kimrhh
@kimrhh.bsky.social
He/Him. Sailing 🤩 DevOps with passion and opinions. All things agile. proud father of two crazy teen girls. Love the people part of SW dev.
Reposted by kimrhh
Det ene øjeblik er Nigel Farage fra Temu så vred og forurettet over Martin Lidegaards udlægning af DF’s åbenlyst fremmedfjendske udspil, at han vil hive ham i retten. Så vred, at han laver en video.
… Hvorefter DF render ud med den ene løgn efter den anden 🫠
Imponerende amatørisme.
#dkpol
… Hvorefter DF render ud med den ene løgn efter den anden 🫠
Imponerende amatørisme.
#dkpol
November 16, 2025 at 5:55 PM
Det ene øjeblik er Nigel Farage fra Temu så vred og forurettet over Martin Lidegaards udlægning af DF’s åbenlyst fremmedfjendske udspil, at han vil hive ham i retten. Så vred, at han laver en video.
… Hvorefter DF render ud med den ene løgn efter den anden 🫠
Imponerende amatørisme.
#dkpol
… Hvorefter DF render ud med den ene løgn efter den anden 🫠
Imponerende amatørisme.
#dkpol
Reposted by kimrhh
I do not want to “Do my own research.” I want to pay trained accredited experts to do the research, trained accredited experts to report on findings, and trained accredited experts to make policy based on them.
I am busy af with my own job, I don’t need to do eight zillion more.
I am busy af with my own job, I don’t need to do eight zillion more.
November 15, 2025 at 1:11 PM
I do not want to “Do my own research.” I want to pay trained accredited experts to do the research, trained accredited experts to report on findings, and trained accredited experts to make policy based on them.
I am busy af with my own job, I don’t need to do eight zillion more.
I am busy af with my own job, I don’t need to do eight zillion more.
Reposted by kimrhh
German court finds Open AI infringed copyright by using song lyrics for LLM training. (This news via @publisherslunch.bsky.social , which notes that this is the first court case to end in an infringement ruling against Open AI) www.theguardian.com/technology/2...
ChatGPT violated copyright law by ‘learning’ from song lyrics, German court rules
OpenAI ordered to pay undisclosed damages for training its language models on artists’ work without permission
www.theguardian.com
November 13, 2025 at 4:43 PM
German court finds Open AI infringed copyright by using song lyrics for LLM training. (This news via @publisherslunch.bsky.social , which notes that this is the first court case to end in an infringement ruling against Open AI) www.theguardian.com/technology/2...
Reposted by kimrhh
Disagreements are not problems. They are opportunities to learn. For example, when I have to prove something that seems intuitively correct to me, I have to drill deeply into things like hard numbers.
1/5
1/5
November 9, 2025 at 7:31 PM
Disagreements are not problems. They are opportunities to learn. For example, when I have to prove something that seems intuitively correct to me, I have to drill deeply into things like hard numbers.
1/5
1/5
Reposted by kimrhh
Why The Best Developers Are Obsessed With Vertical Slices
Read it on Medium:
https://medium.devleader.ca/why-the-best-developers-are-obsessed-with-vertical-slices-86926acbf734
Read it on Medium:
https://medium.devleader.ca/why-the-best-developers-are-obsessed-with-vertical-slices-86926acbf734
Why The Best Developers Are Obsessed With Vertical Slices
In the realm of software development, the way we structure and deliver our work can significantly impact the project’s success. Enter the…
medium.devleader.ca
November 1, 2025 at 9:00 AM
Why The Best Developers Are Obsessed With Vertical Slices
Read it on Medium:
https://medium.devleader.ca/why-the-best-developers-are-obsessed-with-vertical-slices-86926acbf734
Read it on Medium:
https://medium.devleader.ca/why-the-best-developers-are-obsessed-with-vertical-slices-86926acbf734
Reposted by kimrhh
Small steps and fast feedback cycles? Specifying with examples? Continuous design, testing, code review, refactoring and integration? Autonomous, self-organising teams?
It's all starting to sound a little familiar...
codemanship.wordpress.com/2025/11/01/t...
It's all starting to sound a little familiar...
codemanship.wordpress.com/2025/11/01/t...
The AI-Ready Software Developer #16 – A Token of Our eXtreme
Some of you reading the posts in this series might be thinking, “This all sounds a bit familiar”. And you’d be right. “AI” coding assistants built around Large Languag…
codemanship.wordpress.com
November 1, 2025 at 9:03 AM
Small steps and fast feedback cycles? Specifying with examples? Continuous design, testing, code review, refactoring and integration? Autonomous, self-organising teams?
