Luke Mewburn <lukem@goanna.cs.rmit.edu.au> writes:

> Julian Assange writes:
> > My observations on this thread:
> >
> > blah. blah. blah. ordering [lets re-invent make]
> > blah. blah. blah. debugging [lets re-invent make]
> > blah. blah. blah. dependences [lets re-invent make]
> > blah. blah. blah. lists [lets re-invent make]
> > blah. blah. blah. pre/post variable over-rides [lets re-invent make]
> > blah. blah. blah. default actions [lets re-invent make]
> > blah. blah. blah. templates [lets re-invent make]
> > blah. blah. blah. variable substitution [lets re-invent make]
> > blah. blah. blah. directory overrides [lets re-invent make]
> >
> > pkgsrc shows make(1) can be succesfully retargeted to
> > strange new dependency domains. rc should gain the
> > same treatment.
> > [this is not to belittle the hard work luke, mrg et al
> >  have put in so far, but why re-invent something that
> >  already works so incredibly well in handling
> >  dependences, and actions in so many other domains?]
> 
> so provide a sample implementation.

There are people better qualified than me to do this. In
particular I'd like to hear the views of our primary
bsd.pkg.mk hackers. hubert? agc? christos? et al, what do
you think?

-- 
Stefan Kahrs in [Kah96] discusses the
   notion of completeness--programs which never go wrong can be
   type-checked--which complements Milner's notion of
   soundness--type-checked programs never go wrong [Mil78].