gsuuon

11 ms·

1. 91.▲
   by gsuuon 3y ago

   There are definitely still Current Thing's though - Game of Thrones, Walking Dead, Squid Games. They'll come and go. There's also the 'in-joke' kind of references from more niche shows that can form a stronger conne
2. 92.▲
   by gsuuon 3y ago

   I think the familiarity when moving from Matlab to Python helps a lot with adoption.
3. 93.▲
   by gsuuon 3y ago

   There actually is a similar project in F#: https://github.com/ScottArbeit/Grace
4. 94.▲
   by gsuuon 3y ago

   I wonder if that means pursuing work-related personal ambitions (i.e. promotion), or outside-work personal ambitions (i.e. side projects)? Could side projects actually reduce burnout?
5. 95.▲
   by gsuuon 3y ago

   You'd need some amount of arbitrary data (the copy) so the user knows what kind of content they can expect.
6. 96.▲
   by gsuuon 3y ago

   I think that point has come and gone. Not sure how society will adapt - I really hope smart people are out there working on this sort of stuff.
7. 97.▲
   by gsuuon 3y ago

   Wow - my first thought was I wonder what framerate they're sending video at. The whole demo seems significantly less impressive in that case.
8. 98.▲
   by gsuuon 3y ago

   I think these smaller models really struggle with the reasoning aspect of writing decent code, I'm getting pretty nonsensical things out when asking it to fix fizzbuzz (though it made the right fix), like: - Replaced "i % 2 ==
9. 99.▲
   by gsuuon 3y ago

   Whenever anything is down: "The system.. is down.. The system.. is down"
10. 100.▲
    by gsuuon 3y ago

    Llamanate
11. 101.▲
    by gsuuon 3y ago

    I think the take-away from Dunning-Kruger is that most people think they're closer to average than they are, on both sides.
12. 102.▲
    by gsuuon 3y ago

    I was recently looking at domains and it's kind of frustrating how many good names are parked or listed on some site for 6 figures. I mean, I'm guilty of it too (hoarding), but I wonder what the internet would be like if there was
13. 103.▲
    by gsuuon 3y ago

    Do we need to sign up to try it? The generate button doesn't seem to be working.
14. 104.▲
    by gsuuon 3y ago

    All the text that I assume should be on the monitor is hovering below the display and I only see it if I zoom out (Chrome on Win 11). Seems like it's working fine for others though?
15. 105.▲
    by gsuuon 3y ago

    The true savings were in front of us all along
16. 106.▲
    by gsuuon 3y ago

    I know nothing about this stuff, but as a bit of off-hand advice I'd guess leaning heavy into social media / influencers would be a good way to get traction for a fitness app. You could also organize local free workout events to h
17. 107.▲
    by gsuuon 3y ago

    I use my neovim plugin note.nvim[1] to brainstorm and plan features as well as track day-to-day stuff. It's one of the few projects of mine that I've spent more time using than building - by a good amount. I have a top level ~&#x2
18. 108.▲
    by gsuuon 3y ago

    The absolute uniformity was a bit disconcerting to be honest, but I can also see it being just a great display of comradery. I'm still unsure about how to feel about the thing with it mostly resolved.
19. 109.▲
    by gsuuon 3y ago

    At least they'll be operating under the original charter - it sounds like the mission continues. Not sure about this new board but hard to imagine they'd make the same sort of mistake.
20. 110.▲
    by gsuuon 3y ago

    Zephyr is coherent enough to bounce ideas off of, but I'm eagerly awaiting when open-source models are on par productivity wise with the big providers. I imagine some folks are utilizing codellama 34b somehow, but I haven't been a
21. 111.▲
    by gsuuon 3y ago

    How has the valuation of OpenAI increased by $20B since this weekend? I feel like every time I see that number it goes up by $10B.
22. 112.▲
    by gsuuon 3y ago

    Will Anthropic ever open up their API to hobbyist / personal use?
23. 113.▲
    by gsuuon 3y ago

    For pure productivity reasons it's not worth it unless you don't fiddle with your config. Messing with config can indeed be a massive time-sink, unless you enjoy it, in which case it simply becomes another hobby. It's definit
24. 114.▲
    by gsuuon 3y ago

    I think trying to judge morality by anything other than actions is probably the wrong approach, you could ask clarifying questions on some of their previous decision making but hypotheticals and such likely aren't a strong signal eithe
25. 115.▲
    by gsuuon 3y ago

    This looks cool! Just a note, it looks like the youtube playlist is in reverse order: https://www.youtube.com/watch?v=DSGXB9G5dxk&list=PLsydD1kw8j...
26. 116.▲
    by gsuuon 3y ago

    I think Roc has a lot of good ideas, especially the backpassing syntax sugar and that there's only one way to declare functions (anonymous or not). Excited to see where it goes!
27. 117.▲
    by gsuuon 3y ago

    I think a general 'explicit > implicit' priority is good enough to cover most cases - a written signature takes precedence over an inferred one. The compiler can also simply emit errors at both sites and let the writer figure i
28. 118.▲
    by gsuuon 3y ago

    Is there some sort of sparknotes or summary out there of the effective altruism and e/acc schools that would help catch-up the uninformed?
29. 119.▲
    by gsuuon 3y ago

    If Altman goes back it could potentially salvage the model he helped create - maybe there needed to be some mechanisms in place to validate decisions like firing the CEO. This drama was made all the more intense because no one still really
30. 120.▲
    by gsuuon 3y ago

    Isn't Ilya even more against opening up models? OpenAI is more open in one way - it's easier to get API access (compared to say Anthropic)