blastmaster :vim:
Erik McClureIt's either very funny or very depressing to watch executives trip over themselves to prove who has the worst understanding of what software development actually entails.
interru 
@cloudhop
From the quoted screenshot:
Mistake not ...@interru @cloudhop I mean I can imagine that at some point I'd rather check diffs from the llm rather than typing them myself ... same way as I sometimes don't want to read compiler logs. But if the plan is to let people work in those mental regimes regularly, count me out. Because taking a nap fixes that a lot better.
FoolishOwl@interru @cloudhop Once again they don't understand that the purpose of a programming language is to allow unambiguous expression of intent.
But then, that's how ruling classes work. They don't do the hard work and they don't understand the hard work. Which takes more knowledge, growing crops with your own labor, or ordering soldiers to collect the grain at spear-point?
Krazov
Truffles
Schneckbert 🐌
Bruno Nicoletti
Pare
Manni
Alien software, human hardware@cloudhop you should have put a content warning around that, because i definitely just got dumber reading that.
Also, makes you roll your eyes so hard it could pull a muscle.
Also, makes you roll your eyes so hard it could pull a muscle.
Jernej Simončič �@cloudhop 20-30% of AI-generated code in Windows seems to be going really well:
https://www.theverge.com/news/867647/microsoft-windows-11-january-2026-update-bugs-issues
https://www.pcworld.com/article/3060768/windows-11-update-kb5077181-causes-startup-problems-heres-what-you-can-do.html
https://www.theverge.com/news/867647/microsoft-windows-11-january-2026-update-bugs-issues
https://www.pcworld.com/article/3060768/windows-11-update-kb5077181-causes-startup-problems-heres-what-you-can-do.html
Matthew Walton@jernej__s @cloudhop I'd be really interested in knowing how the bugs relate to areas with LLM-generated code in.
Matt Hardy 3.11 for Workgroups@cloudhop I used to buy magazines that had programs that you could type in. Think might have been the last time my programming was constrained by typing speed. The hard part has always been understanding the task at hand and formulating the best way to achieve it.
Captain of the SS El Faro@cloudhop as a composer I am no longer constrained by how fast I can play the piano
@cloudhop now the AI makes songs in thousands of BPM's, soon - millions of BPM's will be within reach, eventually the songs will become sentient and rule with an iron fist
🏴🏳⚧🏴☠
Jordan@coolcalmcollected (re: alt text: https://time.com/4979608/beethoven-trick-question/ )
Henk LangeveldNice to see the Sony vs Philips CD capacity conflict mentioned.
bencourtice@coolcalmcollected @bangskij @cloudhop I think we should ask the ladies from the typing pool to do all our software development, have you seen how fast they can type?
Wulfy—Speaker to the machines@bencourtice @coolcalmcollected @bangskij @cloudhop
I always suspected software developers were holding back human progress by being slow typists...
...they even talk to rubber duckies to infuriate C-execs!
Cat!@coolcalmcollected @bangskij @cloudhop
Everyone knows Beethoven's 9th Symphony takes 24 hours to perform.
Monica Rooney@coolcalmcollected @bangskij @cloudhop obviously less time because there are half as many people coming in and futzing around with chairs and stands and sheet music (assuming you include prep time)
@monicarooney @bangskij @cloudhop
excellent point. I wonder if the extra credit question was "which musicians, men or women, played faster and why?"
mathew
Sabrina Web 
📎@cloudhop typing speed? I'm speechless, do they even know what they're talking about?
ersatzmaus@sabrinaweb71 @cloudhop No, no. They don't know anything. We're not seeing true consciousness here, just an (admittedly surprisingly convincing) example of mimicry.
And then there are the LLMs…
Twenty Twenty Sixhorns
DressToKILT
refraction 
James 🦉 #FBPE 🇪🇺@cloudhop Show me the code printouts!
@cloudhop Soon, soon my pretties, CEOering will no longer be constrained by counting beans
Loke Mani
🇨🇦🇩🇪🇨🇳张殿李🇨🇳🇩🇪🇨🇦@cloudhop I've heard the claim from my SO that he gets about 12 lines of code in the final product per day. Is that true? If so, it's definitely not typing speed that restricts him since he types waaaaaaaaaaaaaaaaaay faster than me and I type waaaaaaaaaaaaaaaaaaay faster than most of my colleagues.
rustedivan
Barry Rowlingson@cloudhop Yeah, I remember all my software engineering skills I got from Mavis Beacon.
Tim Ward ⭐🇪🇺🔶 #FBPE@geospacedman @cloudhop I have often wished that I could restrict my hiring of software engineers to people who could actually touch type (they're more likely to write things like comments and documentation) but sadly that would have limited the available pool. We're talking about the days when typing lessons in schools were only offered to girls, and most software engineers were boys.
Oggie@cloudhop
I mean it's a pretty accurate understanding of programming measurements as 'lines of code' without ever once considering 'that compiles', and somehow people have used THAT as a metric!
(which is to say, this is hilariously awful, but they've made a lot of really dumb decisions before now too!)
Mad Engineering@cloudhop Software development is about as typing constrained as storytelling.
Abram Kedge🏴🇨🇦@cloudhop seriously... I spent far longer planning and designing a complex embedded system than actually coding it. Typing in the code is the easy part.
