Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
fwlr
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
28 ms
·
361.
▲
by
fwlr
3y ago
My understanding of the situation is this: At the request of Rust Project, JeanHeyd is asked to give a keynote at RustConf. JeanHeyd is reluctant, worried that their topic is too speculative and too much like a proposal for implementing a R
362.
▲
by
fwlr
3y ago
I remember this technology being a plot point in a Clive Cussler novel, and then being standard issue on new boats in most Cussler novels after that. I see Tom Clancy also featured it in his books even earlier. It’s always cool when you fir
363.
▲
by
fwlr
3y ago
While I don’t at all doubt the public health advantages of the Covid vaccine, I do have one reservation about the whole thing that I feel was never properly addressed: when I was growing up, I was taught the reason we make vaccines mandator
364.
▲
by
fwlr
3y ago
I have a lightning talk (each person takes turns giving a ~5 minute lecture on something useful) I sometimes give when the opportunity arises, jokingly titled “How to be Sincerely Grateful for any Criticism”, it might be helpful for you her
365.
▲
by
fwlr
3y ago
Swapping batteries probably isn’t a good idea but swapping planes might be. A lot of desirable routes like SF <-> LA become possible with two flights, if an operator sets up a halfway point with charging stations you could disembark o
366.
▲
by
fwlr
3y ago
I’m glad you saw the irony in my reply. I was a little worried it would come off acerbic, like “since you use ChatGPT, you only deserve ChatGPT”. The real intent was double-edged, of course - to demonstrate that ChatGPT isn’t just dumb-code
367.
▲
by
fwlr
3y ago
This… is a very strong objection and I think you’re right that users would be convinced. I didn’t think capabilities would be a panacea but I did think they would be pretty close, I have to revise that down somewhat.
368.
▲
by
fwlr
3y ago
For over 20 years now, professional woodworkers have had SawStops, devices that literally use an explosive charge to ram a block of aluminum into the blade of a table saw when it detects that the blade is touching something that might be a
369.
▲
by
fwlr
3y ago
A counter-example where a permissions system mostly still works is Mac apps. They also have to request permissions for things like using the webcam, accessing important areas of the filesystem, reading keyboard input when they’re not the ac
370.
▲
by
fwlr
3y ago
Capabilities-based security prevents these supply chain attacks! (Even though in this case most of the downloads were of packages including unwanted telemetry and not something more dangerous.) “…And now if a [color theme] wants to read you
371.
▲
by
fwlr
3y ago
It does re-ignite an old interest I have in, I don’t know the right name for it, but I think of it as “seafloor” neural networks, where they aren’t equally deep in layers all across the network (as opposed to “swimming pool” nets which are
372.
▲
by
fwlr
3y ago
I asked ChatGPT and it said “This limit can be adjusted with the `+P` flag when starting the BEAM… To adjust the limit, you could start the Elixir application using a command like the following: elixir --erl "+P 1000000"
373.
▲
by
fwlr
3y ago
Another major factor that goes into “the main colors in an image”, but doesn’t get picked up by these services, is the “visual importance” of different colors. For example, in the image used in the post, you probably want the color of the b
374.
▲
by
fwlr
3y ago
“I don’t expect privacy in an airport” doesn’t imply “I don’t expect privacy in my bathroom”.
375.
▲
by
fwlr
3y ago
That is a fair point. I don’t think it saves this particular analysis, of course. Probably what’s important to know is something we might call the “Pareto naive performance”; with just a little bit of effort (the first 20% or so), you can g
376.
▲
by
fwlr
3y ago
They do get closer to their rated volume than I expected, that’s true - I honestly expected the best would top out at 120dB even close up, just because these things are necessarily mounted within a few feet of the truck cabin and that would
377.
▲
by
fwlr
3y ago
Big fan of quiet. My home environment sits at about 30dB as measured by NIOSH’s sound level monitoring app, 32 during fridge cycle, 35dB when I have air conditioning running. In previous places I wasn’t able to control external noise as muc
378.
▲
by
fwlr
3y ago
Yes, it could. I believe the paper does mention this.
379.
▲
by
fwlr
3y ago
To play Devil’s Advocate, this is a good test of “how many concurrent tasks will an uncritical developer relying on ChatGPT end up achieving with low effort”.
380.
▲
by
fwlr
3y ago
Unless I’m missing something in the writeup, it doesn’t mention how much RAM is on the system. A little odd to mention the CPU and the OS but not the RAM, for a benchmark that doesn’t care about execution speed but does care about memory us
381.
▲
by
fwlr
3y ago
They introduce “height” as another architectural dimension, alongside the usual width and depth. If you imagine the usual diagram of a neural network, the difference when a neural net is of height 2 is that in the middle layers, each indivi
382.
▲
by
fwlr
3y ago
The idea that text contains more information than language is fascinating , wow. Thank you!
383.
▲
by
fwlr
3y ago
I’m not really sure how to respond other than to say no, that’s not what I was doing at all. I don’t think progressives, conservatives, liberals, neoliberals, etc., are infantile, I’m not trying to feel like I’m better than any of them. And
384.
▲
by
fwlr
3y ago
I’m not as surprised as many people (I saw a great tweet once that said “Language was the first artificial intelligence, writing was the second. I literally believe this”, and that’s a broadly accurate description of my worldview too). That
385.
▲
by
fwlr
3y ago
> A human being who did that without ever playing a game would also start out better than a typical novice. I am quite skeptical of these arguments along the lines of “imagine a human read everything written on the topic…”. What humans a
386.
▲
by
fwlr
3y ago
> The large language models are a little surprising. I’ll give you that. I think this is the key point about LLMs that kind of explains the wide and polarized views on whether it understands or parrots, whether it can think or is the pre
387.
▲
by
fwlr
3y ago
Might be fighting the tide on this one, I’m afraid. The basic moral intuition of the progressive world view is that there is good in everyone, we should empower that goodness, and those with the most good in them should be most empowered to
388.
▲
by
fwlr
3y ago
Fantastic, I just tried it out and it makes a lot more sense. I should have played around more with the form!
389.
▲
by
fwlr
3y ago
I’m a little bit interested in intuitive properties that high dimensional space retains as well, just to get a handle on how weird these shapes get. Like, you can inscribe a circle inside a square and have it touch all sides without exten
390.
▲
by
fwlr
3y ago
The history of 20/20 being considered normal is quite interesting, essentially it comes down to being simply what Herman Snellen defined as normal (on the plus side, he had a few years practice as an ophthalmologist before he invented
More ›