Date:         Fri, 17 Dec 1999 01:12:21 GMT
Reply-To:     Melvin Klassen <Klassen@UVIC.CA>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Melvin Klassen <Klassen@UVIC.CA>
Organization: University of Victoria
Subject:      Re: SAS IPO?
Content-Type: text/plain; charset="us-ascii"

On Thu, 16 Dec 1999 21:41:53, Howard_Schreier@ITA.DOC.GOV (Howard
Schreier) wrote:

> Parts of SAS might be open-source. The most likely candidates are statistical procedures.
> One scenario is that at some point research results derived with "black box"
> (non-open-source) software might not be accepted, or might suffer reduced credibility.
> That would more or less compel SAS and other vendors to go open-source.

Really?  I think that the "opposite" would be true,
namely that any peer-reviewer should be able to replicate the
statistical
analysis, to verify that the results were not "cut-and-pasted",
by running his/her personal copy of the "black-box" software, after
certifying that the submitter and the reviewer each obtained the
"black-box"
software from an "authorized" source.  Trying to confirm the results,
given "white-box" software (with various levels of
official/non-official
patches/enhancements/alterations) would be a logistical nightmare.

((posted and mailed -- some people have flaky news-servers))