On Wed, Jul 19, 2006 at 03:06:10AM +0900, Francis Cianfrocca wrote:
> >regarding ruby stuff being 'easy' to install i'd say that opening up the 
> >world of editing source and compiling it will certainly bring in patches and
> >contributions - and those may be steps towards making ruby extensions easier
> >to install be they binary or not.
> 
> This point was made upthread and is entirely valid. I'm not proposing 
> that people stop working with extensions in source code. But as Ruby 
> applications become more successful and acquire credibility for 
> production use, the lack of really seamless binary support in a range of 
> environments (including Windows flavors) becomes a barrier to adoption. 
> In a sense, I'm talking about enabling a community of users that today 
> have some good reasons to avoid Ruby.
> 
> You are right, though, this is orthogonal to the VC-MinGW choice and 
> belongs on its own thread.

I believe that it is not entirely orthogonal. My gut feeling is that MinGW
would make it easier to provide binaries, thanks to:
* the toolchain being closer to that you have on un*x: this means that
  third party libs are often easier to compile than with VC 2005. Ara cited
  gsl. As you said, in the long term not only the extensions, but also the
  libs themselves would have to be provided in binary form.
* the possibility to cross-compile (this is something I've repeatedly used myself)
* its availability in time (is VC 2005 to end like VC6?)

-- 
Mauricio Fernandez  -   http://eigenclass.org   -  singular Ruby