I don't really have any strong opinions about this, but I agree with Gabe's arguments. Another problem with the current approach is determining how the authors list should be updated when moving code between files.
We recently had a related discussion about copyright notices and whether the git log could be reasonably used to find copyright holders. Some of the lawyers in the discussion argued the repo history (the git metadata) and the source code are distinct entities and that many distribution forms would lack the necessary data to track copyright holders. It's not uncommon to lose revision history, for example, GitHub offers a 'download as zip' function, and many grad students tend to check in their base version of gem5 as a single commit without history. Tracking for kudos/blame is slightly different though since copyright has an important legal function. Jason's solution sounds a lot better than manually tracking authors. It wouldn't be too hard to write a script that automatically maintains an AUTHORS file and lists contributors (potentially sorted by commits or lines of code) per directory / keyword. That would convey the same information as the authors field. However, I'm not convinced we need this in the repo to begin with. The MAINTAINERS file already lists subsystem experts, so we don't need it to track exports. To track kudos, wouldn't it make more sense to mine the repo nightly and publish the results on the new website instead? Cheers, Andreas On 18/01/2020 18:09, Jason Lowe-Power wrote: > This would be OK with me. However, I think it should be coupled with us > writing good release notes with each release which gives the code > contributors credit for their contributions. > > IIRC, the purpose of the "author" field was to know who to call when things > broke. I agree that this purpose has now been subsumed by git blame. > > On the other hand, I think it also served as the de facto way of giving > people credit for their contributions. We should be sure to find another > way to do this, too (e.g., release notes). > > Cheers, > Jason > > On Fri, Jan 17, 2020 at 6:00 PM Gabe Black <gabebl...@google.com> wrote: > >> Hi folks. I would like to suggest dropping authors from the header comments >> in the various files in gem5. That field is not really maintained, and it's >> not clear even when it should be updated. The history is a much more >> accurate and complete record of who did what. >> >> What do people think about that? >> >> Gabe >> _______________________________________________ >> gem5-dev mailing list >> gem5-dev@gem5.org >> http://m5sim.org/mailman/listinfo/gem5-dev > _______________________________________________ > gem5-dev mailing list > gem5-dev@gem5.org > http://m5sim.org/mailman/listinfo/gem5-dev IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you. IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you. _______________________________________________ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev
