What even is error handling?
I philosophize about error handling, what it actually means and how to characterize Go’s approach to it.
I philosophize about error handling, what it actually means and how to characterize Go’s approach to it.
I come up with a couple of useless, but entertaining ways to generate entropy without relying on any packages.
When talking about static type systems, we often tend to focus on one side of the equation. I’m trying to make explicit how I view the question as a tradeoff and why I neither agree with “more is always better”, nor with “a little is enough”.
In light of recent discussions about its removal, I try to discuss what kinds of problems context.Value tries to solve. I then try to describe a design which would address most (but not all) of the criticism surrounding it.
Logging in Go is a notoriously lacking topic in the standard library. There are 3rd-party libraries trying to work around this. I’m trying to explain, why I find them still lacking
I take a look at the pattern of optional interfaces in Go: what they are used for, why they are bad and what we can do about it.
Trying to provide some advice on how to do easy, readable, scalable routing in go, without relying on any muxers/routers (or writing your own).
There are next to no “backwards compatible API changes” in go. You should explicitely name your compatibility-guarantees.
I did lazy evaluation in go.
A short list of four things that I might want to add to go (but probably wouldn’t).