The irony is, unlike the old days - actually AMD (ATI) is recommend for Linux now because the drivers are better.

This is in stark contrast to the fglrx days where that driver was an absolute abortion and NVIDIA was really the only usable one.

Not sure when you started your Linux journey but I avoided AMD for years based on that.

Now the tables have turned but I didn’t realize until after I purchased my NUC which has NVIDIA RTX graphics. So I guess I’m stuck on NVIDIA for the foreseeable future

unishittification, that’s a new one.

I like it!

^ this guy corporates

Also, new manager would be part owner in a UX design firm of “experts” that conveniently, via their expert advice, convince management that a major redesign is needed and their firm is the only one that can do it (since everyone knows you can’t get expert advice internally)

80% of the way through the project, the manager gets promoted and moves on, leaving a new manager with no vested interest in their predecessors project to try and clean up the steaming dumpster fire that is now 300x over budget

Yeah, but, like, how many hours will it take for you to deliver 5 complexity points?

I hope this never happens to me but based on the Peter Principle I won’t know when it happens.

Oh shit maybe it’s already happened…

I was thinking about this the other day. Because Lemmy instances keep defederating from each other, I don’t really experience Lemmy. I experience a fragment of Lemmy as determined by the admins of the instance I’m connected to.

Even if I run my own instance, I guess there’s nothing stopping instances from defederating from me (or just refusing to federate to begin with because my instance is too small to bother with).

Is there even a way to experience all of Lemmy, including spam and things some people don’t agree with?

But it’s genuinely what we were all doing not so long ago

Jokes on you, my first job was editing files directly in production. It was for a webapp written in Classic ASP. To add a new feature, you made a copy of the current version of the page (eg index2_new.asp became index2_new_v2.asp) and developed your feature there by hitting the live page with your web browser.

When you were ready to deploy, you modified all the other pages to link to your new page

Good times!

The first thing I noticed. I was confused, thinking maybe they had an old XP machine lying around to plug in after the main one failed, but then I read further and it was just a stunt

Hybrids: the worst of both worlds.

If you want to keep relying on gasoline then just buy an ICE car

It’s like rebrands.

Most rebrands occur because the average marketing person is pretty average and “rebrand” looks good on your CV.

A couple of million later, half way through, customers hate the new brand and the marketing people who started it have already left for greener pastures

Redesigning a perfectly good design that everyone is used to allows you to put “designed Netflix user interface” on your CV, and since management has to spend a ton of money on it, suddenly your team is worth something

Well, as far as I’m concerned Skype for Business set the benchmark for terrible. Teams isn’t even close to being that level of bad

Teams is relative.

At a previous job (Microsoft shop but in the public sector so 10 years behind), the standard messenger when I started was Skype for Business.

In case you’ve never used Skype for Business, it’s “Skype” in branding only and actually has nothing to do with the Skype software that Microsoft purchased and is more like MSN Messenger.

Compared to that, Teams is a huge step up.

Also, at a Microsoft shop, you have to use what Microsoft provides even though it’s usually balls.

It’s 90% of the reason I now refuse to work anywhere that’s bought into the Microsoft ecosystem. It’s just so… mediocre

Actually, it makes perfect sense.

The loose terms like morning, noon, night etc are related to the suns position in the sky and exist regardless of what the wall clock happens to say

var context = RuntimeSingletonFactory.getCurrentFactory().getCurrentRuntimeSingleton().getContext()

As someone in Java land, you might be more impressed about its memory footprint rather than its performance.

Your Java hello world that takes 4GB of JVM heap space or it will fail with OutOfMemory would likely be only a few mb in Rust (or even less)

Came here to mention this

There’s nothing wrong with putting Rc<> or Rc<RefCell<>> around data

It’s mainly the visual pollution that bothers me. Wrapping everything in the reference counting smart pointers just because you can’t be bothered dealing with the borrow checker seems like an antipattern

At the end of the day, the first thing managers do is convert story points / tshirt sizes / whatever other metaphor back into time estimates. So why bother with the layer of indirection.

I’ll die on the hill that most teams do not need scrum / agile and all the ceremony that always goes with it.

A kanban board with a groomed Todo column is all you need. Simple and effective and can easily adapt to unexpected scope changes a.k.a production incidents.

*yes I’m aware that if you’re getting bogged down in ceremony you’re doing Agile wrong. I’ve never seen or worked in a place where I’ve felt it’s been done right

I don’t know if YouTube died per se, but it certainly became enshittified.

Heaps of content creators sold out to advertising interests and degraded their own platform. Not that I blame them really, money always talks

Front end programmer doing full stack:

Apes together strong!