Hi, I think that for any non-coding/boring task the best bet (barring a paid employee) is to crowdsource the effort. You could compare this to the task of enforcing code style: it would be a huge task to convert all NuttX files to follow the code style and most likely no one would ever offer to do it on their own. However, distributing the load to every PR author to adapt files they touch is a good way to make fair distribution of the load.
I think for IP clearance a CI check could be added by using the scripts I made: each file touched by a PR could be checked to see if it can be "Apachized" safely. If so, the PR author would have to include this change in the PR in a separate commit, with a commit message holding relevant output of the tool for tracing. If some files become problematic for this process, they can be added to an "ignored" list (or a "already converted" list). If that sounds reasonable, I could look into it. Of course, that is the slow approach and can only considered a help (to slowly reduce the problem and avoid continuing to increase it), so for something more critical as IP clearance I think it should be possible to further distribute batches of files to verify/clear to committers over some period of time. From our website, I count ~15 non-mentor commiters. If each of those would clear 50 files, that gets us 1500 files cleared. That is ~15% of .c/.h files. That could be repeated periodically (every couple of months, for example) and it would get us much closer to compliance, not counting the files cleared via the CI procedure detailed before. A final point I would make is this: for such a large codebase, this kind of non-coding tasks (for example, documentation) will require the biggest effort and thus require more help. In that regard, growing the number of committers is very important. A call for non-coders to help with CI and documentation may attract some users interested in NuttX but not confident yet with the codebase. Best, Matias PS: similar the the "can be Apachized" CI check, I feel we should add a "has the author signed an ICLA" check eventually for non-Apache licensed files. Sometimes I see PRs from authors who appear not to be commiters and I wonder if we're not making the problem worse by further introducing authors without ICLAs. On Wed, Oct 7, 2020, at 10:24, Gregory Nutt wrote: > > > I do not think there is zero progress as at least I’ve signed a SGA to > > change all the files from Xiaomi to Apache License. > Zero files have been officially cleared. That allowed us to change the > Xiaomi/Pinecone copyrighted headers to Apache headers, but we still have > no idea who all contributed to those files > > For the work on cleaning the license for other files, I could ask Peter and > > Xiang Xiao again to let them be more aggressive on this task. It is the > > long holiday in China, will ask them when they come back. > I think there are many of us who would participate. But there has to be > a key stakeholder and there has to be a dataset of contributor/copyright > information. If such a database were available, it would be easy for us > to clear individual files. > > But in general, I prefer that someone in the community promise to take the > > responsibility of this nasty task, not by an order from their manager. I > > know this is not an easy work, and much boring than writing code for > > engineers, but we need to do these type of works to drive the project > > forward. > I have the same preference. But we need to be honest: That is not > going to happen. > >
