Path: utzoo!mnetor!uunet!steinmetz!davidsen
From: davidsen@steinmetz.steinmetz.ge.com (William E. Davidsen Jr)
Newsgroups: comp.lang.c
Subject: Re: Why NULL is 0
Message-ID: <10283@steinmetz.steinmetz.ge.com>
Date: 6 Apr 88 14:12:32 GMT
References: <2550049@hpisod2.HP.COM> <7412@brl-smoke.ARPA> <3351@chinet.UUCP> <10229@steinmetz.steinmetz.ge.com> <7792@alice.UUCP>
Reply-To: davidsen@kbsvax.steinmetz.UUCP (William E. Davidsen Jr)
Organization: General Electric CRD, Schenectady, NY
Lines: 36

In article <7792@alice.UUCP> ark@alice.UUCP writes:
}In article <10229@steinmetz.steinmetz.ge.com>, davidsen@steinmetz.UUCP writes:
}
}> People (usually ;>) have no problem with the idea the while assigning a
}> zero to a float gives it the vaule 0.0, in most implementations the
					      ^^^^
	Foot in mouth time... I think "some" would be more correct.
}> float value does not have all bits set to zero.
}
}I am overwhelmed by curiosity.  Can you give me examples of
}three machine architectures in which the floating-point value
}0 does not have all its bits set to zero?

No. I can tell you two series of systems which have non-zero 0.0
(Honeywell 6000 and DPS), two which I'm told have non-zero 0.0 (mid 70's
IBM minis and Data General). My reasoning is as follows:

  given:	there are machines which do have C and don't have
		flat 0 expressed as all bits zero,

  and		the proposed C standard doesn't say that any machine
		running C must have float 0 be all bits zero,

  and		I have no reason to think that innovation and
		standards have stopped evolving,

  therefore:	I conclude that assuming float 0 to be all bits
		zero, or any other machine independent integer value,
		is non-portable, now and in the future.

  In light of the previous discussion, I was discussing "really
portable" as opposed to "portable to many machines."
-- 
	bill davidsen		(wedu@ge-crd.arpa)
  {uunet | philabs | seismo}!steinmetz!crdos1!davidsen
"Stupidity, like virtue, is its own reward" -me