I think my point is that I no longer not want to wait until Test2/ is thoroughly comprehensible and documented. (But to be fair, I think it is quite comprehensible as is.) I want to stop using Test/ now, at least stop using it routinely. I.e., I want to: * rename Test/ to Legacy_Test/ * rename Test2/ to Test/ * remove Test2 from Jenkins’ processing * remove Test2 from instructions for updating p5subset * consider how best to attack all the things Joey listed In that order. If there are others besides Hugh who also want to see Test/ just deleted (rather than renamed to Legacy_Test/), I can live with that. It is not my favorite option, because I was not actually suggesting we stop using at at all, just that we take it out of the routine build process. E.g., I think it would make working on Test/ easier if the old system were readily available as Legacy_Test/. But it is really not a big deal, whereas fighting with it every time you want to update p5subset or change a description is. ________________________________ If we’re talking about moving Test to a legacy directory, my vote would be to kill it instead. It will only get more and more out of sync over time. It’ll be useless inside 6 months. I think we should do what Martin says and make Test2 thoroughly comprehensible and documented and then we can get rid of Test.