Sad to see people predominantly jumping from Twitter to other tech moguls walled gardens. This feels more and more like a missed opportunity for the fediverse. That said I'm amazed at how efficient Musk has been at killing the network effect of his platform. This proves it's actually doable.
If you're just doing the minimum to deal with a task to "mark it done" you're probably not doing enough and missing out on learning opportunities.
Will we see more deployments of C++ standard library with bound checking by default? It definitely looks tempting.
Interesting tips to keep learning on the technical side of the job as you get more managerial responsibilities.
Looks like a nice tool. Maybe it'll replace my trusty cgdb in some cases.
Nice to see open hardware for VR hitting such a price point.
Seeing the amount of PHP code open on the internet, it's indeed important to harden the runtime (at long last).
More signs of the generative AI companies hitting a plateau...
Interesting reasoning about what is hard in systems with concurrency. It's definitely about the state space of the system and the structure of that space.
Good musing about complexity. Very often we need to move it around, the important question is where should it appear. For sure you don't want it scattered everywhere.
Now this is definitely a smart trick to estimate position in tunnels.
It shouldn't be, but it is a big deal. Having such training corpus openly available is one of the big missing pieces to build models.
Nice chain of attacks. This shows more than one vulnerability needs to be leveraged to lead to root access. This provides valuable lessons.
This is a nice view into how a query planner roughly works and a nice algorithm which can be used internally to properly estimate the number of distinct values in a column.
The title says it all. This is very fragmented and there are several options to fulfill the task. Knowing the tradeoffs can be handy.
Interesting musing on the heuristics we use when solving problems. There are good advices in there to make progress and become a better developer.
Avoiding boolean parameters in library APIs should be a well known advice by now. Still they should probably be avoided when modeling domain types as well.
Definitely a good post. No you don't have to go all in with cloud providers and signing with your blood. It's often much more expensive for little gain but much more complexity and vendor lock in.
This is an interesting and balanced view. Also nice to see that local inference is really getting closer. This is mostly a UI problem now.
Fascinating research about side-channel attacks. Learned a lot about them and website fingerprinting here. Also interesting the explanations of how the use of machine learning models can actually get in the way of proper understanding of the side-channel really used by an attack which can prevent developing actually useful counter-measures.