If you only read one thing🚢 The dream of an alpine waterway (20 minute read) What a vision. Building a waterway across the alps! It failed, of course. But the vision was great. At the T3 Leadership Summit, I had a discussion about how to get engineers in the office. If you have a grand vision that is inspiring, you will get the best engineers, and to do something great, they’ll work in the office. If you are another mobile app with a database backend, not so. Work on your vision. Work on your engineering vision! (The waterway is an engineering vision). Most tech departments I look into lack a tech vision. Or have one which is outdated and never referenced. Have a great tech vision and you will attract great people. And do something that matters. Not another dating app. https://blog.nationalmuseum.ch/en/2024/06/the-dream-of-an-alpine-waterway/
Stories I’ve enjoyed this week🐞 Ladybird Web Browser becomes a non-profit with $1 Million from GitHub Founder (5 minute read) This is huge. It will change the web landscape. Ladybird is a new browser without using Googles HTML engine like most others do. People said it can’t be done, one developer creating a new browser without using Chromes rendering engine. But Andreas proved them wrong. Now he got $1M in financing. Firefox dropped the ball a long time ago; they were more interested in having a nice life with high salaries, instead of driving their browser to new heights. Firefox squandered more than $6.000.000.000 in money. There has been no real innovation in browsers the last ten years. Ladybird will take up the ball and bring back innovation. The IE6 browser wars will be back. If you like it or not. 🧠 MIT robotics pioneer Rodney Brooks thinks people are vastly overestimating generative AI (22 minute read) I do think people overestimate AI in the short run, and vastly underestimate AI in the 10 to 20 years time frame. AI will blow everything away. dolt fetch Gets Much Faster (15 minute read) Dolt is a database, and it is getting faster by using async co-routines. I recently wrote a small framework (ha!) for Inkmi that renders parts of a page in different go-routines, with the nice side effect of Go parking IO go-routines automatically. Nice. Which, in the end, reduces service times for a request. As shown in one of my last newsletters, this does decrease latency a lot and makes for a much better (faster) user experience. Not enough people leverage async page rendering, or async parallel work in general. Do your devs? https://www.dolthub.com/blog/2024-05-08-dolt-new-puller/ Einstein and his peers were ‘irrationally resistant’ to black holes. This illustrated story explores why (10 minute read) Once a physics professor at university told me, people don’t change their mind. Science only changes because these people die of old age. One physicist said about the proposal of black wholes, *“Various accidents may intervene to save the star [..] but I want more protection than that. I think there should be a law of Nature to prevent a star from behaving in this absurd way!” Engineers are sometimes like this—they have problems accepting new realities. This was when we moved from manual memory management with malloc/free to garbage collection in Java. And it happened when we moved from command hierarchies (“Tell me what to do!”) to Agile. And it will happen with AI. A Eulogy for DevOps (28 minute read) A very long text on DevOps. Did it fail or not? Coming from a time with silos and huge gaps between operations and development, I would say it succeeded. Even if no one knows what DevOps means (like Agile). Developers today deploy on their own, and infrastructure as code is the standard. Development changed for the better. Even perhaps if DevOps never achieved, what it was set out to do. The movement is gone and has folded into mainstream-as they always do. Join the CTO newsletter! | |
