Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!pasteur!ames!killer!igloo!bhv
From: bhv@igloo.Scum.COM (Bronis Vidugiris)
Newsgroups: comp.sys.apollo
Subject: Re: Flame Wars
Message-ID: <1372@igloo.Scum.COM>
Date: 26 Apr 89 21:25:33 GMT
References: <8904242133.AA16384@umix.cc.umich.edu>
Reply-To: bhv@igloo.UUCP (Bronis Vidugiris)
Organization: igloo, Northbrook, IL
Lines: 42

In article <8904242133.AA16384@umix.cc.umich.edu> FERGUSON@TMASL.EXXON.COM writes:
]
]I've got an idea! Let's all start flaming on IBM, instead of Sun or Apollo!
]I hate IBM PC's minis, mainframes, and big companies (this one, at least)
]seem to be infested with them. The software, graphics, and general ease of use
]is lousy! With Apollos, Suns, Mac's, and even (what?!) Amigas, why would
]anyone think of IBM as a choice? So how come IBM is called Big Blue?
]
]Hope that distracted some attention...
]
]Scott
I've had the misfortune to have to use IBM mainframes, so I agree thoroughly
with Scott.  What do I dislike about IBM software?  Well, I could pick on
TSO and JCL - but mercifully, I have forgotton most of what I had to learn
about that OS, and that misbegotten son-of-an-assembler turned into a
'language'.  I might also mention the joys of EBCDIC, but I'd rather not.
It's really more fair to pick on IBM for their current offerings, such as
CMS.  Relatively speaking, (to TSO, that is) this is not a bad OS.  However,
it still lacks any concept of hierarchical directory structure, follows a
weird convention of seperating the components of filenames by spaces which
prevents any reasonable parsing of a filename EXCEPT by ALWAYS specifying
all three components (Filename FIletype and Disk).  There are, of course,
still mutiple 'formats' of files, so that if one gives a fortran program
a 'variable record' format file, one can get the highly explanatory error
message 'Premeature EOF on FT06F001 - buffer contents are - standard fixup
taken, execution continuing'.  And watching a relatively new user who has
attempted to use the system without thoroughly reading the manual is always
entertaining.  Eventually, the user will hit the 'break' key to interrupt
some long output - not realizing that this will put him into CP mode.  
The only error message available from CP is CP ?.  He probably won't realize
that hitting the break key again will get him out of this mode - so he'll
have to yell for help.
   Anyway, to try and maintain a small amount of relavance to this newsgroup,
I'll mention that the Apollo workstations are a much, much, much friendlier
environment than any IBM mainframe.  Now, if they would only fix the fact 
that allocating more than 270k - odd bytes of stack trashes the OS, things
would be just peachy :-).  [Or, maybe this has been fixed in SR10 - half
a :-) - it might really HAVE been fixed for all I know].  Well, at least 
(as near as I can tell) one can't crp on another node and trash IT'S os -
that's something, I guess.
-- 
bhv@igloo