This is nonsense. Old software was not fast. Computers used to take minutes to start up. That's not an exaggeration. MS Word could easily take 30 seconds to open.
Obviously if you run it on modern hardware it's going to be lightning fast, but not because the authors cared more about performance and efficiency than we do now.
Yeah it's obviously an improvement but I guess you can't really sell a distro on "it has a more logical filesystem layout which may cause some issues, and it's super super niche which will definitely cause some issues".
Probably a lost cause. I imagine Flatpak will actually work properly before the Unix filesystem hierarchy is made sane.
Wow you can tell from the first paragraph that this isn't worth reading. I read it... just out of curiousity...
For some reason the whole discussion around this Rust/C/Linux/GNU/thing is mostly focused around superficial and irrelevant things like the sexualities and genders of the Rust people
Err....
Rust people seem to be focused mostly on identity politics and dividing people into groups that are then supposed to fight each other. As I wrote earlier, I didn't invent the term "Rust people" myself - those people themselves identify as "Rust people", which is not a good thing. I code mostly in C and assembly, but I certainly don't identify as a "C person". I can also write other programming languages, and I would even learn Rust if it wasn't such a horrible Trojan horse that is clearly designed to destroy computing freedom.
.... yeah. I can confirm he has zero sane points. Let's not give this lunatic any credence.
So if you're important you can be an arsehole.
TL;DR: yes.
Git's autocrlf feature causes more issues than it solves in my experience. I don't think there are really any tools on Windows that can't handle Unix line endings any more. Even notepad can now.
I recommend you set it to input which will fix them to be Unix line endings on commit, and not change them back on checkout.
Before virtual memory was a thing, almost all memory was accessible.
Virtual memory has nothing to do with whether 0 is a valid address. You can have a CPU where it is valid, or one where it isn't and you'll get an access fault if you try to access it. You can also have virtual memory where page 0 is mappable, or not.
I think the author knew that, based on the later points so it's probably just bad wording. Interesting point about wasm too!
Have a go on your free time and see if you like it. There is an absolute ton of free learning material online. You don't need to pay anyone.
Most programming jobs (e.g. making web sites) are easy enough for the average person to do, but I think most people would find programming far too tedious and boring to learn.
It's like law - there's nothing particularly difficult about it but most people find it incredibly mind numbing to read legal documents.
So I would have a go in your free time first to make sure it is something you could do.
Ok first impressions:
I only made a very simple part, but I am impressed. This is significantly better than when I last tried it, when it basically didn't work at all. I haven't tried assemblies or anything complex (e.g. extruding up to a non-flat surface, degenerate geometry etc.), but definitely for simple tasks you could use it.
Yeah I think that's mostly a myth. When I looked up salaries they were definitely good (for programming; amazing for the average person), but not "I would write COBOL for that" good.
There aren't really that many old COBOL systems around. I think it's mostly just over-reported because you can write an article about how some government department still uses COBOL but you can't write about one that switched to Java.
Maybe not dumb but I've definitely been forced to at least partly learn a few terrible languages so I could use some system:
ityp, nsec, ef_bin... The sort of names where you already need to know what they are.
This doesn't explain why you'd want to use Lisp at all really, and it pretty much ignores the many disadvantages.