From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Thu Jan 24 2008 - 02:23:22 CST
Kenneth Whistler wrote:
> About this:
>
> > > ... about the Soviet "Buran" program:
> > >
> > > A significant part of the technical information about the flight is
> > > inaccessble to today's researchers as it is recorded on the magnetic
> > > tapes for BESM-6, of which none are still working ...
>
> Philippe noted:
>
> > There's no more any tape reader working that can be adapted to work with
> > today's computer interfaces? It looks incredible.
> >
> > If those are tapes, may be a professional tape reader (used for audio,
> or
> > for reading some video) could be used, by rewinding the band on another
> > compatible roll. It just looks like a few hours of work from a competent
> > electronician, even if this is used temporarily on a single
> installation,
> > just to process the bands. And this is what some companies are doing
> > worldwide for reading crashed harddisks or tapes, suing various
> instruments,
> > and some custom PLA programmed to emulate the past reader's decoder and
> > connect it to some modern data bus interface.
>
> This is just one more example of the now very widespread
> problem of digital data archaeology required to access
> surviving, important data stores that may happen to be
> 5 or 6 generations of media back, stored in no longer used
> or forgotten digital formats, and with all kinds of other
> potential problems of interpretation.
It jst remembers me with some creations I did at end of the 1980's where the
data was stored on mini-tape cartridges built specifically for the Sinclair
QL... I think there's no more tape reader to process such thing, but it may
still exist some working QL computer for those that really need to read such
antiquity and send it to its serial or parallel port without much work.
I know an old friend that wrote his job reports on a QL. Today there just
remains the printed copies. But the interesting things (some algorithms and
technical specifications for a computer-aided design tool used in
architecture, and the mathematic implementation of a statistic/heuristic
analysis and decision process was also stored on such tapes), and he could
never find at that time a way to transfer it at least on floppies for use on
PC, despite both computers were still built and sold: he was missing the
necessary interface. It's possible that he did not search a lot, and
abandoned it, but if the data had some important value (or was needed for
legal purpose) the cost and time for recovering it would be higher than the
initial cost for recreating it.
And another of his job was in the implementation of a B language compiler,
with a meta-language compiler (with an interesting graphic representation)
that could still be useful for helping designing/compiling other languages.
Such thing has been recreated nearly from scratch, using the printed
specifications and reports.
And I'm sure that now there are lot of people that still have some old
archives on large floppies, and that can't find a floppy reader at
reasonable cost (it already starts to be difficult to find a new 3.5" floppy
readers on PCs, but it's still not difficult to find one in some old PC; but
try to find 5"1/4 floppy readers!).
Soon we'll be exposed to data stored on one of the NVRAM formats used for
photography, or only usable on devices with DRM restrictions already
forbidding their transfers to non similar devices. Look at the same problem
that exists for old TV records: the display and encoding standards have
changed (including color specifications). Consider those video records on
old Betamax; and also those stored on the initial versions of optical disks.
Archivists are already exposed in many places to the problem and cost of
conservation or conversion of these archives: this is often considered not
urgent to handle the problem, but the more we have to wait, the more it will
cost to recover this data. Often they are left without budget to perform the
conversion, and if there was no passionate people to provide help and time,
many archives would be longer usable or could be definitely dead (as all
media supports are also degradable with time, especially numeric supports
whose conservation is much more problematic than paper).
Another similar problem comes as well when preserving printed copies: the
quality of modern papers is poor (due to acidity) and the inks used are not
stable (this includes photocopies and prints with modern laser printers or
some faxes, for accounting purpose some billings "burned" on metalized
paper); I have some books I bought (in the middle of 1980's) when I was
student whose pages are becoming brown, and the glue used to maintain pages
is fragmenting, the books are turning into a pile of separated pages. I can
compare it to the (now rare) books that my father used (when he studied
electricity and electronic for TV/radio transmissions) and printed in the
1950's: they are still perfect today.
Our "modern" age has "exploded" the number of support formats, whose usage
does not past more than a dozen of years before being obsolete. There really
needs to define an international standard for the adoption of
recommendations (to be used by archivists and all legal or accounting
departments) for the preservation of data on formats that will be easily
convertible for longer periods, and defining priorities/schedules for the
conversion of formats that are in danger of becoming very costly to convert
or impossible to convert (such schedule would also expose other problems
such as copyright and patent restrictions on the technologies needed to
process those formats and which are still valid). Many people and
organizations do not even realize that what they are keeping for long is on
formats belonging to classes of supports that are already very costly to
process.
This archive was generated by hypermail 2.1.5 : Thu Jan 24 2008 - 11:10:30 CST