rhombus-brainstorming VS racket-multimethod

The Racket team has been embarking on the Rhombus project with something they're calling shrubbery notation. The most recent Racket Con (https://con.racket-lang.org/) has several presentations on the topic, and there's a GitHub repo (https://github.com/racket/rhombus-brainstorming) tracking discussion. For what it's worth, I strongly dislike the heavy use of : in this notation.

Once you become a skeptic, other warts began to appear such as the culture (1,2) of the language and one of the active academic members putting a lot of effort into changing the syntax. I concluded that I might as well spend my time just learning common lisp as then I might have a tool I could use. Everything is old in common lisp, and I am coming to recognize that as a good thing. Re-inventing the wheel is not always useful, and a few libraries that everyone uses means they probably function and build.

Matthew has been working on a [prototype](https://github.com/racket/rhombus-brainstorming/pull/163) that works on the current snapshot version of Racket.

It is possible to use to write some basic things and most existing Racket libraries are usable with the prototype.

It is called rhombus and there is an open discussion going on at https://github.com/racket/rhombus-brainstorming

rhombus-brainstorming

In my case the "because" was the lack of multi-methods. And in looking into that a little bit and finding generics and a package to extend them that was unmaintained, and finding a long discussion about multi-methods for porting the code for Functional Differential Geometry that was active for a bit and then dropped, and then looking again at some of the libraries that drew me in (like DrBayes) that was no longer maintained, I concluded that as an academic language there are a lot of elegant interesting packages that are one and done projects. The student graduates. The package rots.