home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.unix.misc:3260 alt.religion.computers:471
- Path: sparky!uunet!cs.utexas.edu!swrinde!sdd.hp.com!wupost!news.miami.edu!cybernet!pineal.sci.fau.edu!spc
- From: spc@pineal.sci.fau.edu (Sean 'Captain Napalm' Conner)
- Newsgroups: comp.unix.misc,alt.religion.computers
- Subject: Re: The need for an assembler (was: Favorite CPU)
- Message-ID: <1992Aug17.185447.2408@cybernet.cse.fau.edu>
- Date: 17 Aug 92 18:54:47 GMT
- References: <id.FN9S.OC8@ferranti.com> <1992Aug12.173328.7160@cybernet.cse.fau.edu> <id.2PCS._X2@ferranti.com>
- Sender: news@cybernet.cse.fau.edu
- Organization: Florida Atlantic University, but who cares?
- Lines: 68
-
- (If this ends up in alt.folklore.computers, then I did something wrong 8-)
-
- In article <id.2PCS._X2@ferranti.com> peter@ferranti.com (Peter da Silva) writes:
- >In article <1992Aug12.173328.7160@cybernet.cse.fau.edu> spc@pineal.sci.fau.edu (Sean 'Captain Napalm' Conner) writes:
- >> -spc (Unix philosophy:
- >> 1. Write cryptic and highly customizable code.
- >
- >Write simple code. Write small modules that hang together well. Write your
- >code as generically as possible... one day you'll have to port it.
- >
- And find is simple? dd is small? Gee, I'd like to have your Unix system
- at work 8-) 8-) (add many many more if needed)
-
- >> 2. Never write original code. Find some that's close, and force it.
- >
- >I've seen some of that.
- >
- >> 3. Never write C. That's what Awk and Perl and Bourne Shell are for.
- >
- >If an interpreted language is fast enough, sure.
- >
- >> 4. Never write Assembly. That's what C is for.
- >
- >If a compiled language will do the job, sure.
- >I've got assembly modules in some of my current code, but it's pretty
- >hairy stuff.
- >
- >> 5. Never remove code. #ifdef it out.
- >
- >I don't hold with this. If I change the functionality of a routine, all
- >the ifdeffed stuff is either brought up to date or dyked out.
- >
- >> 6. Write sloppy code. The compiler will optimize for you.
- >
- >Write good algorithms before you worry about micro-optimisation.
- >
- >> 7. Never worry about memory. It's virtual anyway.
- >
- >Oh, this is the MIT/Berkeley philosophy. Don't check the return code of
- >malloc. Assume your network is fast and your workstations are slow.
- >
- Hmmmm, that's funny, because I always check the return code of
- malloc, my computer is fast, but the network is kinda slow ...
-
- >> 8. If it's not running Unix, it's not a real computer.
- >
- >If you can't run UNIX code on it, it's not a real computer.
- >
- Can you run UNIX code on the Amiga? You can run C code written
- with the Standard C Library on the Amiga, PC, Unix, etc, but not
- UNIX code, per say.
-
- A person I know doesn't consider the Amiga to be a REAL COMPUTER(TM)
- because it doesn't run Unix. I don't consider the 386 to be a real
- processor because it's a deranged 8/16/32 bit mutant CPU from 1980
- (well, based on a deranged CPU from 1980, I should say) 8-) 8-) 8-)
-
- >> (Personally, I can't stand the Unix philosophy 8-) 8-)
- >
- >I can't stand people confusing MITnix with UNIX.
-
- -spc (And I can't stand people that defend a 20 year old
- OS that's about as many patches, kludges, fixes and scotch
- tape as a certain 10 year old OS for PC's, but only
- 100 times larger 8-) 8-) 8-) 8-) 8-)
-
- P.S. Add a 8-) after EVERY word. Letter, if it would make you
- feel better.
-