jjnoakes

9 ms·

61.
by jjnoakes 3y ago
I see people rant about this, but not about other equivalent ways of getting scripts from the author to the system and running them. If the instructions were "download, unpack, and run ./configure; make; make install" would y
62.
by jjnoakes 3y ago
I've been working in a similar environment, and once a month (or more) someone comes to me with a crash they don't understand. It usually takes a couple of days to debug (since the easy bugs don't make it to me), and every si
63.
by jjnoakes 3y ago
It's trivial to store sensitive data in a collection, for example, and then index it with the wrong index when trying to access and export adjacent non-sensitive data.
64.
by jjnoakes 3y ago
You absolutely can.
65.
by jjnoakes 3y ago
> I have all the definitions and vars in my REPL and so has my IDE. I get completion on all symbols, I can look at call trees, usages, refactor things with confidence, run functions in isolation, replace them, wrap them, all from the REP
66.
by jjnoakes 3y ago
I find that for many projects I start, I seem to want: - static types - small, static binaries for deployments - lower memory requirements and value semantics where appropriate - a reasonably large ecosystem - cross-platform wit
67.
by jjnoakes 3y ago
I think you can choose nuances that minimize unexpected behavior in practice, and I think the go team did a good job here.
68.
by jjnoakes 3y ago
There's a second motivation in my opinion. Code might work today without the change, but it could be because the author originally wrote buggy code, caught it in testing, and had to waste time tracking it down and understanding nuances
69.
by jjnoakes 3y ago
I think of Java as more bloated than C++ at runtime - with C++ I can make smaller executables that start up faster, use less memory, etc. But if you are talking about the source code itself, then things are quite different.
70.
by jjnoakes 3y ago
The safe alternatives are in the article.
71.
by jjnoakes 3y ago
I've effected positive change in the companies I've worked with in the past, so I guess I know firsthand of anecdotal counter-arguments.
72.
by jjnoakes 3y ago
Corporations are made up of people, and people can influence the actions of the corporation, if they try.
73.
by jjnoakes 3y ago
I tried it on Android and within the first few minutes I hit the following annoyances: - Changing the title of a note and then backing out of the note does not save the title change; I had to switch focus to the body of the note bef
74.
by jjnoakes 3y ago
Lifetimes are still important in managed languages. You just have to track them in your head, which is fallible. The difference is that if you get it wrong in a managed language, you get leaks or stale objects or other logic bugs. In rust y
75.
by jjnoakes 3y ago
You build them. Or if the latest compilers don't build against the older glibc (which is rare in my experience, but may be possible), then you just don't support the older glibc any more. But I've never had trouble building a
76.
by jjnoakes 3y ago
Depending on glibc can be portable enough if you build against the oldest glibc version that you need. Running on newer systems works.
77.
by jjnoakes 3y ago
That post describes "something followed by a colon is a label in certain contexts", so you need to extrapolate a tiny bit to realize that "$:" is a label, even if the post doesn't explicitly enumerate all possible l
78.
by jjnoakes 3y ago
> it allows players who get stumped at a certain level to save their high score to the leaderboard without finishing the game Have you considered saving the high score after every level automatically?
79.
by jjnoakes 3y ago
If by "learn" you mean "native mastery" then I suppose you are saying only kids reach that level, so they are "faster" simply because they are the only ones to finish. The article considers "learn" to
80.
by jjnoakes 3y ago
This is definitely true. I remember being able to quickly press and release the hangup button on a single phone and if I was quick enough the other person would remain on the line. I don't recall exactly where the threshold was, but I
81.
by jjnoakes 3y ago
The only thing clear to me is they are on a Mac and want a dev environment of some unspecified kind.
82.
by jjnoakes 3y ago
It isn't so clear to me. As I read it, OP might be asking about a dev environment on Mac, not necessarily for Mac-only development.
83.
by jjnoakes 3y ago
Config files work fine. But so does expressive code with the right abstractions. And I think code scales up better and down almost as well. Just my opinion.
84.
by jjnoakes 3y ago
Well-written code in an expressive language can be very close to a config file, but also has the nice ability to scale up in complexity when required.
85.
by jjnoakes 3y ago
I will look it up later, but maybe someone here knows off the top of their head. Does the rust tcp listener accept not fail with an EINTR equivalent upon receipt of SIGINT?
86.
by jjnoakes 3y ago
The claim above was that a for loop implied that the code couldn't be declarative. > Please stop calling it such I'm not claiming it is always declarative, I'm only claiming that a declarative example above can contain a f
87.
by jjnoakes 3y ago
> When someone tries to make a sophisticated argument that up is down and white is black This is where the deep misunderstanding is coming from.
88.
by jjnoakes 3y ago
The claim above isn't "imperative is impossible".
89.
by jjnoakes 3y ago
In general maybe, but in the specific context above, I think calling that loop declarative is accurate, and laughing at that classification is a poor response rooted in a deep misunderstanding.
90.
by jjnoakes 3y ago
This seems extremely dismissive and shallow. The imperative part of that code appears to be analogous to templating. The actual work done under the covers is not imperative, but is based on the difference between the result of the template

More ›