home *** CD-ROM | disk | FTP | other *** search
- Path: jhunix.hcf.jhu.edu!hbien
- From: hbien@jhunix.hcf.jhu.edu (Harold Bien)
- Newsgroups: comp.object,comp.lang.c++,comp.ai.alife,sci.comp-aided
- Subject: Re: [Q] Growing Objects
- Followup-To: comp.object,comp.lang.c++,comp.ai.alife,sci.comp-aided
- Date: 3 Feb 1996 09:55:30 GMT
- Organization: HCF - Johns Hopkins University, Baltimore, Maryland, USA
- Message-ID: <4evbei$55v@news.jhu.edu>
- References: <30FE8297.41C6@bme.jhu.edu> <RMARTIN.96Jan20093428@rcm.oma.com> <4ef3bo$mjo@news.jhu.edu> <310C2EE2.8171C34@alcyone.com>
- NNTP-Posting-Host: 128.220.2.5
- X-Newsreader: TIN [version 1.2 PL2]
-
- Erik Max Francis (max@alcyone.com) wrote:
- : Harold Bien wrote:
-
- : Hmm -- I never saw anyone say that your approach was incorrect.
-
- That's why I put it in quotes. I didn't mean literally
- incorrect. Instead, I meant that I felt it might be incorrect.
-
- : > The Run() function is the main function called and will
- : > dispatch to the changing RealRun() function. However, the Run() function
- : > doesn't really do much except call _OTHER_ subfuctions. Therefore,
- : > should I need new methods like RespondToLight() et al. I will only have
- : > to put them in the RealRun() function relevant to the state. Am I
- : > overlooking something? I have yet to implement anything, but am afraid
- : > of commiting myself to something which will not stand up to the test of
- : > time. Thanks for your respons.
-
- I understand the utility of abstract/virtual classes, but my
- problem is that the classes are held in a container class. If I were to
- morph a class, I must somehow squeeze it into the same space as the
- former class in order for the container class (which holds pointers) to
- maintain a valid pointer to the class.
-