NEC-LIST: NEC Web User Manual

Hello everyone!

It would not be a major matter to scan the NEC2 User's Manual and place it
on the Internet as a PDF file.

The question is one of copyright: From the IEEE copyright release form, it
is my understanding that US government work is not protected by copyright.
But since NEC2 was not exactly always freely available because of all of
these export restriction, is there possibly a special convention applying to
the reproduction of its documentation?

Could anyone presently or formerly associated with LLNL lurking on this list
help clarify the matter?

Adobe offers several excellent tools, which I admire and endorse, to produce
scanned documents. Adobe Distiller and Adobe Capture range in price from
US$400 to US$1000. I unfortunately do not have them (yet).

This expense is not essential, as very acceptable results are possible with
only simple tools and freely distributed software.

The elements required are the following:

1) A FAX machine with a sheet feeder, which is used as a scanner.
2) A PC with a fax modem and software. (I use Eclipse FAX)
3) The Adobe PDF-Writer driver, which is bundled with the Acrobat Reader

At an average production rhythm of 2min/page, the complete ~200 pages NEC
users manual could be processed in a single long evening session.

The tasks are:

1) Making sure the pages are feeding correctly in the FAX machine, and that
the the scanning head remains clean. The pages are grabbed by the FAX
software and stored in TIFF/G4 format.

2) Interleaving the pages once they are read into the computer, as they were
scanned in groups of odd and even pages. This is the most time consuming
step. It could possibly be automated by hacking the TIFF file page
directory, but would still require thorough checking for missing and out of
sequence pages.

3) Eliminate the shadows from the binder holes and also the FAX machine's
CSID message, which can't be turned off. (On my model).

4) Output the PDF using the PDF Writer driver.

The result would be a bitmapped PDF, scanned to a resolution comparable to
FAX high definition, which is about 200x200DPI. The bitmap compression used
in the file is CCITT/G4, with ASCII85 coding which results in a size of
about 20k/page. This is inferred from a test that I have just made on pages
45-47-49-51 of the NEC2 user manual, which I have decreed typical.

For the user manual, the resulting file size is about 4MB, which is a long
but manageable transfer on the Internet.

With much work and compromise, one could probably reduce the file size by
about 50-60%.

Printing the PDF document produces something comparable in quality to the
original. Windows produces 130k of spool file per printed page. Reading on
the video monitor is also possible, but Acrobat does not produce the most
readable on-screen presentation. (This is one aspect of Adobe products which
could be improved.)

Once the file is created, the generation of an index, and hypertext markup
of the file could be performed with someone with the proper tools.

The issue of selecting a PDF page size eventually compatible with both B and
A4 paper formats needs to be resolved, but this is a trivial matter.

Pages which are NEC source decks, FORTRAN listings, or program output could
be generated directly from the corresponding computer files, reducing the
PDF file size, and somewhat improving the appearance of the corresponding
sections. This raises however the issue of conformity to the original document.

My questions to this distinguished group are:

1) If the copyright is not a problem, would a 4MB PDF version of the NEC2
manual be acceptable and convenient? How many takers?

2) Would there be interest in producing a PDF of the NEC Theory manual (81
pages - about 1.6MB)

3) How about Part II, the program code, (433 pages long, about 9MB PDF)?
Anyone thinks it is essential?

If anyone wishes to see the result of the test mentionned hereabove, I could
e-mail or FTP the four pages mentionned. That file is about 80kB long.

My own feelings on the questions above are:

1) Maybe, as much work has been done already. Two versions of the document
could coexist: an online reference manual, and the PDF file for printing.
Once the document is captured, I could contribute to the former scans of the
missing figures, and a crude OCR of the missing text.

2) Yes, as it would otherwise be too much work to create a LaTeX document,
because of the profusion of formulas.

3) No, or maybe partially, because of size. Assembling the pages from both
source files and scans would take several days.

If there is sufficient interest, I could undertake to produce PDF files
within two months, when I have some time. A server would be needed to host
the file.

Greetings,

Alexandre Kampouris

Radio-BIP
Montréal
Received on Fri Jan 10 1997 - 00:01:00 EST