ftxbro

33 ms·

481.
by ftxbro 3y ago
> "an unnamed Effective Altruism organization that made YouTube videos promoting that troubling ideology." Excuse me, what? Effective altruism isn't a troubling ideology it's an amazing ideology. Just because they don
482.
Paul Graham Observes That Wokeism Seems to Be on the Retreat(twitter.com)
12 points by ftxbro 3y ago|18 comments
483.
Paul Graham Quantifies Social Media Usage of 10th Grade Girls(twitter.com)
2 points by ftxbro 3y ago|0 comments
484.
by ftxbro 3y ago
> you'd need to see something more existential than slightly increasing margins in worlds where existence is related to health care which costs money which is related to margins, tampering with margins can be existential
485.
by ftxbro 3y ago
Maybe he's talking about maturity, but maybe it could also have something to do with the world of acting versus other careers like medicine or science or engineering, or even possibly have something to do with being at that stage of li
486.
by ftxbro 3y ago
ok but it's an odd take for a tech site like hacker news i mean what does that say about the llm paradigm vs for example object oriented paradigm or functional programming paradigm were they just playing around before and now it's
487.
by ftxbro 3y ago
It's Crawford Lake in Canada which they are proposing as a kind of geological 'type specimen' https://en.wikipedia.org/wiki/Type_(biology) or 'golden spike' for the onset of the Anthropocene h
488.
by ftxbro 3y ago
meanwhile on hacker news https://news.ycombinator.com/highlights
489.
by ftxbro 3y ago
in case anyone doesn't know it, the latest hotness is to use low-background steel as an analogy for pre-ai-generated internet content. they will 'mine' this ancient content in the future bc it's relatively unpolluted by
490.
by ftxbro 3y ago
its ok you are one of hacker news lucky ten thousand i guess https://xkcd.com/1053/ probably i just am too old and go on these sites too much it would be like watching sesame street for twenty years and then complainin
491.
by ftxbro 3y ago
suckless is some hacker news nerd bait, it explains the seductive aesthetic of computer environments where every cruft is removed and your computer has exactly the things that are useful to you in it. also some of the main people in it are
492.
by ftxbro 3y ago
He is basically saying they are elites and that for some reason human society needs some people to be better than others. It's literally elitist in the worst sense.
493.
by ftxbro 3y ago
this is hacker news all time favorite wikipedia article. it needs to ascend to the pantheon in a list within a new tab at the top of the hacker news website and removed from new submissions.
494.
by ftxbro 3y ago
This seems like a step backwards to me. If AI will make like every job obsolete then we will have to add more fake jobs like this instead of remove them. The purpose of these jobs will be to filter which people are deserving of health car
495.
by ftxbro 3y ago
in a way yes and in a way no yes: it would actually be theoretically possible (but openai wouldn't allow you to do it for business reasons and it would be expensive to host and run) to download all weights to a large inference server c
496.
by ftxbro 3y ago
It's because novels are an older and more established and prestigious medium. There are "old masters" solo novelists like Dickens or Tolstoy or Jane Austen or whatever. There aren't such analogous dead indie devs and the
497.
by ftxbro 3y ago
maybe article would be good but this hacker news title makes me want to vomit, it makes me physically ill i'm sorry
498.
by ftxbro 3y ago
yes it's clickbait, which is sad because that blogger is an OG computational science popularizer EDIT: whoever is downvoting me, the article title is objectively clickbait
499.
by ftxbro 3y ago
so you're saying the title is accurate
500.
by ftxbro 3y ago
well it got your attention enough that you appeared to have read at least some of the article so i guess that's a win. everyone got to hustle.
501.
by ftxbro 3y ago
> Data quality like you're describing just doesn't matter. GPT is trained on PEBIBYTES of data. It matters a little bit, in a quantitative but not qualitative way. Probably with good data cleaning you could get as high quality
502.
by ftxbro 3y ago
I mean 'better result from less data' is at least a little bit possible. For example you can just clean out obviously bad data from the trillions of tokens data sets. It's things like the subreddit where they are counting to
503.
by ftxbro 3y ago
I've been following the LLMs and I don't think it's 'running out of road', except in the sense that it's expensive. The model perplexity keeps decreasing when you scale up the parameters and exaflops and datase
504.
by ftxbro 3y ago
at some point social media posts will put mistakes like this just to bait engagement
505.
by ftxbro 3y ago
i think the real reason is the enshittification of online spaces by seo and spams and bots and other things, and some reddit mods were somehow able to hold the line against it (wait am i defending a reddit mod i would never do it). i'v
506.
by ftxbro 3y ago
well yeah thats because gpt shamelessly bullshits while the truthcel humans are afraid their tribe will exile them. still not agi but nice try.
507.
by ftxbro 3y ago
Hello, human resources?
508.
by ftxbro 3y ago
i didnt read the article but im just going to assume they are talking about leaving for college or something like that
509.
by ftxbro 3y ago
is this some new hacker news poes law usually i can tell but this one is too spicy for me
510.
by ftxbro 3y ago
what if lying is the real superpower of ceos and all those marketing blogspams were also lies

More ›