• Please review our updated Terms and Rules here

How can we preserve more software and documentation?

Ragnarock

Experienced Member
Joined
Sep 15, 2017
Messages
268
Location
Texas
How can we preserve more software and documentation?
This stuff is being lost every day.
There's so much, one person can't do it, it's gonna take a crowd.
How can we crowdsource and/or crowdfund this archiving process?
 
There's so much, one person can't do it, it's gonna take a crowd.
You end up with the Internet Archive. randos doing shitty scans or not having the knowledge of how to properly image and archive media

What we do need is more people who can recognize rare things at flea markets or estate sales and storage lockers to get to this stuff before
the flippers can get to it. Or, in the case of estates to convince people that there are people trying to preserve old computer (as in 80s and
earlier) paper/catalogs/media and they shouldn't just throw it out. Flippers are going after keyboards and throwing everything else away.

I've cut WAY back on what I'm willing to spend money acquiring on eBay.
The collectables pricing is getting way out of line there.
I have been trying to restore a Megatek vector graphics system. A keyboard for it came up and I
was outbid at $680 for it. Clearly either a keytard or someone thinking it was for Apple because
the seller misidentified it. On the other hand, I got the knob and button box for $18. WTF?
 
Last edited:
Really, it is up to us to hunt stuff out, get it scanned/imaged and point Al at it to "do his thing" on bitsavers.

Dave
 
I should point out I really don't "do" consumer computer documentation/software collecting
That is taken care of by lots of other people.
I have some up, but it really is pretty minimal.
 
The poster child for what is missing is minicomputer and mainframe software.
The amount of interest in the PDP-11 vs what software was preserved is embarrassing.

I'm saddened by what is going to be lost for LCM's running SDS Sigma 9 system dissappearing
into some unnamed private collectors hands.
 
By making copies and keeping those apart one from another in order to have redundancy.
 
Well, the fist thing is to hunt down and find the software or manuals. I praise anyone who has the time and resources to do that. Poke at social groups to let them know someone wants this stuff, search garage sales, flea markets, and so on. If you are rich, buy stuff on eBay.

If you can't archive it yourself, at least store it somewhere where it won't rot too much more.

Ask around on places like this, there are various people, such as myself, who are set up to archive such things and may be able to assist.

For the inexperienced, keep in mind that it is not a good idea any more to just plop a random floppy disk in to a drive. They may need cleaning or treatment, otherwise they may rip themselves up. Copy protected software requires special flux-level hardware to archive properly.
 
Internet Archive takes a high quality scan and reduces it's resolution to save space, so doesn't preserve that well, but better than nothing.
I did try OCR'ing old manuals, but there was a huge amount of post scan work cleaning it up.

Maybe the so called 'AI' has a use if it could do he task well?
 
I would NOT trust an "AI" to do post-cleanup work. Many scanning tools have already had automated tools like that built in for a long time, and they frequently get things wrong. Hole punch removal algorithms, for example, often remove any similar sized circles from diagrams on the page.

For common regular things like adjusting cropping, contrast, etc, I use ImageMagick.

But if a page needs repair (such as removing doodles, page rips, mold stains, or such) it is better to do things manually.
 
I use Adobe Acrobat X for PDF generation and OCR. It does do minor page alignments if necessary but unless it's an awful photocopy or lots of equations it gets the in-document OCR search correct.
I don't generally bother with pen/pencil marks, smears or wrinkles unless necessary. I'm documenting and archiving. Not preparing a masterpiece.
 
You can't crowdfund to pay $300 for a manual someone is extorting on ebay.
Yeah that's what happens I've been at the end of both sticks and every time I have a chance to hold something for ransom, I don't do it I put it out there for free or beer money, because even if it's something I'm not into somebody out there needs it.
I got burned by a guy once was buying something and he found an extra manual but then wanted buku bucks for it. This was several years ago so the price wouldn't see me zuberant now but at the time I had to pony up. I swore then I would never do that to anybody else if I were in the position and I haven't
 
I found some useful technical manuals at work that would have gone to the skip if I hadn't rescued them.

I gave them away to someone with the equipment but no technical manual (just to recover my postage costs).

People have given me things in the past in exchange for help, and I am obviously happy to reimburse them for their costs plus a beer!

Long may this tradition continue within our circle...

Dave
 
I'd love to learn how to do a better job with scanning and archiving. Is there a best-practices or guide to how to do this well somewhere?

"Best Practices" is more like whose religion you believe. I have my recommendations for scanning on the home page of bitsavers
Common sense isn't the amateur scanner's strong suite on IA. There is no point in scanning a document at some absurdly high
resolution that you couldn't see in the original document, or so low that it looks like crap. Many, many manuals were scanned in
the 00s at too low of a resolution. The boatanchor archive and even HPs microfiche scan archive is full of them and people with
access to long-format scanners are slowly redoing them with full-length adequate resolution replacements. My recommendation
of 600dpi bitonal for text and 300 jpeg (or jpeg2) for images has been a compromise for resulting document storage size.
Bitsavers total size is still under 2tb for tens of thousands of documents. The people who provide my bandwidth and disk space
would have my head if I started pushing out 1gb documents. The biggest I'm willing to go is a few hundred megabytes and that
was for full-color magazines. https://www.worldradiohistory.com went too far in the opposite direction. They've scanned a lot of
stuff but made the resulting files with too much contrast and too low effective resolution (you can't read schematics in them) This
is sad for the amount of work he's done. Hopefully he started with better looking files and didn't discard them.

You could write a book on media archiving. I'm actually writing one for internal use now at CHM from my experience doing it
there for almost 20 years. Two common sense things, take a picture of the media especially the label for establishing provenance
and digitize and verify the contents at the lowest level practical. Someone mentioned the past week that they are working on
a catalog of copy-protected media, something I've wanted for a long, long time.

https://archive.org/details/@f15sim is an example of someone doing excellent work on document and floppy archiving on IA
 
Last edited:
Thanks @Al Kossow for the tips. I'm guilty of going for too high a resolution without thinking about the storage implications. I appreciate your thoughtful common sense balancing resolution vs file size.
 
Back
Top