Diego Biurrun wrote:
It is bad. Why would you want to run make recursively in the first place?
Usually since you partition your problem better and usually because you just depend on getting the subdir built in order to build what is in the parent dir. Having .mk or .inc doing the same makes this issue pointless, beside the fact is nice doing "cd foodir ; make".
It's worth noting that Miller's paper was fairly well known 10 years ago, and I believe the use of recursive make has only increased since then. So he apparently didn't convince a lot of people.
It is not quite as well-known as it should be, yes. However, this excuses nothing. People have simply gotten accustomed to running make clean or distclean all the time because they have come to distrust all the broken make systems. It's a real shame.
Most of the people are accustomed to autotools that got that right (probably just that), many of the people that got the Miller's paper just put it in the "Considered Harmful" basket (is usually round and full of other trash if you are wondering which is it), some found something good in it and applied in real life case with good results to a point. Having a jihad open just because somebody found the paper useful and made a better build system from practices suggested by it isn't the smartest thing to do. Thinking that non-recursive is the holy and only way seems as stupid to me. Backing statements by hard numbers seems too obvious and un-creative than using flawed logic rhetoric? lu -- Luca Barbato Gentoo Council Member Gentoo/linux Gentoo/PPC http://dev.gentoo.org/~lu_zero