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=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,821f293084877cad X-Google-Attributes: gid103376,public From: st92j0gw@dunx1.ocs.drexel.edu (Chris Papademetrious) Subject: Re: Adding a "Project Types" layer to a project Date: 1996/05/06 Message-ID: <4mlolb$g27@noc2.drexel.edu>#1/1 X-Deja-AN: 153355606 references: <3186E51C.250@post.drexel.edu> <31886BD1.51E@comm.hq.af.mil> organization: Drexel University newsgroups: comp.lang.ada Date: 1996-05-06T00:00:00+00:00 List-Id: Thank you everyone for your replies! I think I was a bit over-eager to use this "project types" concept. Chad is definitely right, for the project I'm working on, it's better to just use the vectors package if I need vectors functionality. I was trying to offer as much flexibility and indirection as I could for the groundwork, but it appears to be superfluous now that I look at this instance. - Chris Chad Bremmon wrote: >We have to keep in mind the "object oriented way." Supposedly, the data and the function are >together. What you have shown with your software problem is that by taking the vector functions and >putting them in another package, they are not available when you use the vector type later on. If >you are using vectors, use the vectors package. The indirection provided by the types package seems >to confuse the problem. >I'm not saying that a "standard" types package is not a good thing. There are times when it may >make more sense that you might consider using a types package. In those instances, you may consider >limiting the types package to be within the subsystem you are using. >cb