Giles Goat@AbramKedge @cloudhop My personal experience goes that coding often involves "quite some time" staring at code that is already there/thinking maybe even 2 ..3 days like that touching few lines at the time. Then you start getting "the new ideas" and could be a few days of "code this and that rinse and repeat" only finally you get "the moment" where maybe you can even write 5000 lines of code in a few hours where 4995 will be correct and 5 will take 2 weeks to debug 😅
@gilesgoat @cloudhop my cube was outside the VP of Engineering's office. For weeks I saw him quietly fuming as he walked past. Often I'd be sketching ideas on a whiteboard, or sitting back staring at it with my feet up on a filing cabinet. Four o'clock each afternoon I disappeared off to the war room to chat with the other three system architects.
Sometimes he saw me actually typing into a code editor. "How's it going?"
"Pretty good - I've got the data structures locked down, most of the function headers in place, just working on the state machine now."
"So no code yet?"
"Not yet."
The code worked the first time it was flashed into the fpga prototype, reading and writing data to a RAM disk. In three months from the start of the project, we were booting Windows from that prototype.
For comparison, the previous ground-up firmware project took 18 months to get to the same point. Code-first only *feels* faster.
@AbramKedge @cloudhop To me coding 'unless I start already with some developed idea in mind' of course always involves quite a bit of thinking/re-watching some code I already done. I tend to 'split a big problem into a set of smaller problems' and work/test them one by one before to attempt "the big merge". Sometime I quickly type things in the editor as 'they are quick ideas I want to test' that then after much rework can turn into real functional code. Erm do I see a brony here 😎 ?
@gilesgoat @cloudhop absolutely - especially when adapting or extending existing code. My process is very much the same as yours.
The scary part of that big project was that it was the frontend processor tightly bound to a hugely complex SAS interface hardware block - I tested what I could by simulation, but that was only about 10%!
@AbramKedge @cloudhop Have you ever found yourself "in the paradox" of having to also write "test programs" to test the code you are writing but then those too would need testing almost leading to 'an infinite recursion of debug' ? 😂 I really almost ALWAYS found the 90% 10% rule working .. 90% of what you wrote will be bug free and doing exactly what you wanted how you wanted .. BUT .. is the 10% that will consume 90% of the time to figure out what is wrong with it, usually few lines of code 😐
@gilesgoat @cloudhop I was frustrated by test-driven design purists who seemed to want to continually test whether the processor could add two numbers!
I tended not to write test programs - except where running the real program could corrupt real persistent data. Then I separated out all the "doing the work" code from the "writing the results" code, and made a parallel data-safe test version of the program.
Other than that, Debug builds of the code that added sanity checking on function parameters seemed to catch most errors.
@AbramKedge @gilesgoat I find that relying on spec driven or test driven development too early is useless when dependencies lie about their capabilities or are just broken. I prefer prototyping a design before writing any tests just so I can work with the libraries and get a better sense of what problems I might run into. I only write exhaustive tests after I have an architecture that has a working, functional end-to-end minimal example.
@cloudhop @AbramKedge Well I don't test 'minimal pieces' at all but at the present I am working on a set of cross platforms libraries ( APIs ) so I want to be sure that "if I call banana() with params P1" I get the same "external" identical result/behaviour in all the platforms. So I need "a test program" to call a number of functions in certain ways and verify "they produce really the same results" and believe me there's ALWAYS SOMETHING that falls through and/or needs to be adjusted 🙄
@gilesgoat @cloudhop That's a perfect use case, nice one 👍
Matt Lacey@cloudhop The number of times in 30+ years my development speed has been constrained by the speed of my fingers: 0.
Jules 🍺@cloudhop
I remember managers at a firm I worked for suggesting that the typists should enter the code to speed things up 😝
I remember managers at a firm I worked for suggesting that the typists should enter the code to speed things up 😝
🇺🇦 haxadecimal 🚫👑@julesbl @cloudhop
That was actually common practice in the 1960s and early-mid 1970s. The people who did the typing were called "keypunch operators". Programmers would hand-print their programs on coding forms.
It may have been the case that most programmers did not have typing skills, but that was not the primary force driving that method of computer usage, and it certainly did not make programming faster.
That was actually common practice in the 1960s and early-mid 1970s. The people who did the typing were called "keypunch operators". Programmers would hand-print their programs on coding forms.
It may have been the case that most programmers did not have typing skills, but that was not the primary force driving that method of computer usage, and it certainly did not make programming faster.
Timothy Wolodzko@cloudhop where did they get the idea that software development was constrained by typing speed? From Zootopia??
Ryan Harter@cloudhop I find this stuff more fun to read if you replace "AI" with "Steve".
scott
Exit Code 1@cloudhop This dude really wants to tell us that he doesn't code without telling us that he doesn't code.
Joris MeysRE: https://equestria.social/@cloudhop/116077882659405785
@cloudhop Equally funny (or depressing) to watch CEOs not seeing how "20-30% Microsoft code now written by AI" at the same time we see "security leak in Notepad" and "Windows 11 performance drop after last update" is not a recommendation for AI...
So can we tap the sign "when a measure becomes the goal, it stops being a measure" a few more times? One day they'll get it. One day...