Idea: Market mogan/texmacs by creating an AI coding benchmark based on the C++ source

#1

For example, the benchmark could ask the AI to port 1000 methods in the mogan/texmacs C++ source to another language such as rust.

You could then evaluate the results by running mogan/texmacs with one ported method at a time to see if it seems to be computing the same thing as the original C++ method.

#2

Hi,
I’m not sure if this will work well.
Both Qt and Guile don’t seem to have a lot of example code available on the internet (it exists, of course, but not nearly on the same level as e.g. python, numpy or javascript). Furthermore, there are incompatibilities between different versions of Guile, so I’m not sure if current language models will be able to interpret/write code well and produce something that can be compiled.