Return to the
comp.compilers page.
Search the
comp.compilers archives again.
| Related articles |
|---|
| How should size of program grow with size of problem? jvn@fermi.clas.virginia.edu (Julian V. Noble) (1993-10-28) |
| interface reuse over code reuse sdm7g@elvis.med.virginia.edu (Steven D. Majewski) (1993-11-04) |
| Re: interface reuse over code reuse throop@aurw44.aur.alcatel.com (1993-11-05) |
| Re: interface reuse over code reuse carroll@hercules.cis.udel.edu (Mark C. Carroll) (1993-11-08) |
| Re: interface reuse over code reuse pardo@cs.washington.edu (1993-11-09) |
| Newsgroups: | comp.compilers |
| From: | pardo@cs.washington.edu (David Keppel) |
| Keywords: | OOP, design |
| Organization: | Computer Science & Engineering, U. of Washington, Seattle |
| References: | 93-10-136 93-11-046 |
| Date: | Tue, 9 Nov 1993 21:37:16 GMT |
pardo@cs.washington.edu writes:
>[Reusing implementations has a flaw: optimizations (in the reused
> code) for one case are often pessimizations for another case (the
> new user/client of the reused code).]
henry@zoo.toronto.edu (Henry Spencer) writes
>[Many argue you want to reuse *interfaces*.]
throop@aurw44.aur.alcatel.com writes:
>[I mostly use OO to reuse "generic", unoptimized implementations.]
One idea from Gregor Kiczales' paper (citation in a previous
comp.compilers posting) is that you want to use "whatever" metholdology to
reuse code; the code is one possible implementation of the interface,
possibly tuned, possibly not. Then you use a meta-interface to tune
(optimize) the implementation of the non-meta operations around the needs
of a particular user (client).
;-D on ( I never meta... joke I couldn't stand! ) Pardo
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.