From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,FREEMAIL_FROM autolearn=ham autolearn_force=no version=3.4.4 X-Google-Thread: 103376,e276c1ed16429c03 X-Google-NewGroupId: yes X-Google-Attributes: gida07f3367d7,domainid0,public,usenet X-Google-Language: ENGLISH,UTF8 Path: g2news1.google.com!news4.google.com!feeder.news-service.com!94.75.214.39.MISMATCH!aioe.org!not-for-mail From: =?utf-8?Q?Yannick_Duch=C3=AAne_=28Hibou57?= =?utf-8?Q?=29?= Newsgroups: comp.lang.ada Subject: Re: Ada is getting more popular! Date: Sun, 31 Oct 2010 15:39:24 +0100 Organization: Ada @ Home Message-ID: References: <4cc60705$0$23764$14726298@news.sunsite.dk> <4cc6753c$0$23756$14726298@news.sunsite.dk> <4cc71e08$0$23758$14726298@news.sunsite.dk> <4cc87d7a$0$23755$14726298@news.sunsite.dk> <4cc912e1$0$23761$14726298@news.sunsite.dk> <10g3vmx6lxyu.bjfg1i5gjcsf$.dlg@40tude.net> NNTP-Posting-Host: gHYtk+mhCrTAX6LNXybaBQ.user.speranza.aioe.org Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed; delsp=yes Content-Transfer-Encoding: Quoted-Printable X-Complaints-To: abuse@aioe.org X-Notice: Filtered by postfilter v. 0.8.2 User-Agent: Opera Mail/10.63 (Win32) Xref: g2news1.google.com comp.lang.ada:15045 Date: 2010-10-31T15:39:24+01:00 List-Id: Le Sun, 31 Oct 2010 11:54:22 +0100, Dmitry A. Kazakov = a =C3=A9crit: > 5) you want to postpone some design decision/implementation to a later= > point. So you describe/implement what you know for certain right now, = in > the form of an abstract type, and let a derived type to define the res= t > when more information becomes available. Makes me feel like the implementation language is used for modeling (thi= s = topic again). I use to attempt that in the past (as I had a feeling for a need), with = = Pascal, but never felt satisfied. The implementation language is mostly = = not well suited for that, because it is intended to express things which= = works (even what is named =E2=80=9Cabstraction=E2=80=9D in an implementa= tion language, = requires concrete interfaces). You have, not only to complete and fill t= he = holes, but also have to change what was the prior abstraction (concrete = = interface again). Then, the prior abstraction is not readable anymore an= d = is no more there to guide your steps as the concrete implementation goes= . = Finally, the abstraction is not recognizable at all, the completion of t= he = implementation tries to follow something which is vanishing as it try to= = reach it (the abstraction) and goes erratic (more or less, will not say = = this produce a pure disaster). So I gave up with this, and tried another= = way : try to see the whole before starting any implementation (not the e= nd = of the story, as this is not more a piece of cake). To be less formal and to talk about feeling, I also did not enjoyed = incomplete implementations, which sooner or later had the taste of a = pending failure (especially as time passed), this was driving me = disappointed. Now I make a wish : never start an implementation without a good model := ) = (a model which will stay there as the implementation goes). The place to= = write down one's first ideas is really the modal (I believe), not the = source. So I may add this one among the other things inheritance/derivation is n= ot = to be confused with : modeling (unless running Simula or SmallTalk, from= = which languages like Pascal and Ada are rather far away). -- = Si les chats miaulent et font autant de vocalises bizarres, c=E2=80=99es= t pas pour = les chiens.