Revision history [back]

Another strategy you should consider is contributing to Sage itself! Sage's goals entail supporting all kinds of specialized research. Sage already includes code for elliptic curves and Steenrod algebra, both of which I think are intended only for working mathematicians. There are probably others too. So are there parts of your library that make sense to include into Sage?

For the remaining parts, which are maybe idiosyncratic, or which you don't want to be public, or which aren't ready for inclusion into Sage, I guess something like what you have is the best I could come up with.

A minor improvement might be to have the top cell update a local copy of the code repository from the server, so that you have all the up-to-date .py source on the local machine and can use the Python module system. But maybe you've already thought of reasons this doesn't work so well.

As an aside, it sounds like somewhere early in your workflow you're converting .sage code to .py code; are you using sage -preparse for this?