Re: [boost] Technical session at the October C++ meeting?

From: John Max Skaller (skaller_at_[hidden])
Date: 2001-08-20 17:07:34

jk_at_[hidden] wrote:
> 
> 17 Aug 2001 18:50:35 +0400 Beman Dawes wrote:
> >meet common needs without extensive parameterization. Someday a highly
> >configurable smart pointer may be invented that is also very easy to use
> 
> If the current variant will be standardized, whether it begins an obstacle
> for improvements?  Whether will leave it an opportunity for further
> optimization of smart_ptr<>, for example, specializations for an intrusive
> reference counting, allocation of non-intrusive reference counter through
> boost::pool and so on?
> 
> Current size fits many, but not all.

        That's enough. Smart pointers have too many variants.
What about polymorphic cloning? The list goes on and on,
for every category of types, there is a category of smart
pointers. In the end, the correct solution is a garbage collector,
since they're completely transparent, and generally faster
than any smart pointer. So we don't need a perfect smart
pointer, because we already have it: a plain pointer
is the perfect kind of pointer.

-- 
John (Max) Skaller, mailto:skaller_at_[hidden] 
10/1 Toxteth Rd Glebe NSW 2037 Australia voice: 61-2-9660-0850
New generation programming language Felix  http://felix.sourceforge.net
Literate Programming tool Interscript     
http://Interscript.sourceforge.net