You can’t add AI chatbot to task manager without breaking a few eggs.
Vibe coding or no, this is a massive QA failure.
That is the reality.
The problem isn’t “vibe coding” (anyone who has ever managed early career staff will be able to attest that… the bar is REAL fucking low). The problem is a complete lack of testing or any sort of “investment” in caring if production breaks.
A lot of it is general apathy induced by… gestures around. But it very much goes beyond just the obnoxious rise in brain drains over “vibe coding”. Personally speaking, I am THIS fucking close to driving over to my company’s head of IT’s house and burning it down with him in it (For legal purposes, this is a joke) as that entire team continues to think “We’ll just wait until people tell us what is broken” is at all fucking acceptable.
But pretty much any SDLC is going to be built around code review. And code review is how you handle developers of different skill and sanity levels. Whether they are old hats who have been in the basement since before you were born, youngins who can’t stop talking about Rust, or chatbots.
Unfortunately a lot of people are trying to outsource code review to LLMs as well. Also, LLM generated code is more likely to have subtle errors that a human would be very unlikely to make in otherwise mundane code. Errors that are easy to gloss over if you don’t take a magnifying glass to it. My current least favorite thing is LLM generated unit tests that don’t actually test what they say they do.
Yeah I hate when I read through a unit test and realize it doesn’t actually test anything other than itself.
Shit code review is not code review. If you just rubber stamp everything or outsource it to someone who will, you aren’t doing code review.
Aside from that:
LLM generated code is more likely to have subtle errors that a human would be very unlikely to make in otherwise mundane code.
Citation requested
My current least favorite thing is LLM generated unit tests that don’t actually test what they say they do.
If I had a nickle for every single time I had to explain to someone that their unit test doesn’t do anything or that they literally just copied the output and checked against it (and that they are dealing with floating points so that is actually really stupid)… I’d probably go buy some Five Guys for lunch.
Its like saying that the problem is that you are using robots to assemble cybertrucks rather than people. The problem isn’t who is super glueing sharp jagged metal together. The problem is that your product is fundamentally shite and should never have reached production in the first place. And you need to REALLY work through your design flows and so forth.
Citation requested
I keep seeing it over and over again. Anyone that actually has to deal with coworkers using this bullshit that isn’t also in the cult is going to recognize it.
If I had a nickle for every singl yada yada yada
Sure, there have always been better and worse developers. LLMs are making developers that used to be better, worse.
Bad developers just do whatever. It doesn’t matter if they wrote the code themselves or if a tool wrote it for them. They aren’t going to be more or less detail oriented whether it is an LLM, a doxygen plugin, or their own fingers that made the code.
Which is the problem when people make claims like that. It is nonsense and anyone who has ACTUALLY worked with early career staff can tell you… those kids aren’t writing much better code than chatgpt and there is a reason so many of them have embraced it.
But it also fundamentally changes the conversation. It stops being “We should heavily limit the use of generative AI in coding because it prevents people from developing the skills they need to evaluate code” and instead “We need generative AI to be better”.
It was the exact same thing with “AI can’t draw hands”. Everyone and their mother insisted on that. Most people never thought about why basically all cartoons are four fingered hands and so forth. So, when the “studio ghibli filter” was made? It took off like hotcakes because “Now AI can can do hands!” and there was no thought towards the actual implications of generative AI.
Nothing outside of the first paragraph here is terribly meaningful, and the first paragraph is just trying to talk past what I said before. I’ll reiterate, very clearly.
I have observed several of my coworkers that used to be really good at their jobs, get worse at their jobs (and make me spend more ensuring code quality) since they started using using LLM tools. That’s it. That’s all I care about. Maybe they’ll get better. Maybe they won’t. But right now I’d strongly prefer people not use them, because people using them has made my experience worse.
The problem isn’t who is super glueing sharp jagged metal together.
I know it’s not related, curious about this part.
I know it has an aluminum based frame which should inhibit it’s use to haul heavy loads, but what else?
