Path: utzoo!utgpu!jarvis.csri.toronto.edu!qucdn!vmsa.technion.ac.il!Path-Lost
Newsgroups: comp.software-eng
Subject: Re: "Correct" IS UNDEFINED!
Message-ID: <17271@duke.cs.duke.edu>
From: crm@romeo.cs.duke.edu (Charlie Martin)
Date: 2 Feb 90 14:47:16 GMT
Reply-To: crm@romeo.UUCP (Charlie Martin)
Sender: news@duke.cs.duke.edu
References: <13211@cbnewsc.ATT.COM> <13212@cbnewsc.ATT.COM>
Organization: Duke University CS Dept.; Durham, NC
Keywords: proof, correctness, theory, balderdash
Lines: 25

In article <13212@cbnewsc.ATT.COM> lgm@cbnewsc.ATT.COM (lawrence.g.mayka,ihp,) w
rites:
>
>An addendum to my previous posting:
>
>The lion's share of telecom switching system development is not
>usually the construction of entirely new products, but the
>addition of new functionality to the existing product line.  Thus,
>the well-understood, textbook requirements have often already been
>met.  Developers of new, innovative features do not always have
>the luxury of firm, consistent requirements agreed upon throughout
>the industry.
>

I don't think I understand this.  I recognize that new features, not
extensions to new peripherals or additions of things like nail-up
connections to existing switches, nmay not have firm requirements at the
time they are proposed.  But do they not have to meet standards set
through Bellcore before they can be integrated into the network.  It was
certainly my impression that Northern Telecom/BNR had to do this.

Doesn't AT&T?

Should Mean Judge Green know about this?

Charlie Martin (crm@cs.duke.edu,mcnc!duke!crm)