're trying to reach are dogs, but I don't think the analogy is intended to be taken th
12.
by treesprite82 4y ago
> needs to be cleaned up[1] as it's way too blurry to be seriously used as an app icon Seems to have been blurred after the fact. The version linked in the article before cropping looked fairly sharp: https://jacobmartins
13.
by treesprite82 4y ago
For convenience features advertised as requiring active driver supervision and not making the vehicle autonomous, I think it only really makes sense for the driver to be liable. Similarly, if some airplane instrument is stated to require re
14.
by treesprite82 4y ago
Something that just flies an aircraft straight and level is still an autopilot, so I don't think it's a totally inaccurate metaphor. Ironically in these situations a more understated name would probably work better for PR. Headlin
15.
by treesprite82 4y ago
> But ask which is heavier, a goldfish or a whale, and it will tell you a goldfish. Or ask what Napoleon said about hamburgers, and it will say, “Hamburgers are the food of the gods.” Both are answered correctly by GPT-3 on OpenAI Playgr
16.
by treesprite82 4y ago
The full video that supposedly causes the auto-ban is in the article, so I decided to just try it: https://i.imgur.com/WEP4dcV.png I assume "additional tasks" means needing to delete the tweet to be unsuspended, a
17.
by treesprite82 4y ago
Note that I'm being intentionally reductive to argue that the net buscoquadnary threw around ML models and future-telling with sheep entrails also includes the scientific method. I do also think that ML as a field progresses through th
18.
by treesprite82 4y ago
> I mean basically that is what ML is, trying to predict the future If being so reductive, that's also the scientific method. Form a model on some existing data, with the goal of it being predictive on new unseen data. Key is in fav
19.
by treesprite82 4y ago
An example: https://i.imgur.com/FtwnUTk.mp4
20.
by treesprite82 4y ago
> There's just something I can't describe with words about what it lacks today - where it's possibly more philosophical than technical My personal viewpoint is that there's nothing inherently special about biological
21.
by treesprite82 4y ago
> based on all the data it's seen before? > it's closer to another virtual layer on top of the ever growing virtual layers I'd claim similar can be said for us. Don't forget to count the billions of years of evolut
22.
by treesprite82 4y ago
> [Video @ 21:50] I don't know about you, I put my car in reverse and if it gets too close to something it beeps because it's aware of its surroundings, right? Does that make my car self-aware? I wouldn't call sensing the
23.
by treesprite82 4y ago
Default behavior being for a single mouse button to paste and run (if it contains newlines) whatever's on the clipboard always seems dangerous to me. Beyond the accidental mistakes, I bet a website could catch a lot of people out about
24.
by treesprite82 4y ago
> The absurdity of Microsoft and OpenAI’s legal justification for GitHub Copilot. > Note 1: See Comment Regarding Request for Comments on Intellectual Property Protection for Artificial Intelligence Innovation (application/pdf) s
25.
by treesprite82 4y ago
A school day ending at 2pm already has this issue - this just splits it between the start and end of the day. For some parents that may be better, for some it may be worse. There's also the possibility for before-school activities or i
26.
by treesprite82 4y ago
> So it shouldn't matter. It definitely does matter that any secrets generated were already public: * Emitting secrets from private repos would be a huge confidentiality issue (though really you shouldn't commit code secrets to
27.
by treesprite82 4y ago
What's described as impossible is "[keeping the license] attached to each chunk of code as it wends its way through the model". Checking generated code for similarity against training set is possible, and is now done by both
28.
by treesprite82 4y ago
> There was many personally identifying information leaks when it initially launched > The aforementioned PII and secrets emitted It's not trained on private GitHub repos. Any secrets it generated would have been already public a
29.
by treesprite82 4y ago
> Asking for two different images in a series that have similar "art styles" is going to be enough work to still need a specialist aka an artist Running a separate style transfer network on the generated images is currently pos
30.
by treesprite82 4y ago
Do you recall/have a link to such examples? Would be interesting to try them again with the filter. The example I can remember was Carmack's* quick square root - but I'd probably call that "folk code" given it was p

More ›