Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!cs.utexas.edu!uunet!visdc!jiii
From: jiii@visdc.UUCP (John E Van Deusen III)
Newsgroups: comp.unix.wizards
Subject: Re: Must UNIX be a memory hog?
Summary: Possible reasons for the /bin/true copyright; more efficient
	alternatives to the use of /bin/true & /bin/false.
Message-ID: <544@visdc.UUCP>
Date: 22 May 89 19:06:13 GMT
References: <159@zebra.UUCP> <1608@auspex.auspex.com> <2@minya.UUCP> <31529@bu-cs.BU.EDU>
Reply-To: jiii@visdc.UUCP (John E Van Deusen III)
Organization: VI Software Development, Boise, Idaho
Lines: 27

In article <31529@bu-cs.BU.EDU> bzs@bu-cs.BU.EDU (Barry Shein) writes:
>
> I wonder about the legal implications that /bin/true which contains
> nothing but a copyright notice ...

Since it is impossible to copyright ideas, (see COMPUTER SOFTWARE
PROTECTION by Thorne D Harris III, 1985, Prentice Hall, 0-13-528373-6),
I can see how AT&T might have been concerned about publishing a
significant, although small, part of the UNIX operating system that was
unequivocally unique.  A software company in Korea, trying to export an
operating system called YUNICS, might claim that although their code was
almost identical to UNIX, certain things can only be done in one way;
exibit A, /bin/true.

I am currently not inclined to use either /bin/true or /bin/false.  The
colon ':' has the same effect as true, but as part the the shell it
avoids the path search and file read operations.  Test(1) is usually a
part of the shell too.  Thus,

	test ""

should usually be more efficient than /bin/false.  Even if it were not
more efficient, it is at least as clear to me what is intended.
--
John E Van Deusen III, PO Box 9283, Boise, ID  83707, (208) 343-1865

uunet!visdc!jiii