home *** CD-ROM | disk | FTP | other *** search
- Path: gaia.ns.utk.edu!mbk
- From: mbk@caffeine.engr.utk.edu (Matt Kennel)
- Newsgroups: comp.object,comp.lang.c++,comp.software-eng
- Subject: Re: Design language?
- Followup-To: comp.object,comp.lang.c++,comp.software-eng
- Date: 13 Mar 1996 15:44:58 GMT
- Organization: University of Tennessee, Knoxville and Oak Ridge National Laboratory
- Message-ID: <4i6qhq$4e8@gaia.ns.utk.edu>
- References: <31437B51.27C3@bitmailer.net> <4i1j1h$b7i@uuneo.neosoft.com> <JSA.96Mar12132737@organon.com>
- Reply-To: kennel@msr.epm.ornl.gov
- NNTP-Posting-Host: caffeine.engr.utk.edu
- X-Newsreader: TIN [version 1.2 PL2]
-
- Jon S Anthony (jsa@organon.com) wrote:
- : In article <Do4GLH.5Hw@ngc.com> jasons@ngc.com (Jason D. Smith) writes:
-
- : > = >I would like to find information about object oriented design languages. I am
- : > = >interested in some language to describe an object oriented design (Booch,
- : > = >OMT, or something like that). Such a language would permit to describe a
- : > = >system in an independent coding language manner an then generate the real
- : > = >code with some tool.
- : >
- : > Isn't this (roughly) where Eiffel came from?
-
- : No. Not even close.
-
- Right. And as soon as you make this language independent design
- langauge, some dweeb (picture Dilbert's boss) is going to ask,
-
- "I want a design-language-independent design-design-language
- to which generates the actual language-independent design-language
- code!!"
-
- I think the reality is that once some ideas and technology get settled
- and reliable enough to be automated, they end up--and should end up--in
- the ordinary language.
-
- Everything else---all that stuff in the design books---is the fuzzy human
- judgement parts that we haven't learned how to automate or formally
- characterize. Yet.
-
- New langauge design seems like 'boring' meat-and-potatoes computer
- science but I think it still matters for this reason. Computer languages
- are the technological embodiment of research and human judgement.
-
- No doubt many things done by standard technology compilers today used to be
- only "rules of thumb" or "heuristics" or "code patterns" known to the
- experts.
-
- This is a good reason why you shouldn't be so unwilling to change
- langauges over time.
-
- : /Jon
- : --
- : Jon Anthony
- : Organon Motives, Inc.
- : 1 Williston Road, Suite 4
- : Belmont, MA 02178
-
- : 617.484.3383
- : jsa@organon.com
-
-
-