|
Boost : |
From: Braden McDaniel (braden_at_[hidden])
Date: 2002-08-07 13:57:31
On Wed, 07 Aug 2002 08:39:21 -0400, Eric Woodruff wrote:
> "There is a very compelling reason to not include any pthreads or any other
> headers. If anyone would like to write even the most minimalist
> platform-neutral application using boost and multithreading, they should be
> able to get a precompiled boost library and link it in with their project
> without any other header dependencies. One should not require pthreads or
> win32sdk installed on their system if they have a statically build STLport
> and boost. The idea is to address the impotency of the standard library
> itself, by providing something that is autonomous and complete.
>
> Every time an abstraction layer is made, it helps remove more and more
> dependencies from a project, only adding more simplicity and
> maintainability.
>
> I think the majority of users would not want to enable
> #BOOST_POLLUTE_HEADER_AND_NAME_SPACE"
>
>
> I'd like to hear a good argument against this... because without libraries
> like threading, boost is made of only templates or other compiled-in
> utilities, which make that aspect of boost an autonomous development
> framework already--this is is exactly what the developer needs and efforts
> should be made to keep it that way, no matter what is added to it.
I've already heard a few good arguments here, but let me try to add to the
list: the scenario you describe strikes me as pretty contrived. Why
*wouldn't* a Win32 developer have the Win32 SDK? Or a POSIX developer
pthreads?
"Dependencies are bad" seems to be an unstated assumption underlying your
argument. Well, dependencies are *not* inherently bad. Specifically, I
don't see any problem with depending on things that one can reasonably
expect to be present in the compilation environment.
Braden
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk