Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
whereismyacc
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
31.
▲
by
whereismyacc
2y ago
That sounds like it'd give you no hope at all of ever learning muscle memory for it.
32.
▲
by
whereismyacc
2y ago
would they have preferred not being able to live/work in the US at all?
33.
▲
by
whereismyacc
2y ago
you can grow the sector. there's not a fixed supply of jobs. getting more talented people into your economy will likely just lead to more companies, more agglomeration effects etc. it's good.
34.
▲
by
whereismyacc
2y ago
I agree with you, I've just struggled to find one that worked well on Android but also synced to desktop. I'm on iOS now, so maybe it's time to look through the options again.
35.
▲
by
whereismyacc
2y ago
My biggest frustration right now is paying for Spotify and not getting any ads in music, but getting ads in podcasts. Not even just the ads embedded in the audio stream, which I can understand Spotify can't do much about, but Spotify a
36.
▲
by
whereismyacc
2y ago
For every year that passes, storage becomes cheaper, but the total size of youtube's video repository grows. I wonder what the net effect of all that is in the end. Ever increasing costs? Or maybe it kinda evens out.
37.
▲
by
whereismyacc
2y ago
I believe the blog poster has an option to toggle this. Maybe the default behaves like this?
38.
▲
by
whereismyacc
2y ago
So to be clear this is just a blog throwing out unfounded predictions?
39.
▲
by
whereismyacc
2y ago
Isn't basically every modern cpu core OOO?
40.
▲
by
whereismyacc
2y ago
i tried an app like this at some point, the one where you had to solve a puzzle. I was just terrified that one day I'd set the wrong alarm and have it blaring during a university lecture when I couldn't turn it off.
41.
▲
by
whereismyacc
2y ago
i've noticed that copilot seems to do this pretty well, i ask about a function and it correctly looks up the relevant lines of code
42.
▲
by
whereismyacc
2y ago
shouldn't this be possible to reverse-engineer since they stream the responses token-by-token?
43.
▲
by
whereismyacc
2y ago
Isn't this one of those things that is going to require a landmark case?
44.
▲
by
whereismyacc
2y ago
Securer DOM.
45.
▲
by
whereismyacc
2y ago
isn't the efficent market hypothesis just about asset pricing
46.
▲
by
whereismyacc
2y ago
by what law of economics?
47.
▲
by
whereismyacc
2y ago
Ah, that sounds fair.
48.
▲
by
whereismyacc
2y ago
A few paragraphs in the author seemed very pretentious in their writing style. > By now, you may well ask why I have paid so much attention to and have spent so much eloquence on such a simple and obvious notion as the radical novelty. M
49.
▲
by
whereismyacc
2y ago
Usually I'm listening to podcast when doing dishes, or grocery shopping, at the gym, or on a walk. I don't really want to have to take up my phone to skip an ad, and the controls on the headphones are too janky. This would definit
50.
▲
by
whereismyacc
2y ago
I assume the characteristics would be pretty different, since your local hardware can keep the context loaded in memory, unlike APIs which I'm guessing have to re-load it for each query/generation?
51.
▲
by
whereismyacc
2y ago
average says more about throughput, right? 95% would be nice too
52.
▲
by
whereismyacc
2y ago
Didn't we have this headline for Sweden like a year ago? Same deposit?
53.
▲
by
whereismyacc
2y ago
I guess the output would be an llvm intermediate representation that you can compile down and run, right? I'm stretching pretty far past my knowledge here.
54.
▲
by
whereismyacc
2y ago
There's been talk in the past about github adding integrations with common build tools (automatically?). What if you could compile every llvm-compiled project on github and run a diffusion model over the intermediate representations?
55.
▲
by
whereismyacc
2y ago
That sounds kind of reasonable? "community solar incentives" sounds like a really strange way to target funding
56.
▲
by
whereismyacc
2y ago
there are a lot of tricky parts.
57.
▲
by
whereismyacc
2y ago
I always thought it seemed likely that most needle in a haystack tests might run into the issue of the model just encoding some idea of 'out of place-ness' or 'significance' and querying on that, rather than actually say
58.
▲
by
whereismyacc
2y ago
oh interesting it actually works that way whether you zoom in or out
59.
▲
by
whereismyacc
2y ago
maybe the layout just messed up on the resolution I looked at it with, there were boxes all over the place, the spinny thing on the main page scaled out of its bounding box. also I think it would be helpful to have some consistent design cu
60.
▲
by
whereismyacc
2y ago
ah sick people from my uni it's pretty messy tbh but it's a good idea
More ›