On Thu, Oct 24, 2013 at 08:40:46PM -0400, James K. Lowden wrote: > I think Ken Thomson may cast the evil eye on you.
I'll take that risk :-) > I would keep usd/smm/psd. It seems to me that the user, the > programmer, and the administrator are different roles, even if they are > sometimes the same person. User is section 1, programmer is 2, 3, & 4 > (and sometimes 9), admin is 5 and 8. I would say that taxonomy has > held up pretty well. Yes it has; that's why I'm proposing to adopt it here as well. I think it's superior to usd/smm/psd as well as being more consistent. Also we'd need at least one more category (ksd?) for kernel programming docs. And then, is e.g. the ffs article a SMM or a KSD? One can get around that question by placing it in section 5, or possibly section 4; either of those avoids the issue. > I would *really* like to see only the sources distributed, with a > Makefile at the top of the hierarchy that produces the preferred output > format. I realize it's traditional, but I think it's ill-advised, outdated, and not suitable for what NetBSD ought to be. The whole point of what I've been doing is to move away from that model. > Current groff produces pdf directly. Only if you have perl available. > > usd/19.memacros reference/ref1/roff/meref usd/20.meref > > I dislike calling reference documents "ref". That one is not my fault, it came that way. I would not be averse to changing it to just "me", but I don't really want to start renaming articles (particularly the roff ones) until we're sure we have all the ones we want going forward. Quite a few of the historic roff-related articles are missing from the tree right now. -- David A. Holland dholl...@netbsd.org
