Re: [ARSCLIST] Memorex CDs

[steven c <stevenc@xxxxxxxxxxxxxx>]

----- Original Message ----- 
From: "Don Cox" <doncox@xxxxxxxxxxxxxx>
> On 12/01/06, John Spencer wrote:
> > Well, it certainly helps perpetuate the myth that CD-Rs on a shelf  
> > will be easily read 100+ years from now.
> > 
> > For some reason, every time I consult my crystal ball, I can't get an
> > accurate answer on the availability of playback equipment for CD-R
> > media 100 years from now!
> 
> We have playback equipment available now for cylinders and piano rolls
> from 100 years ago. Provided there is no general collapse of
> civilisation, it seems very likely that there will be equipment
> available for archivists to read common media such as CDs in 100 years'
> time.
> 
> DAT might be more of a problem, but I expect some ingenious inventor
> will come up with a device. 
> 
> There is a set of CDs issued by Andante of a live Toscanini performance
> from Salzburg of Die Meistersinger, which was recorded on an obscure
> film soundtrack recorder. A device was constructed to play it back.
> 
The technology of a century ago was all based on comparatively
large-scale mechanics. This meant that we could, and still can, figure
out how it works just by carefully examining it...and we can then
fabricate a similar device to recover the information. In fact,
simply looking at a recording will let someone figure out how it
needs to be played...and build a working, if crude, player. This
probably isn't true of magnetic media (tapes, computer discs, etc.)
and certainly isn't true of optical media (CD's, DVD's)!

Even if a microscope of high enough power was used, all that would
be visible are the pits in the surface...and any information they
carry can only be accessed by taking the apparent information and
applying a computer algorithm to it, but only the CORRECT such
algorithm.

This is something like trying to figure out how a software program
works by looking at the ones and zeroes that make up the program
file (and that is only one of several files that operate together
to accomplish any tasks!) without knowing the details of the
machine language of the computer that ran it...

Steven C. Barr