andrews@ubc-cs.UUCP (James H. Andrews) (07/11/85)
Well, here's the good news for all you SDI freaks! (Well, it may not be
news to some of you, but it was news to me.)
D.L. Parnas, the software engineering expert who has a long history of
working with the U.S. military and still has many ARPA contracts, has released
a report very critical of the whole SDI software plan. He says the Pentagon
really has no idea of the problems involved with a project of the massive
scale they want, they are relying too much on the software's freedom from
bugs, and they could not even supply the upper bounds on computation they
would need to effectively plan the real-time systems.
On "As It Happens" last night, he said he had sent copies of drafts of
his report to many people on his network (presumably ARPAnet) and "the
response was amazingly uniform: they all said 'Yawn... Dave, we *know* all
that. But we see this as a great source of research funding.'" He also said
that he knew he was in danger of losing support from the military because of
his report, but that he was involved with them only to find ways of making
the world more secure for the future, and there was no way that he could see
the SDI software plan doing that.
I will undertake no moralizing -- I invite you to draw your own
conclusions. Note that this article represents my recollections of the "As
It Happens" interview, and is not intended as a verbatim report.
--Jamie.
...!alberta!ubc-vision!ubc-cs!andrews