> However, it is in printed format (about six LARGE three ring
> binders). Is there any better way rather than just keying in the source?
I tried OCR'ing some of the HP-41 microcode listings that were released
through the PPC club many years ago. Some were dot-matrix, and some were
impact printed, but none of it OCR'd very well.
I ended up typing in all 500-plus pages of it by hand. I'm a reasonably
fast and accurate typist, so it didn't take me too long. On the one hand,
it was somewhat tedious, but on the other hand I learned much more about the
inner workings of the 41 than I ever would have otherwise. Overall I thought
it was fun even though it was somewhat tedious.
Maybe OCR has gotten better since then.
I used to use Adobe Acrobat 3.01's "capture" OCR module. I now have Caere
OmniPage Pro 8.0 (and they announced 9.0 just a few weeks after I bought it,
damn!), and Xerox TextBridge Pro. I haven't tried to OCR those old listings
again; maybe I should do so for comparison purposes.
If OCR isn't up to it, I'd be willing to do some typing of TSB listings.
Even if the OCR does work, it is easier to merge two files, one OCR'd and
one typed, than to proofread either, and the result will likely be more
accurate.
Another idea would be to run the scanned images through two or even three
different OCR programs, and compare their output.
Did you get the listings of the I/O processor?
Cheers,
Eric
Received on Thu Feb 04 1999 - 15:58:33 GMT
This archive was generated by hypermail 2.3.0
: Fri Oct 10 2014 - 23:31:59 BST