Real answer: those things matter to me because a quick frictionless experience very heavily dependant on muscle memory really helps with my ADHD. Laggy interfaces, having to hold left key for several seconds, and similar issues quickly pull my out of my train of thought.

It’s not about shaving 2 minutes off my day, it’s about not interrupting the flow.

I’ve been dailying the exact same arch installation since 2014 without reinstalling it a single time.

Now to be fair I did have it non-bootable at several points. Worst of which was a PAM update which broke it completely because the new config was in a .pacnew file and the old one was not compatible anymore. But since it was a edge-case there was no forum post about it. Still recovered it just fine after an hour or so of troubleshooting.

It’s all open-source and usually decently documented. The only reason anyone should have to reinstall a Linux desktop is lack of experience, but I would always advise to persevere because troubleshooting my system is how I gained much of my expertise. If that’s not what you want, stick to Debian.

Well, yeah, that’s what Scrum is. From the guide which takes maybe 10 minutes to read

Scrum Teams are cross-functional, meaning the members have all the skills necessary to create value each Sprint. They are also self-managing, meaning they internally decide who does what, when, and how.

That’s not a throwaway sentence - it is fundamental to how scrum works and that is reinforced throughout the scrum guide.

Every conversation about Agile and/or Scrum being “the worst”, after some prodding it turns out that their company has refused to read or implement one or several of the fundamental principles, often without even being aware that was an essential requirement. You’re baking a cake and you decided to not use any butter, that’s on you champ, don’t blame the fucking recipe.

The biggest valid criticism of scrum is that the thing that makes it so great - its structural empowerment of individual teams - is also what makes it structurally incompatible with any traditional top-down management style. The company must fundamentally be (re-)organized to have a flat corporate structure within its R&D department - most are simply incapable of mustering the necessary changes, if only because too many middle managers’ jobs are at stake. So they call their middle managers “POs” or “Scrum Masters” and wonder why their version of Scrum sucks.

The last thing I want is merge conflicts in my issue tracker. The git data model is simply not right for conversational histories.

ActivityPub is the obvious solution to decentralize public communication. We’re using it right now and AFAIK Forgejo is working to implement it for their issue tracker.

I mean, Nestlé killed hundreds of thousands in impoverished countries. I think the Oracle service APIs can’t be that bad, though I’m sure they would be if it made Larry Ellison $50.

Ideally you’d use the docker executor with a dind service instead of docker commands in the shell. You’ll have better isolation (e.g. no conflicts from open port forwards) and better forward-compatibility (the pipeline won’t break every time a major upgrade is applied to the runner because the docker - especially compose - CLI is unstable).

For gitlab this is only correct with a shell executor which is to be avoided in the general case in favor of a docker or k8s executor for isolation&repeatability.

Those you can actually run locally with gitlab-runner, but then you won’t have all your gitlab instance’s CI variables so it’s a PITA if you need a CI token which you probably do if you actually make decent use of gitlab’s features.

In most cases I just end up committing my changes to avoid the headache. :!git commit --amend --no-edit && git push -f goes pretty dang fast and 60 % of the time third time’s the charm.

For systems programming it makes the most sense out of the languages you mentioned. Languages requiring a runtime (Java/Python) do not fill the bill for system tools IMO. Golang is more arguable, but its memory safety comes through GC which many systems programmers aren’t fans of for a variety of technical and personal reasons.

Rust is meant to be what C++ would be if it were designed today by opiniated system developers and didn’t have to be backwards-compatible.

Those are the technical arguments I would use in a corporate setting.

All that aside, there’s personal preference, and my point is that for FOSS projects that matters too. Rust is fun in a brain-teasy kind of way in the same way that writing C is fun, but without nearly as many footguns. Golang is practical but arguably not as fun. That’s the same logic that draws many programmers to write Haskell projects.

The story of the Fish shell illustrates it quite well; the project gained a lot of development attention and contributions when they decided to rewrite from C++ to Rust, where they achieved a stable release with feature-parity a few months ago. It would have been a remarkably dumb decision for a private company to make, but makes perfect sense when you are trying to attract free talent.

The counterpoint is that, especially with FOSS that does not receive much (if any) corporate backing, developer retention and interest is an important factor.

If I’m donating some of my free time to a FOSS project I’d rather not slug through awful build systems, arcane mailing lists, and memory unsafe languages which may or may not use halfway decent - often homebrew - manual memory management patterns. If the project is written in Rust, it’s a pretty clear indicator that the code will be easily readable, compilable, and safer to modify.

if its controlling expression is not a constant expression

Pretty big caveat. If I’m reading this right true definitely qualifies as a constant expression and the loop in the meme would therefore not be optimized away.

Strongly disagreed. A lot of removable storage benefits very strongly from async writes. The performance benefits are night-and-day and worth the hassle of explicitly unmounting.

