home *** CD-ROM | disk | FTP | other *** search
- Path: newshost.netinfo.com.au!usenet
- From: Alan Brain <aebrain@dynamite.com.au>
- Newsgroups: comp.lang.ada,comp.lang.c++,comp.lang.c
- Subject: Re: C++ vs Ada for large project
- Date: 25 Feb 1996 04:06:32 GMT
- Organization: Netinfo Pty Ltd - Canberra Australia
- Message-ID: <4gon88$gqu@fred.netinfo.com.au>
- References: <w4wx5wc1a2.fsf@cln46ac> <4ffjrq$i8k@qualcomm.com> <w4bun3a7ji.fsf@cln46ac> <4fsm6l$36j@dawn.mmm.com> <w4u40t3rc3.fsf@cln46ac>
- NNTP-Posting-Host: dialup5.dynamite.com.au
- Mime-Version: 1.0
- Content-Type: text/plain; charset=us-ascii
- Content-Transfer-Encoding: 7bit
- X-Mailer: Mozilla 1.2N (Windows; I; 16bit)
-
-
- >>>>>> "Kevin" == Kevin J Hopps <kjhopps@mmm.com> writes:
-
-
- > Kevin> So let me see if understand you. Less flexibility is better; rules
- > Kevin> are better than discipline; the more rules the better?
-
- (Sarcasm mode On)
- No, The fewer rules the better, discipline is better than rules, and the more
- flexibility the better. That's why compilers should never reject any code, there's no
- such thing as a compilation error, all programmers always write exactly what they mean,
- and no compiler-writer should ever dare think otherwise.
- (Sarcasm Mode Off)
-
- In summary, what you said sarcastically is often true. Make the language smart, so the
- poor benighted programmer can concentrate on the difficult tasks, rather than making
- the trivially easy so prone to error (hence difficult). I'd far rather be concentrating
- my limited time on deciding which AI method to use on the problem domain, than trying
- to trace the one mis-casting that's causing a coredump in 1.5 million LOCs.
-
-
-