I have seen at least 1 out of every 5 comments from coderabbitai that lead me down a rabbit hole looking to see if the suggestion is correct. It can waste so much time trying to validate their suggestions only to find out it’s complete BS.
And automatic test systems. It’s not that hard.
You think the QA was done by a person? I’ll bet this was Vibe QA’d.
The public, aka Microsoft QA Team, found the bug. It’s a QA success!
Well, who’d ever think of testing that closing the app works. Some things you have to take for granted or you’ll never test anything meaningful.
If you are a company the size of Microsoft, you have more than enough resources to test absolutely everything.
You will never have resources to “test absolutely everything”. It is ALWAYS about building out personas and deriving tests from those.
What this tells us is that one of two things happened:
- This was not tested at all
- The testing harness resets the environment after every check (e.g. “does process close when killed”) rather than involving a manual reset (i.e. “close and re-open task manager”)
The latter is a lot more common than you would think since it makes it much easier to automate these harnesses rather than having a human at a VM. But… this is what happens when you don’t step through the entire workflow.
Didn’t Microsoft fire their dedicated human testing team in about 2014?
Probably?
This is the kind of thing that a LOT of companies outsource. Mostly for ill.
They used to have very comprehensive automated testing processes to exercise all sorts of things. Unfortunately, like many tech companies these days like Apple, Google, etc., they’re all punting QA as a concept because they just don’t care - what are you going to do, go use another oligopoly platform?
Well, there may have been a period when MS was trying to improve product quality, and in that time, yes maybe they did have very comprehensive automated testing processes. But before that, up to the time of Windows 7 I guess, their quality was dog shit.
In the early days, MS was an undisputed monopoly though, and not only did they not test thoroughly, they hardly even tried to fix bugs - the userbase had to take care of that too. Earlier versions of Windows had all sorts of workarounds and 3rd party tools to try and get things to work properly.
I suspect that once they’d achieved their objective of improving quality, there just weren’t the incentives there any more for middle management to allocate resources to things like comprehensive tests.
Isn’t that the final step in the testing plan for every app though? The first step is always opening it.
A bug like this means literally nobody tested it at all on this build, or was so apathetic they didn’t file an obvious issue.
For people to test, you need management that is willing to invest in QA. But that incentive disappears for a corporation when there’s no free market of competitors who can poach your customers by making a better quality product or service.
The person who fiddles with the always on top option in this monstrosity of a compositor maybe?
No problem! I can just kill the process in the…shit.
No problem! I can just kill the process in the…shit.
No problem! I can just kill the process in the…shit.
Nothing fucking works anymore.
I don’t think I could go back to Windows 10 permanently to do any non-web based work. I couldn’t even touch Windows 11. It’s so disgusting compared to my KDE plasma desktop on OST.
What is OST? Sorry, I’m just interested because I’m getting tired of Arch but I still KDE plasma 6
OpenSUSE Tumbleweed
TIRED OF ARCH!?!??!!!
Cachy or Endeavor are a nice step back from Arch while keeping things familiar. For even more stability without sacrificing your ability to fiddle, there is always Fedora.
+1 for Fedora. I’ve been using the same installation for 3 years. That’s multiple dirty upgrades. The only reason I’ll be most likely reinstalling is I got a new SSD and it’s a nice opportunity to start fresh.
I just finished my 42nd Fedora upgrade. I’m not counting multiple machines, just the number of releases where I’ve done an upgrade.
KDE apps work on windows generally so that’s nice.
I am more scared of powershell, it has OOP.
Same lol, but some people swear by it
Sorta related, i found a copilot pull request (merged) on an npm project repo and I was so disgusted I wanted to delete the dependency and do my own thing manually, but alas this was at work and my time on this earth is finite (for now)
(I just wanted to share this)
my time on this earth is finite (for now)
Let me know when you get that immortality hack working!
You are the first I will call :)
All week I’ve been sending these to my coworker, currently hip deep in AI propaganda, at the Microsoft Power Platform conference in Vegas.