It's all starting to sound a little familiar...
codemanship.wordpress.com/2025/11/01/t...
It's all starting to sound a little familiar...
codemanship.wordpress.com/2025/11/01/t...
Reposted by kimrhh
Jeg har mine hvalplakater op og håber ikke, jeg har misforstået noget.
October 26, 2025 at 9:03 AM
Jeg har mine hvalplakater op og håber ikke, jeg har misforstået noget.
Reposted by kimrhh
Reposted by kimrhh
I hope this email never finds you. I hope you’re free.
October 23, 2025 at 3:41 PM
I hope this email never finds you. I hope you’re free.
Reposted by kimrhh
"investigators believed that three or four thieves were involved and that the robbery bore the hallmarks of an experienced team of veteran criminals given its precision and speed."
www.nytimes.com/2025/10/19/w...
www.nytimes.com/2025/10/19/w...
October 19, 2025 at 6:38 PM
"investigators believed that three or four thieves were involved and that the robbery bore the hallmarks of an experienced team of veteran criminals given its precision and speed."
www.nytimes.com/2025/10/19/w...
www.nytimes.com/2025/10/19/w...
Reposted by kimrhh
I just found out that Finland teaches kids in kindergarten to fact check, and actively teaches children how to spot fake news.
Why isn't this standard everywhere! The best time for other countries to follow suit was at the same time. The second best is now.
www.helsinkitimes.fi/themes/theme...
Why isn't this standard everywhere! The best time for other countries to follow suit was at the same time. The second best is now.
www.helsinkitimes.fi/themes/theme...
Finland trains six-year-olds to spot fake news – and they might be better at it than you
Finnish schools have made media literacy a key part of education, equipping even six-year-olds with the skills to spot fake news and online manipulation.
Finland has consistently ranked as Europe’s m...
www.helsinkitimes.fi
October 16, 2025 at 5:56 AM
I just found out that Finland teaches kids in kindergarten to fact check, and actively teaches children how to spot fake news.
Why isn't this standard everywhere! The best time for other countries to follow suit was at the same time. The second best is now.
www.helsinkitimes.fi/themes/theme...
Why isn't this standard everywhere! The best time for other countries to follow suit was at the same time. The second best is now.
www.helsinkitimes.fi/themes/theme...
Reposted by kimrhh
Reposted by kimrhh
From the tech bro gospel: “someone was going to coarsen society and further immiserate people, it may as well be me.”
October 10, 2025 at 4:07 PM
From the tech bro gospel: “someone was going to coarsen society and further immiserate people, it may as well be me.”
Reposted by kimrhh
alan cooper (who is hiding out on mastodon) published this in 1999:
October 4, 2025 at 10:54 PM
alan cooper (who is hiding out on mastodon) published this in 1999:
Reposted by kimrhh
as a girl with a PhD in natural language processing and machine learning it's actually offensive to me when you say "we don't know how LLMs work so they might be conscious"
I didn't spend 10 years in mines of academia to be told ignorance is morally equal knowledge.
We know exactly how LLMs work.
I didn't spend 10 years in mines of academia to be told ignorance is morally equal knowledge.
We know exactly how LLMs work.
October 4, 2025 at 10:23 PM
as a girl with a PhD in natural language processing and machine learning it's actually offensive to me when you say "we don't know how LLMs work so they might be conscious"
I didn't spend 10 years in mines of academia to be told ignorance is morally equal knowledge.
We know exactly how LLMs work.
I didn't spend 10 years in mines of academia to be told ignorance is morally equal knowledge.
We know exactly how LLMs work.
Reposted by kimrhh
🚀 𝗔𝗜 𝘄𝗼𝗻’𝘁 𝗸𝗶𝗹𝗹 𝗰𝗼𝗱𝗶𝗻𝗴—𝗵𝗲𝗿𝗲’𝘀 𝘄𝗵𝘆 🚀
I got a thoughtful reply on one of my posts: we’re in a transition phase where LLMs cover 80% of the work, and once the last 20% is solved, we’ll just be writing apps in natural language instead of code.
It’s a smart take—and it’s also something I hear a lot.
I got a thoughtful reply on one of my posts: we’re in a transition phase where LLMs cover 80% of the work, and once the last 20% is solved, we’ll just be writing apps in natural language instead of code.
It’s a smart take—and it’s also something I hear a lot.
October 2, 2025 at 4:40 PM
🚀 𝗔𝗜 𝘄𝗼𝗻’𝘁 𝗸𝗶𝗹𝗹 𝗰𝗼𝗱𝗶𝗻𝗴—𝗵𝗲𝗿𝗲’𝘀 𝘄𝗵𝘆 🚀
I got a thoughtful reply on one of my posts: we’re in a transition phase where LLMs cover 80% of the work, and once the last 20% is solved, we’ll just be writing apps in natural language instead of code.
