(It is actually all available on a hidden web page that I disclose if someone sends me a pointed email, but I'd hate to stress my little T-1.)

Back in March 2001 I posted about a cache of 20,046 pages of scanned docs I received from someone on the net. See the TOC below, followed by his explanation of how he did it. It consumed several CD-Rs, compressed. I now have a DVD burner as well, so I'd be glad to make copies on new or old media. (It is actually all available on a hidden web page that I disclose if someone sends me a pointed email, but I'd hate to stress my little T-1.) Anyone care to upgrade it to OCR'd PDF or whatever would be considered a next-best method of preservation and search-ability? I know it's possible with a handful of Linux, but my to-do list is already too long. - John I've made contact with a guy who's scanned 20,046 pages of the docs listed below, at 300 to 400 DPI. He first told me about the UCSD p-System docs he'd scanned. Below the list is his description of the process he followed. I'm planning to get a copy of what he has and burn it to CD-R. Does anyone else have an interest in these docs, or have any ideas about distribution without massive copyright violation? - John 6502 MOS 6502 datasheet 6502 Assembly Language Subroutines (Leventhal) AMD AMD 29000 Memory Design Handbook Am29027 Arithmetic Accelerator Am29C327 Floating Point Processor Data General C Language Reference Manual GATE User's Manual AOS/VS Internals Manual AOS/VS Programmer's Manual, volume 1 AOS/VS System Calls Dictionary CEO User's Manual Eclipse 32-bit Principles of Operation Eclipse 32-bit System Functional Characteristics Fortran-77 Environment Manual Fortran-77 Reference Manual Fairchild Clipper User's Manual IDT RISC System Programmer's Guide R3000 Assembly Language Programmer's Guide R3000 Hardware User Manuals R3000 Language Programmer's Guide High-speed CMOS databook Motorola 68000 Family Reference 68020 User's Manual 68851 User's Manual 88100 User's Manual 88200 User's Manual Linear Interface Integrated Circuits NCR 53C90A/B Advanced SCSI Controller (2 different manuals) 53C94/5/6 databook 53CF94/96-2 Fast SCSI Controller Disk Array Controller Firmware Disk Array Controller Hardware Disk Array Controller Software Floppy Disk Controller (SCSI-to-FD) National Semiconductor NS32532 Datasheet Series 32000 Programmer's Reference Manual DP8490 Enhanced Asynchronous SCSI Interface NS32CG16 Programmer's Reference Supplement Graphics Handbook Series 32000 Databook DRAM Management databook Embedded Controller Databook Ohio Scientific C4P User's Manual (2 different manuals) 65V Programmer's manual Schematics for: 502 CPU board 505 CPU board 527 24K memory board 540 Video board 542 Polled Keyboard Pinnacle Systems 2 User's manuals for their 68k machine (My P-system machine) P-system manuals IV.12 Operating System Reference Program Development Reference Application Development Guide Fortran 77 Reference Assembler Reference Weitek WTL4167 Floating-Point Coprocessor datasheet Most of these are from about 1988 to 1992, with the exception of the OSI documentation, of course, which is from 1979. --- As far as the process, I scanned a manual in and checked to make sure all the pages were there. If they weren't, I'd scan the pages that didn't make it, and go through all the pages again. I'll admit this is a little anal, but better safe than sorry. (When you're using a lot of shell scripts, you never know if you accidently deleted a page with an "mv" command.) When all the pages where there, I'd go through the manual one more time to check for general quality (no folded corners, no torn pages, etc.) If all was good, the manual would be moved to the directory that would be the root directory of my CD-ROM. That's pretty much it. The big manuals of more than 1000 pages really sucked, because I'd generally have to make 3 or more passes to get those completely correct. If I was going to do it again, I'd probably break the larger manuals into smaller chunks to avoid this problem. One thing that made the whole process a lot easier was the netpbm utilities. I wrote a script to convert the manuals from ~2500x3300 TIFs to ~500x600 GIFs. My machine takes about 2 seconds to process a 300-400 DPI TIF, but only a fraction of a second for a 75 DPI GIF. I'd run my script, then do something else for a while. When it was done, I could flip through the GIFs with GQview and inspect about 2-4 pages per second. That saved a lot of time. I assume that, by "devices", you mean what type of scanners I used. I started with an HP 6350cse (with ADF) that I bought for this very purpose. However, having never owned a scanner before, I was a little disappointed with how slow the "fast" scanners are. Fortunately, imaging is an integral part of the software my company sells and, as luck would have it, we were demoing a new scanner from Fujitsu. This thing literally does 60 pages/min at 300 dpi - *both* sides. It's about half that fast at 400 dpi, which I had to use for the IC databooks to get the fine print. Needless to say, I did most of my scanning on that. By the way, to date, I've processed 20046 pages. I'm kinda burned out, though, so it'll be a while before I do any more.