NEC-LIST: NEC-BSC status and availability

Bonjour!

I've been interested for a quite a while to use NEC-BSC (especially in
conjunction with NEC2), but haven't been able to procure a source (or
object) version of the code, because of the code's restricted
circulation status. I tried last year to go the official route, and
first contacted OSU, which referred me to the military sponsoring
agencies, which referred me ultimately to Ottawa, but I gave up after
a few calls. (To be honest, I didn't try every possibility, but it
seemed obvious to me that it wasn't going to be easy). The source for
these difficulties appear entirely independent of OSU.

I don't understand why this getting this code should be so
complicated, when the unclassified user's and code manuals for 1979
version of the code are *freely* available from NTIS, which I ordered
and received promptly. So if there had been a military critical
technology or a state secret to protect, it has been in the open for
quite a while. (But what about copyright? AFAIK, this was US
government contract work.)

It would be technically possible to recreate a version from the source
listings. (I once heard a rumor that someone actually did this
somewhere). The legality of that might be technically similar to the
issues regarding Phil Zimmermann's PGP (Pretty Good Privacy), where
it's OK to print a program, and recreate it from the listing, but not
send the actual file across the US border.

But this would have to be done manually, as the quality of the FORTRAN
listing pages of the code manual made from the NTIS microfilm is
somewhat poor, precluding the use of OCR. (But the other pages and the
user's manual are perfect, though).

Would this be worthwhile? I believe that the differences between the
features of this 1979 version and the current one are relatively
small. Current versions have more source options, and can treat
transparent materials. (correct?)

The program computes paths with special case routines for
Reflected-diffracted, Reflected-reflected, diffracted-diffracted,
etc., for each type of entity and possible interactions. I suppose
that this is related to the non-reentrant nature of Fortran 66/77. I'm
not sure that a program written today would use the same approach.

But on the other hand, I think that there would be a great demand for
a minimal version of NEC-BSC, or a general purpose GTD program, which
could be available as easily as NEC2.

I'm posting this to NEC-LIST to trigger a discussion, and hope to have
your views on this topic.

- Do you know of a version of NEC-BSC re-created from the printed
documentation?

- Do you know of a free or quite inexpensive user-oriented GTD code?
(Last time I looked there were none).

- Would it be possible to obtain release an old or obsolete version of
NEC-BSC to the EM community? (Freedom of information act?) What was
the experience regarding the red tape with NEC2?

- Am I putting my mouth in my foot, [er] and it's actually not *that*
difficult for a civilian foreigner to obtain NEC-BSC?

And now for a bold suggestion:

- Would you be interested in eventually participating in a
collaborative effort to create a distributable version of the 1979
published code, which I would coordinate? Sections of the printout
would be farmed out by post for typing, and the results collected by
e-mail. The more volunteers, the better. Supposing 5000 lines of code
(I haven't counted them yet) and 30 volunteers, this would mean
roughly 4-5 pages to enter per head. Volunteers would need to have
some Fortran programming experience (some pages are quite hard to
read). I'd scan the documentation, and place it on the web. The
volunteers would also participate in a second stage in the testing of
the result. This would then become the basis of a user supported and
improved program, distributed GNU-copyleft style. ATTENTION: THIS IS
NOT A FORMAL CALL FOR VOLUNTEERS, BUT ONLY A INVITATION FOR A
DISCUSSION, TO SEE WHETHER THIS WOULD BE A FEASIBLE AND WORTHWHILE
OPERATION. I'm still unsure of the legal issues, I'd have to
check. Your comments on NEC-LIST would be most welcome.

Thanks,

Alexandre Kampouris
Montreal
Received on Mon Oct 12 1998 - 06:05:35 EDT