I think a better compromise would be to mount non-journaled + removable storage (e.g. vfat USB keys) as sync.

Real answer: it depends.

• Deleting a file in use: no problemo. File is removed from the directory immediately, but exists on disk until last program who had the file open closes. Everyone wins! (Unless you’re trying to free up space by deleting a huge file that’s being held open by a program and not understanding why the filesystem usage didn’t go down)
• Unmounting a hard drive in use: Will error out similarly to Windows. lsof can tell you which process has which files open. There’s nuance with lazy unmounts and whatnot but that should not be used in most cases.

Now in practice you should be wary of one very important thing that changes compared to Windows: Writes are asynchronous on Linux. First the kernel writes to RAM, then it flushes to disk at a later time for performance reasons (this is one of the reasons why writing a bunch of small files is many times faster on Linux than Windows). The upshot is that just because your file copy is “done” doesn’t mean you can just yank the USB cable. Always safely unmount before unplugging a storage device on Linux.

GIMP apologists are what Adobe users think FOSS apologists are like. The principled refusal to even consider fundamental UX principles is astounding.

Being able to assign a nameserver per interface with a domain wildcard is a fucking godsend. I use it every day with a hook script because my job uses some private domains but I don’t want to send my entire DNS history through the VPN. Now ~job.com goes to tun0 and that’s the end of it.

systemd-resolved is not perfect but with libnss’s overly rigid nature the only alternative for my use-case would be to recreate similar functionality to resolved with dnsmasq – which is just objectively worse especially when you want to use DHCP sometimes but not always. Why reinvent the wheel? resolved does its job and does it well. I had some issues with it a few years ago but have been using it for the past couple years without complaint.

On windows maybe. Never ran into that on Linux. I understand it’s inconvenient but that’s not the format’s fault, it’s windows developers’.

It’s one of a plethora of scripting languages from the '90s which were designed to be the antithesis of “fail fast” and kept going no matter what.

I guess what with C/C++ being the Mainstream Option at the time, not having to deal with a strict compiler must have felt like freedom. As someone who has had to maintain, cleanup and migrate ancient PHP code, I call it folly. That mindset of “let the programmer just do whatever and keep trucking” breeds awful programming practices and renders static analysis varying degrees of useless, which makes large-scale refactoring hard to automate which is just amazing when your major versions aren’t even remotely FUCKING BACKWARDS COMPATIBLE.

PHP’s original design is just fundamentally atrocious. It became popular in large part because unmaintainable code is usually someone else’s problem.

A language that I would definitely use for server-side rendering and that was already good from its first stable release is Go. It was thoughtfully designed and lends itself really well to static analysis, while still being easy to write and decently performant.

If I am not mistaken the 47.0.0.0/8 ip block is for Alibaba cloud

That’s an ARIN block according to Wikipedia so North America, under Northen Telecom until 2010. It does look like Alibaba operate many networks under that /8, but I very much doubt it’s the whole /8 which would be worth a lot; a /16 is apparently worth around $3-4M, so a /8 can be extrapolated to be worth upwards of a billion dollars! I doubt they put all their eggs into that particular basket. So you’re probably matching a lot of innocent North American IPs with this.

Classful IPv4 was obsoleted 32 years ago. Only 8 years left before it’s literally older than a standard career.

It’s fascinating the sheer inertia that leads formally-trained IT professionals to use and perpetuate such profoundly useless and obsolete nomenclature. You’d think that having an incorrect use of the term “class A” and not having any use for classes B and C would tip off academia that they should cordon off classful networking to the “History of Computing” course next to ARPANET.

Maybe next time someone refers to 10.0.0.0/8 as a Class A network I’ll refer to it as the ARPANET Network. That’s only very slightly more anachronistic (3 years).

Nowadays “buggy” is not how I’d describe it, though there were certainly teething issues at the beginning. By now other DEs have learned to deal with it.

However it’s still true that the GTK4 design is ill-fitting, and very opinionated. Quite exemplary of this are the applications that hardcode the GTK file picker (like Firefox and chrome) even though it’s inferior in every way to the Qt file picker and forces the infuriating GTK “design” choice of doing fuzzy search when you type in the file list instead of jumping to the relevant file. Very annoying when dealing with organized directories especially when no other file browser on my system works that way!

• Eventually Company decides “agile will fix things”
• Developers are told to work agile but the only stakeholder they talk to is the PO, who talks to PM, who talks to Sales, who talks to Customers
• PM&Sales don’t want to deliver an unfinished/unpolished product so they give a review every sprint, by themselves, based on what they think the customer wants (they are Very Clever)
• A year or two later the project is delivered and the customer is predictably unhappy.
• Management says “how could this have happened!” and does it all over again.