News for hackers. Stuff that matters.
Re:ModelingToolkit.jl, an IR and Compiler for Scientific Models by marmaduke on Thursday May 16, 2019 @04:06PM (#19932566)
as a research engineer in data science/modeling, posts like this just look too good to be true. there's no free lunch. what is the catch with all this stuff? who is going to maintain all this when Chris Rackauckas gets plucked by MathWorks or Wolfram?

(this is a compliment! it's great stuff)

Re:ModelingToolkit.jl, an IR and Compiler for Scientific Models by ChrisRackauckas on Saturday May 18, 2019 @04:33PM (#19949583)
If you're looking for a catch, I guess the issue is that the modeling domain is relatively simple, restricted to "mathematical models", i.e. no conditional-based control flow like while loops. Like systems like Jax, for loops are supported by simply unrolling the expression into something without control flow. This lets the resulting IR be relatively simple and then all of the math is easy to define, but of course means the domain for this is "math" and not "programs". That said, it's still large enough for things like Modelica models, which is a large enough domain for a lot of the algorithms I want to explore like automatic index reduction of DAEs.