Date: 2001-10-12 12:28:30
--- In boost_at_y..., Ross Smith <ross.s_at_i...> wrote:
> David Abrahams wrote:
> > It's strange that you should say that. If you mean the rules that
> > the Jambase, you can replace any of these with your own
definition at any
> > time... do you mean rules like DEPENDS, NOTFILE, etc., which
form the core
> > Jam funtionality? Why would you want to override them?
> Sorry, it's been too long and I can't remember the details. I'll
> probably give it another try sometime, but I'm not in any hurry.
> Jam may well offer some advantages over make, but I find it
> imagine the advantages being sufficiently overwhelming to overcome
> enormous disadvantage -- from the point of view of anyone I
> my code to -- of nonstandard build tools.
> The only thing I'm using Boost in at the moment is a personal
> that's sufficiently large and long-term that I'm confident in
> Boost to be fairly mature and available in instant-install form
> RPM) before I'm ready to release. In Boost's current form, I would
> use it in any project intended for either open source or in-house
> distribution (i.e. anything involving giving other people my source)
> because installing and building Boost is currently far too
> to expect people who aren't interested in Boost itself to go
> The new build system is intended to fix that, of course. But the
> sort of problem -- nontrivial build process -- is the reason I won't
> consider using anything but make + autoconf for my own projects. I
> want potential users to react with, "You mean I have to install
> thing I've never heard of _too_?" (Distributing Jam along with my
> as Boost is intending, is more trouble than I'm prepared to go to.
> to mention being a fast route to the Unix equivalent of DLL hell.)
Well, from a Windows perspective, the Jam solution is lightyears
better then make + autoconf, since we don't have autoconf and though
we do have variants of make they are all a pain to maintain in
comparison to Jam. So, while I understand your opinion here, I don't
agree with it.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk