If you compare a compiled language you should also mention the time it takes to compile. Although that will probably still be less than 200s.

Also with the right libraries python would probably be a lot faster.

Well, time to build our own cpus...

If you access something not by ip but by dns name from inside a docker container it doesn't always work because docker has its own dns server that does not respect all local dns resolver settings.
But can't remember how I fixed that...

"bigger team" lead. Or I guess just some sort of manager. Like "Project manager". Or you will just get some other random made up title.

I strongly disagree. While pseudo xml inside js files is not ideal. Writing a lot of small self contained components is still way better than writing one big ass html template and manipulate it later on.
Also react native is way better than native Android (if it works for you use case). But I still prefer flutter or kotlin like sytax for building ui over xml and jsx.

My problem here is that $item is not scoped to the foreach. That really drives me nuts. That would make this behavior impossible and is in general what you actually expect.

@CptFox sadly I have to agree on qa of Lenovo. My T460 is great except a couple of months ago it refused to boot. And now I noticed yellow tint at left edge of the screen.

Maybe there were a lot of other applicants that were faster than you.
But it still is a stupid reason to exclude someone.

@highlight I think there is a crate to do something like that. But not sure what it was called.

@Demolishun No. Rust enums are just ADTs. Meaning a variant of the enum can be a struct and contain data.
In contrast enums in Java are fancy ints

@12bitfloat == doesn't work if the type in option is not eq. But you don't need a match block to pattern match. If you just care about one case you can use

if let Some(_) = a {... }

But for Option there is also a.is_some()

I think the biggest "disadvantage" of rust in comparison to e. g. Java (because I don't know C) is that you have to think about your program before you write it. Otherwise it won't work or you have to rewrite it.

@Codex404 it's always more complicated than the studies make it look and I haven't looked at them myself. That is just what the book sais. But the author is referencing every study she talks about so I might take a look at some.

The book also sais brainstorming via online tools is more productive than in personal meetings. It also talks about that our perception is influenced by what others in a meeting say. Not just what we think about a topic but actually how we look at it.

I'm currently reading a book called Quiet by Susan Cain. It's mostly about introverts and how the "team work culture" is fucking them/us up.
And it quotes a lot of studies that say working alone is way more productive. Even for things like brainstorming. An open office is better for social interactions and asking your coworkers for help. But it is way worse for productivity and quality of work. Home office might also not be great because there are probably a lot of distractions.

When they taught sql in university they used like 3 different sql servers because none of them had all the features. And if I remember correctly postgresql had all the features. And even with the same syntax as the sql standard.

Depending on how you use OOP I strongly agree with that statement. I think class hierarchies are the wrong way to make your program extendable. I think typeclasses/traits like in haskell or rust are much more useful and intuitive.

@irene my point is you can use the same function to calculate the nth fib and to calculate the first n fibs. And if it's lazy it only uses as much space as it needs to store what you actually use later

@irene if you write a function to return the first n fibs and you just want the nth fib and not the others the calculation will take up the space of n ints. With lazynesa it doesn't because the compiler knows you only want the nth.
If the function is lazy it is more versatile

@irene we were talking about lazy evaluation a few comments above. And it is one of the main reasons e.g. Haskell is so fast and efficient. If you calculate the 10000th fib you take up literally no ram because you throw away the others. But you can still use the same function to calculate a range of fibs if you want.

@irene Fibonacci is just a very simple example everyone should understand. And it demonstrates nicely why lazyness is a nice feature

@arthurhoch I haven't tried it myself. But if it doesn't work that sucks... But then it is probably a bug. Maybe open a bug report if there is not already one?

I found out that you can specify the timeout per unit. See
https://freedesktop.org/software/...=
In the case of stop job running... It probably enough to specify it on you login manager. Btw there is also a DefaultTimeoutStopSec

If you already know js and especially if you already know react I would learn react native.
With flutter it is probably easier to build material design apps but in my opinion the ecosystem is to immature.

The problem with systemd is there might be better alternatives (and I'm not even sure there are) but I'm not gonna build my own distro and changing a distros init system does more harm than good. Why systemd is not perfect it works just fine. I just wish it was a little less complex.

John Cena would be a good skill to have😂

From what I heard the extensions broke because a certificate expired not because of a forced update.
And at least for me it was fixed the same day it broke. But I'm also using nightly so I might have gotten the update earlier.

@Fast-Nop if you implement them both the same way I expect they have almost the same performance. But if you implement fibs in one line in haskell it probably takes more time to print the numbers then calculate them. I can calculate the 10000 fib pretty much instantly. To get the same performance you would have to implement it either iteratively or cache all fibs in other languages. Also in other languages you need something like BigInt which is usually annoying to work with.

I would recommend writing property tests and not only unit tests.

@AlgoRythm @AlgoRythm lazy evaluation is also one of the reasons why you can write e.g. Fibinacci function in one line and it is so much fucking faster than all conventional methods

I really don't think there is a point in investing a lot of time into thinking about this. Either way someone will get killed at some point. And if it is the driver or a passenger their family will be angry and if it is someone in the street their family will be angry. It will be dramatized in media either way. And politicians will probably say they will do something about it. But ultimately you can't. The only thing you could do about it is analyze how human choose in these situations. At least that way autonomous cars won't be less ethical than humans.