I was not aware of it, but I am now! :) It seems like I have some research to do in looking at other tools. When I get another chance to work on muster (maybe not until August?) I'll have a look.

(FWIW, it seems like there is a fundamental tradeoff between accuracy and speed in these things.. and I'm not sure which direction to take things. Maybe the best tool would be one that is configurable, so you could tell it "be as accurate as possible, I don't care how slow you are" or "please sacrifice accuracy for speed, I want the results asap".)

An additional data point here: fabric8-analytics (the dependency analysis engine for openshift.io) recently switched to using ScanCode Toolkit as its license analysis backend: https://github.com/fabric8-analytics/fabric8-analytics-worker/pull/56

So if there are use cases where ScanCode isn't fast enough, effort would likely be most usefully invested in making it faster (e.g. with Cython, PyPy, or appropriate use of extension modules), or in allowing some aspects of the scanning process to be skipped.