[AMIA-L] Reply: [AMIA-L] Preserving digital video vs. film

[Leo Enticknap <ldge@U.GENIE.CO.UK>]

Roger Smither writes:

> Readers who have been following this strand may be interested in a story
> of hardware obsolescence carried in the British newspaper 'The Observer'
> last Sunday (3 March 2002).  The headline was "Digital Domesday Book lasts
> 15 years not 1000" - the full text can be found on
> http://www.guardian.co.uk/Archive/Article/0,4273,4366728,00.html
This story is full of errors, inaccuracies and exaggerations.

For those who aren't familiar with the hardware and software specifics of
the BBC Domesday Project, it was delivered on 2 double-sided 12"
laserdiscs.  Sides 1-3 were CLV and carried text, images and sound
recordings (I seem to remember that one selling point of the project was
that it included a photo of every town in the UK), whilst side 4 was a
conventional, CAV PAL video containing some BBC national TV news
reports.  You could even play side 4 on a conventional, unmodified video
laserdisc player.

Hardware wise, it could be made to run on x86 PCs fitted with special
graphics cards and a laserdisc drive controller interface.  I was a high
school pupil aged 14 in 1987 and distinctly remember it running on an
MS-DOS platform.  The claim that it only worked on specially modified BBC
Micros is just plain wrong.  True, the more common version consisted of a
computer specifically designed for schools, the 'BBC Micro', which was
built by Acorn computers.  A quick web search will reveal large numbers of
BBC Micro fanatics still out there and I'm sure the hardware and software
needed to run the laserdisc drives and Domesday discs could be obtained
after a little bit of sniffing.

I suspect the real problem was that the data sides of these discs stored
information in a weird mix of analogue and digital formats.  They didn't
follow the standard CLV pattern for video laserdiscs.  I seem to remember
that while the text was stored as digital data, sound recordings were
encoded as analogue waveforms and I can't remember (if I ever knew) how the
control hardware and software differentiated between the two.

But this project is, as the report points out, only 15 years old, and I
can't believe that the technicians and engineers who developed the disc
format and created the masters did not keep records of the data storage and
retrieval processes they used, and I find it difficult to believe that the
glass masters for the discs were not preserved (the article states that the
computer scientist trying to preserve the data only had two scratched sets
of release discs to work with).

Whilst I, like everybody on this list, appreciate the potential (and
actual) problems of data storage obsolescence, this particular article
seems to me to be grossly exaggerated and, as I say, there are also
inaccuracies.  Many thanks for the link, though.

Leo