I was recently pulling my hair out because a recent update caused a bug in the Windows recovery media creator that prevented the Windows RE from recognizing USB input devices. The few weeks between the faulty update and the fix just happened to coincide with me needing to use it.
WTF are they doing messing with something as basic as the Recovery Environment? And further, in a way that breaks its utility entirely?
And no regression tests that caught it
They probably don’t even test
They vibed their way out of the tests
they ask copilot if the code works
Will windows run ok? Yes? Good.
Captive audience on the contracts side, so they can do whatever they want as crappy as they want and the contracts still generate revenue.
You wanted a usable product? Stay away from Big Tech anymore.
Marketing discovered AI can code (terribly).
The version of windows 10 my Lenovo laptop can’t see wifi. I don’t like factory resetting that machine (actually I can’t anymore since I wiped the recovery partition when I moved it to Linux) my framework laptop came with no OS so was immune to bad factory OS version problems
How the hell do you fuck up the task manager, an application that hasn’t needed to be touched in decades? At best, you stick a new skin over it to match the aesthetic. The core functionality of the task manager should have remained untouched going back to… What, Windows 3.0?
I really don’t agree with this comment seeing how useful task manager has gotten over the years. It used to be pretty simple process explorer, but has evolved into almost a full fledged resource monitor. The only thing that feels like missing now is afterburner-like overlay in games.
The only thing that feels like missing now is afterburner-like overlay in games.
There are a thousand applications that provide such an overlay, one is already built into Windows, do you really need yet another?
And temperature?
And fps… Frametimes… Recordable graphs… Power… Voltages… Amps… Task manager is a baby compared to serious monitoring apps. But the main purpose is not monitoring ig, it’s fine as it is.
It has temps for GPU I know that
Never really needed that since I built my own rig, do maintenance and trust my build. Never had problems with temps, even during prolonged stresstesting
Okay, but literally all of that is reason to need temperature monitoring?
Only during the finish phase to make sure your system stays optimal during full load. No need to do that during actual use afterwards.
Think of it like servicing the car vs using the car
When I used to be on Windows, I shifted to Process Explorer. It is developed by Microsoft only I guess as part of their Sysinternals suite. I think it retains an older style UI but is significantly more powerful (has/d virus total integration for one).
Correction: Sysinternals Process Explorer was developed by Sysinternals long before MS bought them.
Because they vibe code the shit out of everything now. Insane shit is bound to happen.
…hail Hydra?
That’s not vibe coding… Its vibe testing…
Both? Both. Both is good.
Well no human looked at it while coding, no human looked at it while testing, so should be good as long as no user is a human looking at it…
what does vibe coding mean?
Using AI to generate code without fully understanding what the code is doing.
what does that have to do with vibing. thats just AI-Generated code. you might as well use Base 44
Because you’re coding “based on vibes,” not logical thought.
They’re ruining LinkedIn and GitHub as well.
I’m not sure it was possible to ruin something like LinkedIn …
No way… Are you serious?
Stares in disbelief
Hugs htop
Personal anecdote, but I’ve had Microsoft apps like Outlook and Teams crash on me on 4 different days this week while at work. Is anyone else getting similar instability issues?
Teams crashes or fails to work for me at least a few times a week and has for months. Outlook glitches out daily. I legit started using the web access instead of actual Outlook because it constantly bugged out.
Both Teams and Outlook are so ridiculously slow for what they do and the hardware they are running on.
Meanwhile in Windows 11: 4 years after release and I still can’t click on the clock on my secondary monitor to look at the calendar.
my windows 10 stability has been horrendous the past month or so
constant file explorer and office program crashes, random reboots and blue screens, lock ups. no consistent pattern to it that I can tell.
also getting weird glitches in teams that I’ve never seen before that require a force close and restart to make it usable again
That’s what happens when you start using LLMs for all of your software development. Garbage code all day long.
I haven’t had teams crash, just behaving badly, not marking chats read, not moving updated chats to the top.
