Here are some links to tools of cool things and projects that I think is great.
Programming
Graphiti
A different approach to achieve what GraphQL wants to fix without giving up REST and the benefits it brings. It is definitely worth reading their reasoning why they built, what they built.
Duckietown
It is not only a town that consists of Rubberduckies, but also a learning opportunity for AI and self driving cars. And if the duckies are off the clock, you can use them to talk your problems through. I haven’t really tried it, but for what I read is it quite well aligns with things I want to do and learn as well, just with my Lego Boost Robot. It definitely looks like fun.
Music
Sonic Pi
Why not make (electronic) music using life ruby programming and teach people about programming and music on the way? That is what Sonic Pi stands for and it is fun too.
Startups
Ready to share your startup with the world?
A list of resources where to share and propagate your startup idea. Might be useful for somebody.
Blog
Geekring
Through the commercializing of the search engines it is harder to find new interesting blogs that do not buy search results or do not sell stuff that you might find on google and there like. But luckily there are some projects out there, that allow small blog owners to promote their blogs to small audiences that do not know to search for your blog and still might find it useful.
System
Leverage Points: Places to Intervene in a System
A very interesting article about points of interest in a system. It also sorts them in order of effectiveness and gives great examples what a system and its leverage points are. The author concludes with saying that mastery might not be “pushing leverage points” instead it might just be “strategically […] letting go”.
Great Articles
Artificial Intelligence / Machine Learning
Why Are We Using Black Box Models in AI When We Don’t Need To? A Lesson From An Explainable AI Competition
Nice read, that questions the use of creating Neural Networks – which you cannot reason about – when there could easily be written algorithms that you actually can reason about.