Interesting, it looks like index scans in your databases can have surprising performance results with SSDs.

The idea is interesting, I wouldn't throw away level based logging but this could complete it nicely.

Long but interesting chapter which shows how GPUs architecture works and the differences with TPUs. This is unsurprisingly written in the context of large models training.

Easy to misunderstand as an elitist stance... But it's not the way I read it. Churning more code faster isn't going to help us, you need to take the time for people to grow and improve. It's not possible to achieve if you're drowning in eager beginners.

Interesting parable, it's indeed a good way to illustrate different leadership styles. Being more strategic is clearly what one should try to do.

A good way to frame the possible models for your organization regarding remote work. The GitLab Handbook stays a very good resource regarding remote work, they really thought about it and documented their findings.

I think this is still one of the best distilled explanation of product ownership. It's also interesting for the other parties on an agile project.

A good reminder that long hours are not a sign of success with your project... on the contrary.

No good tricks to optimize your code, but knowing the tooling knobs sometimes help.

Let's see if this gets merged. This could be interesting convenience.

OK, this is completely useless but definitely a fun project.

Or why the XML roots of the web are important to keep in shape. I'm not necessarily in love with how verbose XML is, but it's been a great enabler for interoperability. That's indeed the latter reason which pushed Google to try to get rid of it as much as possible.

A bit of a long read, but does a good job explaining the use of assertions and how to introduce them in your organization.

A good list of things to consider when designing systems. And indeed in case of success the result looks probably boring.

There's a need for clearer vocabulary about testing indeed. The write up is a bit dry here but that's a start.

Indeed, CPU prefetchers are really good nowadays. Now you know what to do to keep your code fast.

Looking forward to Git LFS going away indeed.

This latest ruling from the German supreme court is rather worrying...

There's a big "if" of course, don't just throw your tests out of the window. But indeed, they need to bring value... so start by having really good tests.

That's an interesting trick to make sure people reevaluate comments when they remove some code. Doesn't work for every language of course.