Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
boxed
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
42 ms
·
811.
▲
by
boxed
1y ago
The article links to a site with this definition: > A monorepo is a single repository containing multiple distinct projects, with well-defined relationships. It would be better if there were terms that delineated "one repo for the c
812.
▲
by
boxed
1y ago
It's a joke.
813.
▲
by
boxed
1y ago
Which emperor though? Flat design or 3d design? I do agree with the article that before the iOS 7 flat design rush the barrier to entry for indie app developers was super high because it's damn hard to make the iOS <7 style look goo
814.
▲
by
boxed
1y ago
Or "function arguments". Either way it's a stupid name.
815.
▲
by
boxed
1y ago
I think your math is wrong there. Look at France and compare it to Germany. "Less ecological benefit" in your statement would mean that taking the entire country of France away from CO2 emissions (for decades now!) is less importa
816.
▲
by
boxed
1y ago
Wind turbines spread microplastics always. Coal spread radioactivity and CO2 always. Nuclear spreads radioactivity very rarely. The semiconductors are just rocks basically so not a problem (there are other components that might be, but the
817.
▲
by
boxed
1y ago
Any? Solar and nuclear would like a word :P
818.
▲
by
boxed
1y ago
If you build enough of these sustained, the total amount of CO2 bound it them could be significant. Similar to growing forests or restoring peat bogs. But yea, growing forests is equally suspicious as a lot of carbon sink forests have turne
819.
▲
by
boxed
1y ago
> I would have assumed of course that wind turbines are net negative emissions, even factoring in the construction and materials. It's net much-less-than-coal and much-less-than-oil, but it's not zero and certainly not negative
820.
▲
by
boxed
1y ago
> AI does work with competent people behind the wheel So does extremely junior devs that are really bad but you code review EVERYTHING. (Except jr programmers can learn, AI models can't really, they can be retrained from scratch by
821.
▲
by
boxed
1y ago
Maybe he meant it was long. Some people seem to think that long walls of text is how you spot AI slop.
822.
▲
by
boxed
1y ago
You missed the point of the article. It's not about clean code. It's about working code.
823.
▲
by
boxed
1y ago
Clean code is nice. Working code is a requirement. You missed the point. AI slop doesn't just fail on point 1. It fails on point 2.
824.
▲
by
boxed
1y ago
It's like every time AI "can code" and then falls on its face when presented with super basic problems that are outside its training data. Why would protein folding be different?
825.
▲
by
boxed
1y ago
Could once is not the same as can predictably first of all. Second, how can they possibly know this fold isn't actually in the ENTIRE PDB? I doubt very much that the can. The PDB is enormous.
826.
▲
by
boxed
1y ago
I honestly have no idea what you meant by any of that.
827.
▲
by
boxed
1y ago
If I was Ives, I'd sell some of that stock.
828.
▲
by
boxed
1y ago
Relative to cave men those sweatshop workers are so privileged they are basically gods. But that's a rather silly statement imo.
829.
▲
by
boxed
1y ago
And you know.. simultaneity is not a thing in reality...
830.
▲
by
boxed
1y ago
Yea, except that we see "breakthrough" stuff like this all the time , and it almost always is quickly found out that it's fraudulent in some way. How many times are we to be fooled before we catch on and don't believe p
831.
▲
by
boxed
1y ago
I'm personally waiting for the other shoe to drop here. I suspect that, since nature begins with an existing protein and modifies it slightly, AlphaFold is crazy overfitted to the training data. Furthermore, the enormous success of Alp
832.
▲
by
boxed
1y ago
Yea, the thing with the transparent APIs thing in iommi is that there's an admin and you can customize it just as much and as detailed as your own code. So you don't get the thing with the Django admin where you get a rough thing
833.
▲
by
boxed
1y ago
Going past a line of cars that were backed up and going like ~10 km/h and some guy in a pickup truck suddenly decided to force his way into the line and hit a car so it jumped into my lane by a meter or so. Only time I have ever needed
834.
▲
by
boxed
1y ago
What I've taken away from DDD is that consistent naming is important. Consistency > correctness even. But I think most programmers should already be pretty much on board on that one anyway.
835.
▲
by
boxed
1y ago
The first time I drove in the US I came up to a turn with the yellow speed sign. I was going faster and I could feel my car strain to not go off the road. This is to this day the second scariest thing I've experienced in a car. After t
836.
▲
by
boxed
1y ago
The searing steak thing doesn't have any real consequence though, except maybe save you a bit of energy when doing the searing. Compare this to the same tradition-repeating bad advice in self defense/martial arts and it becomes su
837.
▲
by
boxed
1y ago
The forms https://hexdocs.pm/ash_phoenix/AshPhoenix.Form.html seem even more low level than Django forms, and those are notorious for being too low level and annoying to work with. This to me seems to run counter to th
838.
▲
by
boxed
1y ago
So it's different in visual design, but the goal here is actually to make shadcn more like bootstrap? Hmm.. I think I like that.
839.
▲
by
boxed
1y ago
Can someone eli5 how this is different from something like bootstrap?
840.
▲
by
boxed
1y ago
I've done some work on making pytest faster, and it's mostly a case of death by a thousand paper cuts. I wrote hammett as an experimental benchmark to compare to.
More ›