It’s a smart take—and it’s also something I hear a lot.
I got a thoughtful reply on one of my posts: we’re in a transition phase where LLMs cover 80% of the work, and once the last 20% is solved, we’ll just be writing apps in natural language instead of code.
It’s a smart take—and it’s also something I hear a lot.
Reposted by kimrhh
Well don't I feel stupid
September 26, 2025 at 11:02 PM
Well don't I feel stupid
Reposted by kimrhh
If you ever entertain by accident the idea that the best technology wins, remember that a resource-oriented protocol is right now the most successful RPC transport, thanks to a spaghetti templating language and Frankenstein scripting one made in 2 weeks.
And we did alright.
And we did alright.
September 27, 2025 at 8:39 AM
If you ever entertain by accident the idea that the best technology wins, remember that a resource-oriented protocol is right now the most successful RPC transport, thanks to a spaghetti templating language and Frankenstein scripting one made in 2 weeks.
And we did alright.
And we did alright.
Reposted by kimrhh
It took *hours* to install and update Windows 11. The installation and patching of Linux Ubuntu was done in less than 30 minutes on the @frame.work notebook! 👌
September 25, 2025 at 3:49 PM
It took *hours* to install and update Windows 11. The installation and patching of Linux Ubuntu was done in less than 30 minutes on the @frame.work notebook! 👌
Reposted by kimrhh
Reposted by kimrhh
Borgerforslag om ikke at stemme for chatcontrol finder I her. Jeg kender 0 (nul) forskere/eksperter i datalogi og computer, som synes at det forslag er godt. Zero. Et er snageriet, noget andet er at det slet slet ikke skalerer. Ignorant populisme. www.borgerforslag.dk/se-og-stoet-...
September 21, 2025 at 5:09 PM
Borgerforslag om ikke at stemme for chatcontrol finder I her. Jeg kender 0 (nul) forskere/eksperter i datalogi og computer, som synes at det forslag er godt. Zero. Et er snageriet, noget andet er at det slet slet ikke skalerer. Ignorant populisme. www.borgerforslag.dk/se-og-stoet-...
Reposted by kimrhh
Justitsminister Peter Hummelgard (S) og de andre socialdemokrater i #dkpol og #eupol er fulde af lort mht. #chatcontrol og regeringens to andre partier, Venstre og Moderaterne, ved det og vil ikke røre Chat Control med en ildtang. Stem IKKE på @socialdemokratiet.dk.
📄 www.dr.dk/nyheder/udla...
📄 www.dr.dk/nyheder/udla...
Omstridt lovforslag splitter regeringens Europa-parlamentarikere: 'Vores sidste rest af privatliv online forsvinder'
Kritikere mener, man går alt for langt i jagten på at komme delingen af overgrebsmateriale til livs.
www.dr.dk
September 21, 2025 at 4:20 PM
Justitsminister Peter Hummelgard (S) og de andre socialdemokrater i #dkpol og #eupol er fulde af lort mht. #chatcontrol og regeringens to andre partier, Venstre og Moderaterne, ved det og vil ikke røre Chat Control med en ildtang. Stem IKKE på @socialdemokratiet.dk.
📄 www.dr.dk/nyheder/udla...
📄 www.dr.dk/nyheder/udla...
Reposted by kimrhh
Was this not… what critical AI researchers have been saying this entire time?
"In a landmark study, OpenAI researchers reveal that large language models will always produce plausible but false outputs, even with perfect data, due to fundamental statistical and computational limits."
www.computerworld.com/article/4059...
www.computerworld.com/article/4059...
OpenAI admits AI hallucinations are mathematically inevitable, not just engineering flaws
In a landmark study, OpenAI researchers reveal that large language models will always produce plausible but false outputs, even with perfect data, due to fundamental statistical and computational limi...
www.computerworld.com
September 21, 2025 at 4:56 PM
Was this not… what critical AI researchers have been saying this entire time?
Reposted by kimrhh
Appeasing fascists somehow never works out the way corporate interests think it will.
Memos: Amazon, which employs the most H-1B holders at 14K+, and Microsoft advise H-1B staff to avoid travel and return to the US before 12:01am ET on Sept. 21 (Yun Li/CNBC)
Main Link | Techmeme Permalink
Main Link | Techmeme Permalink
September 20, 2025 at 5:28 PM
Appeasing fascists somehow never works out the way corporate interests think it will.
Reposted by kimrhh
