Path: csiph.com!news.mixmin.net!eternal-september.org!reader01.eternal-september.org!.POSTED!not-for-mail From: Tim Rentsch Newsgroups: comp.lang.c++ Subject: Re: CHAR_BIT is not eight Date: Wed, 28 Dec 2022 20:49:33 -0800 Organization: A noiseless patient Spider Lines: 28 Message-ID: <86358yvmoy.fsf@linuxsc.com> References: <361b04cd-8be4-48c5-bc78-229f145eb343n@googlegroups.com> <86r0y3c8yg.fsf@linuxsc.com> <87mt8qkan0.fsf@nosuchdomain.example.com> <86359t63ie.fsf@linuxsc.com> <7f3d41f3-3fd5-4f66-9d7e-e5816f5f8933n@googlegroups.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Injection-Info: reader01.eternal-september.org; posting-host="77b286a689752ca270ac0d7d8e983965"; logging-data="254579"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX189FTF9r8BgaMOzQM4uKgYk3eaAisC1A8g=" User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.4 (gnu/linux) Cancel-Lock: sha1:wZmhHZXFW2jqYoZ39opYSnN1TWo= sha1:NRECoUuFuSLHmDddIonLEKKHFr4= Xref: csiph.com comp.lang.c++:88287 Tiib writes: > On Monday, 5 December 2022 at 21:43:08 UTC+2, Tim Rentsch wrote: > >> Testing can be used to show the presence of bugs, but never their >> absence. The same principle applies to determining whether code >> is platform agnostic. Showing that part of a program is platform >> agnostic can be done using formal methods and formal semantics, >> just like other kinds of formal verification. It can be more >> work to take into account the range of variation allowed by >> platform variability, but the principles involved are the same. > > Formal proof of complex system is usually impossible as general > solution is missing even to simple system of three point > masses (three body problem). All platforms are way more complex > than that ... yet made by fallible entities under time-to-market > pressure. >> [.. commenting on the level of hyperbole in the statement "Software >> without tests is just a work of literature." ..] > > That hyperbole seems to be in our laws. All jurisdictions that I know > of address copyrights of software as those of works of literature. > So software that has not been tested to be useful for something > is on general case just a work of literature written by author of it. You are either remarkably clueless or you do a very convincing imitation of being remarkably clueless.