From ... From: Erik Naggum Subject: Re: Problems with EQUALP hash tables and Arrays Date: 1999/05/28 Message-ID: <3136843901742945@naggum.no>#1/1 X-Deja-AN: 482936142 References: <374AF040.F0978582@mindspring.com> <1WU23.43$KM3.15733@burlma1-snr2> mail-copies-to: never Organization: Naggum Software; +47 8800 8879; http://www.naggum.no Newsgroups: comp.lang.lisp * Lieven Marchand | I think it would make more sense for the standard to demand of | implementors to document the expected performance of their implementation. this would mean that no implemntation would ever be conforming to the standard, and reasonable people would shrug off that requirement, because it is actually an unreasonable thing to request people to do. the reason for support channels back to a vendor is to be able to ask when you need to know. perhaps this is foreign to people who don't like commercial vendors in the programming language market, but the support I get from Franz Inc when I need it is tremendously valuable. | Alternatively, you end up with requirements that are so loose as to be | almost meaningless. The C++ standard demands certain performance bounds | in terms of O(f(n)) of the STL library components, but since the | implementation is not required to support objects or arrays above a | certain size any vendor can claim O(1) performance of everything by | choosing a suitably large constant. I don't think you know what the big-O notation means, but I'm sure the people who read the C++ standard don't either, so it's probably a useful requirement for that community. #:Erik -- @1999-07-22T00:37:33Z -- pi billion seconds since the turn of the century