Very good points. Solving problems is not necessarily what happens when they are identified.

The title is a bit too much of a blanket statement. Still there's indeed a lovely no between pair programming and merge requests. If possible you should favour the former. Yet it rarely happens in practice, there are reasons for that.

This is a good point. I feel unease at the current trend pushing toward cooldowns. The proposed rollout scheme is much better and fairer.

Nice suggestions on how to structure larger Rust code bases. The proposed error handling is particularly neat and tidy. This is doable in other languages but tends to be more verbose.

Ultimately, they just want people to stay on the pages they fully control and not have them visit anything out of their mall.

The responsibilities drop on people before they're ready for it (I see it first hand regularly at customers). Such tips are thus welcome and helpful during the transition.

Nice little facilitation formats. I'll try those for sure.

Knowledge management is hard. It's almost never a tool problem despite what people claim.

Time to retire std::function in new code.

Not sure it warranted the "dumb" mention in the title. Still it's likely a good idea to have a list of the ways projects can die.

fork() doesn't want to die. But help is coming it seems. Maybe the day it disappears from kernels is "near".

I agree with this short history tour. It's the composability which matters.

Really smart SIMD trick which packs a punch.

It's not complicated, and a good thing to do.

Good overview of why we don't see a speed up in development processes when AI tools are introduced. The bottlenecks don't magically get destroyed.

A proof that you don't need much to write a test suite.

We've seen a stream of those security issues lately. It says something about the security practice in the industry right now. Things need to be improved.

The GitHub exodus continues. Looks like Forgejo is really benefiting from it, I wonder how far this will go.

As usual with this author it feels a bit too much like advertising toward the end. Still this is an important post, it shows quite well why you can't limit yourself at only the language used when picking a stack. You definitely need to look at the standard library and the wider ecosystem as well. Rust is no different there and has its own issues.

Another type checker for Python gets stabilised. So many options and fragmentation in this space. This is odd.