> Date: Mon, 22 Oct 2018 22:28:35 +0900 > From: Rin Okuyama <rokuy...@rk.phys.keio.ac.jp> > > I'm really disappointed by this. I do not like the way you got things > going (and also I really do not like to write a message like this). > > I asked you to discuss further before rashly committing it. Didn't I? > People actually provided some useful proposals to improve your draft. > However, instead of finding a point of agreement, you shutted down > discussion one-sidedly: > > https://mail-index.netbsd.org/tech-toolchain/2018/10/10/msg003307.html
Hi, Rin! There are some good suggestions in that thread, and parts of the thread became a bit of a bike shed. We don't have a formal review process for evolving changes before merging them into HEAD (though we may be moving toward one with hg). Perhaps it would have been better to see another iteration or two of drafts, but (a) bike shed discussions can be long and tedious and can discourage contributions, and (b) we can always revise README.md as a living document -- no need to revert altogether when we can just edit it to improve it. So, what would you like to see improved in the document? - We should definitely mention the relation of the mirror on Github to the master version in CVS. - We can pick a standard style for the capitalization of example architecture names, or pick a different selection of them that doesn't make capitalization difficult, or change all of them. - We can tweak the suggested build.sh incantation, and expand on what the parts of it mean. Personally I use `../obj.amd64' instead of `~/obj'.