I’ve used juli a few times over the years. For some numerical things it’s nice with easy interfaces to modern algorithms. I solved a differential equation rece

I’ve used juli a few times over the years. For some numerical things it’s nice with easy interfaces to modern algorithms. I solved a differential equation recently and it just worked. And Julia feels like a much more proper programming language than matlab. At the lower level (which I haven’t looked at in a while so may have changed) I found it a bit confusing and messy. The subtyping and method selection are tricky to get right and they are fundamental to important parts of the language like it’s numeric tower. But libraries seem to just work. Macros were horrific and the ast is inscrutable and liable to change from one version to the next. Quasiquoting was also tricky. So I wouldn’t recommend trying to do anything weird with them. But maybe they are good now. Pluto notebooks seem a great concept. I tried them recently and mostly they worked (sometimes they didn’t get dependencies right but it’s still beta). I felt like I was fighting a bit with plots.jl. I don’t know if there are things that weren’t obvious to me that I was missing or if it can just be a bit annoying. I haven’t tried gadfly but I would like to at some point. I’ve heard good things about ggplot2 which it is inspired by. I felt like documentation was a bit lacking in good straightforward tutorials and examples. As well as documentation in general. But I don’t want to put too much emphasis on that.