Re: [post-1.2] proposed api extension mechanism ["Lou D. Langholtz" ] (fwd)

[Randy Terbush <ra...@zyzzyva.com>]

Jim echos my thoughts as well, however I am never apposed to 
learning. It seems hard to pick up any new language today that does 
not require an OO thought pattern.

Ben has pointed out in all of this that the guts of this would need 
to be done by someone who knows C++. In theory, that should make 
life easier for anyone trying to use that part of the code. I agree.


> Ben Laurie wrote:
> > 
> > I'm concerned that this discussion is based on vapourware - the assertion that
> > C++ isn't going to work on most platforms. This is not my experience at all,
> > but we really need some hard data. Perhaps we could get together some kind of
> > minimal test suite that we can ask people to try, and actually find out what
> > the real deal is?
> > 
> 
> A test suite would work, but it would most likely be something so
> basic that it may not prove anything. After all, C works on most
> platforms but some require some hoops and weird things to work.
> I would guess that based solely on _that_, and that C++ isn't
> as pervasive as C, the situation would hardly improve.
> 
> For me, the switch to C to C++ (or any real OO language) requires
> me to switch my mental gears; to program well in OO, you have to
> think OO. It may be that way for many, so we would wind up with
> C++ that's really C using some very basic OO concepts (ie: not really
> using the full power of C++). Add into the mix the fact that unless
> we update all modules, decide that all future modules must be C++
> and decide that all "appropriated" code (like sfio or pthreads) to
> be used in 2.0 be "ported" to C++, this mental-switching-of-gears
> will be going on a lot.
> 
> -- 
> ====================================================================
>       Jim Jagielski            |       jaguNET Access Services
>      jim@jaguNET.com           |       http://www.jaguNET.com/
>                   "Not the Craw... the CRAW!"