14 Apr
2009
14 Apr
'09
2:09 p.m.
On 4/14/09, Al Kossow <aek at bitsavers.org> wrote:
I have all of the diag listing on fiche. The scanner
that I have is a bit
difficult to use, so it takes about an hour per sheet to set up and scan.
Wow. That's fiddly. I can understand things being difficult - the
information density on a fiche is rather high.
Speaking of fiche scanning, I was going through some boxes last week
and ran across the source fiche for VMS 2.0. I know it's far from the
only copy in the world, but it did make me wonder if anyone has an
electronic copy. It'd be a hell of an OCR project (or an interesting
Captcha dataset), but even scans could be interesting to read through.
What I have is one 3-ring binder with a special fiche-storage insert -
you can read the top 1/4" to see that you have the right fiche, and
the lower 90% is protected from dust and scratches. There's at least
20 fiche to a "page" in the notebook, and it's one or two
"pages".
Elsewhere, I know I have VMS 4.0 and probably other versions of source
fiche. Of course, I do have a reader (more than one), but looking at
things on a modern machine without having to go to where the gear is,
make room to set the reader up, etc., means that most of the time, the
fiche just sit - no casual browsing and no easy way to share or
preserve the contents.
I know the group has hashed and rehashed fiche scanning and I don't
mean to re-open that debate. Obviously fiche scanners exist,
reinventing the wheel isn't cost effective, etc, etc. It's
unfortunate that there's so much manual fiddling to get a good scan,
but those letters are awful tiny.
If I were on the left coast, I might consider volunteering to drive
the scanner just to be able to share the results. I have a pile of
fiche myself, mostly from the late 1970s through the mid-1980s, IBM
and DEC docs primarily, and it would be great to know that it's not
only accessible as slivers of film.
-ethan