Anyone interested in this should definitely read the description of a lot related work of recovery in the excellent paper by Warren Toomey on the restoration of early UNIX artifacts. There was a lot more involved than just doing an OCR scan of old documents, it is really impressive. http://www.usenix.org/events/usenix09/tech/full_papers/toome...
Back in April 2006, the great Al Kossow of Bitsavers uncovered some unknown papers and documented his findings on The Unix Heritage Society mailing list:
For as strange as it may sound, this amazing finding went unnoticed for 2 years. Nobody paid any attention to this precious paper...until April 2008. That's when the whole restoration process really started:
Another Vernor Vinge concept I love is the "software midden heap", layers of software standing on the shoulders of and papering over the bugs of earlier software. Will anyone be able to excise rotting middle layers without breaking software compatibility?
For example, Vinge's A DEEPNESS IN THE SKY describes a software system thousands of years in the future that still uses the time_t epoch (and none of the system's space-faring users know the significance of 01970-01-01).