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=0.2 required=5.0 tests=BAYES_00,INVALID_MSGID, REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 109fba,b87849933931bc93 X-Google-Attributes: gid109fba,public X-Google-Thread: fac41,b87849933931bc93 X-Google-Attributes: gidfac41,public X-Google-Thread: 1108a1,b87849933931bc93 X-Google-Attributes: gid1108a1,public X-Google-Thread: 103376,b87849933931bc93 X-Google-Attributes: gid103376,public X-Google-Thread: f43e6,b87849933931bc93 X-Google-Attributes: gidf43e6,public X-Google-Thread: 114809,b87849933931bc93 X-Google-Attributes: gid114809,public From: harry@matilda.alt.net.au (Harry Protoolis) Subject: Re: What is wrong with OO ? Date: 1996/12/09 Message-ID: #1/1 X-Deja-AN: 203068356 references: <5883q1$oae@news3.digex.net> organization: alt.computer pty ltd reply-to: harry@alt.net.au newsgroups: comp.lang.c++,comp.lang.smalltalk,comp.lang.eiffel,comp.lang.ada,comp.object,comp.software-eng Date: 1996-12-09T00:00:00+00:00 List-Id: On Fri, 06 Dec 1996 16:36:36 -0600, Robert C. Martin wrote: >In article , harry@alt.net.au wrote: > >> Despite being a 'self centered hacker' (I like that title :-)) I >> actually agree with you Elliot. You should at least sketch your >> architecture out after your analysis is complete, (where analysis == >> primary use cases), and design to it. > >Actually, I prefer to do much more than just "sketch it out", the architecture >should be very well defined, and very detailed. However, I also prefer to >begin producing the architecture well before all the analysis is complete. >And I prefer producing code that is conformant to the architecture - and the >analysis - long before the architecture is complete. > >In other words, I like to do them all concurrently. > >This doesn't mean that I *ever* code something that is not designed. >It also doesn't mean that I design something that has not been analyzed. >It's just that I don't wait for *all* the analysis to be completed before >I begin on designing the architecture. And I don't wait for the complete >architecture before I begin on the code. Sure, but I tend to do a overall sketch to give me a big picture before diving in to the detail at any point. I find this helps to size the problem up. One force at this point in a project lifecycle is the need to estimate effort. I usually use a preliminary analysis as a significant input to the estimation process, but you can rarely afford to do very detailed architectural work up front (prior to contract signing). As a result I find you begin the 'natural' process with at least a preliminary analysis. From that I tend to sketch architecture, then pick the hardest/highest risk and begin the iterative process of detailed analysis, architecture, design and implementation. The initial sketch helps to give some overall uniformity to the architecture without being a straitjacket to the process. H - Harry Protoolis alt.computer pty ltd harry@alt.net.au software development consultants