On 26/04/07, Danek Duvall <danek.duvall at sun.com> wrote: > On Thu, Apr 26, 2007 at 10:02:40PM +0200, Roland Mainz wrote: > > > In theory I would agree. But there is much more wrong with /usr/bin/man > > - during the OSDevConf2007 in Berlin this year I've met with Michelle > > Olson and discussed some details and she got two hand-written DIN-A4 > > pages from me, full of tightly-written comments listing the issues about > > "what's wrong with the manual page subsystem". Fixing these issues alone > > is much work, I estimate 4-5years/enginee to hunt down and deal with > > these issues... > > But you don't have to deal with all those issues at once. You can deliver > the (pretty simple) improvement to man to make it derive MANPATH from PATH > if MANPATH doesn't already exist without having to spend the next five > years rewriting man from scratch, and have it in within the next build or > two. That is, if you wanted to. It took me about twenty minutes, given > that I hadn't looked at the man source before, to whip up a working > prototype. > > The only thing in my mind that makes this more than a trivial feature > enhancement is the possibility of considering this behavior an interface > that would need to be ARCed -- how would it evolve, if at all, once man has > taken all the changes you proposed in your two pages.
I have to agree with Danek's sentiment here. In my personal experience, my PATH and MANPATH are fairly close to being identical. I was always annoyed that I had to set both. To me, this perfectly fits the whole serendipitous discovery conversation that occurred recently. -- "Less is only more where more is no good." --Frank Lloyd Wright Shawn Walker, Software and Systems Analyst binarycrusader at gmail.com - http://binarycrusader.blogspot.com/
