The next big thing for WebAssembly, today’s ultimate run-time: beside compiling languages to WebAssembly and run it side-by-side with JavaScript, thanks to WasmEdge you can now run JavaScript even inside the WebAssembly run-time! secondstate.io/articles/run-j…

Stunning infographic from South China Morning Post: “a world of languages — and how many speak them” scmp.com/infographics/a…

Yesterday, 2022-11-17, my long-time "Project Fifty" was successfully completed after half a century — with a celebration, attended by my family, friends and most of my colleagues. Many thanks to all who were there or at least sent congratulations. I love you.

A milestone in the Unix history, where nearly everything is based on the C programming language for 5 decades: the Linux kernel in version 6.1 finally added support for the Rust programming language. Now kernel modules can be written in Rust instead of C! lwn.net/Articles/91076…

For those who still think ChatGPT is really intelligent, just check out Susam Pal’s “I Asked ChatGPT To Explain Some Jokes to Me“. The experiment at least shows that you can have fun with ChatGPT, as long as do not expect too much from it… susam.net/maze/chatgpt-e…

In “The Myth of the Myth of the 10x Programmer” you can read that “10x developers[…] absolutely do exist. You’re just not going to find them with a coding test.” I personally fully agree: developer productivity is unrelated to bare coding skills. payne.org/blog/the-myth-…

Images and voice are the domains where AI still shines most: Microsoft unveils VALL-E, a new AI-based Text to Speech (TTS) synthesizer which can replicate speech with only a 3-second enrolled recording of an unseen speaker. valle-demo.github.io

Well written and illustrated and a must-read for anybody doing any kind of visual designs, including designing websites and apps: “Visual design rules you can safely follow every time” anthonyhobday.com/sideprojects/s…

Watch a studio recording of my presentation “Multimedia Didactics for Software Engineering”, which I originally gave at the SE2023 conference in Paderborn, DE. The presentation language is german. youtu.be/L3FINQT82gQ se-2023.gi.de

Worth to check out: Amazon Prime Video team published the surprising information that they moved from microservices to monolith. Adrian Cockroft analysed and cleared it up: they did “Serverless First” and just refactored some parts. primevideotech.com/video-streamin… adrianco.medium.com/so-many-bad-ta…

Awesome new AI application on images: “Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifold” vcai.mpi-inf.mpg.de/projects/DragG… arxiv.org/pdf/2305.10973…

Today I held the keynote at Munich Cyber Tactics, Techniques and Procedures! To better support me on stage, I’ve beforehand ad-hoc developed a special Open Source prompter for PowerPoint: PPTX-Prompter github.com/rse/pptx-promp… It now allows me to use dedicated PowerPoint slides(!) as my speaker notes(!)…

[german] Gestern habe ich den “Balzert-Preis für Digitale Didaktik 2023” von der Gesellschaft für Informatik (GI) für meine “Multimediale Didaktik für Software Engineering” verliehen bekommen. Eine zugehörigen Website multimediale-didaktik.de mit allen Details ging soeben live.

Interesting to read: Blair Fix’s article ”The Dunning-Kruger Effect is Autocorrelation”. There is no systemic bias of unskilled people to over-estimate their skill, but more skilled people are just more accurate at assessing their skill. economicsfromthetopdown.com/2022/04/08/the…

I would like security experts to understand that in practice the equation is “security = 1 / experience” and hence if you always just seek for ultimate security, nobody wants to — or even can — use it. So, security is not a boolean 0-OR-1 decision, it’s a gradual 0-TO-1 decision.

After 39 years of programming experience, I can only agree: "The first 90 percent of the code accounts for the first 90 percent of the development time. The remaining 10 percent of the code accounts for the OTHER 90 percent of the development time." — Tom Cargill, Bell Labs

I get tired on the AI hype. Hopefully we soon get into a time where people really start to understand that Large Language Models (LLM) of this GenAI era are just clever token auto-completers and do NOT understand anything, even if their answers are always looking convincing.

My current favourite locally running Generative AI building blocks: Mistral 7B LLM, executed on the awesome Ollama runtime and accessed through its new OpenAI API. The only essential thing now still missing is direct (non-WSL) Windows support in Ollama... ollama.com

On YouTube youtu.be/aSCpr17MPQU you can now find an interview (sorry, in German language only) with me, due to my new role as CTO of the msg Group. This interview is intended to allow people to better get to know and assess me personally.

One of the most difficult virtues in IT is adequacy. For this one must understand the problem well enough and reflect on reason for the solution. Most people find both very difficult. — me