Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.rmit.EDU.AU!news.unimelb.EDU.AU!munnari.OZ.AU!news.ecn.uoknor.edu!solace!nntp.uio.no!news.cais.net!bofh.dot!world1.bawave.com!newsfeed.internetmci.com!in1.uu.net!news.artisoft.com!usenet
From: Terry Lambert <terry@lambert.org>
Newsgroups: comp.unix.misc,comp.unix.bsd.misc
Subject: Re: How to delete files within C programs
Date: Sun, 12 May 1996 19:17:15 -0700
Organization: Me
Lines: 96
Message-ID: <31969BAB.24B87830@lambert.org>
References: <Oum-El-Kheir.Benkahla-3004961724540001@mac-ugm-3.imag.fr> <4mv7jj$fl7@innocence.interface-business.de> <4mvdoj$6e2@news.rhrz.uni-bonn.de> <3193FA77.7BC5F06A@lambert.org> <4n5e9p$33m@news.rhrz.uni-bonn.de>
NNTP-Posting-Host: hecate.artisoft.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 2.01 (X11; I; Linux 1.1.76 i486)
Xref: euryale.cc.adfa.oz.au comp.unix.misc:22668 comp.unix.bsd.misc:1025
Henry G. Juengst wrote:
] >There is no method of arranging software such that it is intuitive
] >for a user to interact with it.
]
] Right. But we can make it easier. It is not too much work to
] use readable identifiers with a hierachy instead of abbreviations.
] X11 and VMS show that (not perfectly).
With respect, the concept of a CLD (or in UNIX, getopt) is not
unique, nor does it do anything intuitive other than allowing
the user to transport training in command use from one command
to another.
This flattens the learning curve, but by no means makes it 0 slope.
And it is inot a feature unique to VMS (actually, DCL, if we are
going to avoid confusing programs with the OS proper).
] Of course, it is not trivial to compile unix specific code for another
] operation system. But the reverse way is also hard. Anyway, the list
] of running unix software on VMS systems is long enough to show that
] it is not impossible.
This is an architecture problem, and speaks to code portability,
which is a programmer issue... again, not an OS issue.
The point of most portable "UNIX specific code" is to be "POSIX
specific code" instead. And it will compile and run on any POSIX
compliant OS, provided that there was no use of POSIX extentions,
and provided that there was no use of ftell without an ungetc(getc())
to force the record pointer to advance to the next boundry, or any
of the hunder other "gotchas" of POSIX non-compliance in VMS that
happen to not be tested by the validation suite, even though they
are mandated by the standard.
This is an intrinic problem in the validation suite, or the
programmers who built the POSIX environment and relied on *only*
the validation suite (depending on your point of view).
] BTW, POSIX compatiblity is one of the arguments which is used in
] most discussions about the compatiblity of a unix operation system.
] But we also know that POSIX is not enough.
Yes, we know it is not enough. That is where programming practice
of implementing the "not enough" required functions in seperable
code modules so they can be replaced caomes into play.
] >VMS does not run on commodity hardware, much as DEC would like
] >economies of scale to kick in in the production of their
] >proprietary hardware line to allow them to twist the definition
] >of "commodity".
]
] I think most hardware is proprietary. This changes in combination
] with hardware. BSD is also running on VAXes. OSF/1 (now 'DEC unix')
] runs on Alphas.
x86 (Intel) hardware is the *only* non-commodity hardware. Apple,
IBM, and Motorolla are working on commoditizing the PPC, but they
are not there yet.
] >Protestations to the contrary will not change these facts, and
] >thus, from that perspective, VMS is *not* open. You can freely
] >choose a different perspective, but indoing so, realize that
] >you will be in the minority.
]
] Perhaps this also depends on the definition of "open".
This may be true, but I'm summarily throwing out non-useful
definitions. For instance, "Multimedia" does not mean that
a machine has a sound card and a CDROM drive. It implies
certain capabilities. "MUD"'s are not "cyberspace". It isn't
"cyberspace" until you can torture somone to death in it and
they actually die. Etc..
So what is required for the definition of "open" to make it a
useful qualifier, where degree of "openness" is a figure of
merit instead of a marketing bullet item?
] May be the situation on the market changes in somes years. I do
] not know. But do not forget what AT&T und the Berkeley University
] did with their unix lines.
What's that? I can't think of anything they mutually did to
their respective products...
Terry Lambert
terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.