🚀 go-pugleaf

In article <4eu4mo$gff@news.microsoft.com> doriang@microsoft.com (Dorian Garson) writes:
>Bloody heck, my newsreader won't let me kill this article!
>
>Oh well.  I realized immediately after sending that no work is required at 
>all -- who cares if you can decode the data stream on a PC.  
>
Yea, but if you don't decode it, it's likely to get pretty big.
Most PCM style recordings do get pretty big.  It would be interesting
to investigate the hardware and see just what you could get away
with in terms of sampling rates, etc.  Also, being able to decode
the tapes might make it possible to restore programs where sections
of the tape have become illegible.  Most of the old machines aborted
when they hit an error.  If you could write a PC program to load
the whole tape, you could probably correct most one character
errors, or maybe use two corrupted copies to fill in the gaps
that each has.  You could also clean up a marginal recording,
maybe the tape has stretched some and the frequency shifts a little.
You could make the program smart enough to detect and correct these
problems and produce a "perfect" program tape.