Leslie Newell <les.new...@fastmail.co.uk> wrote:
...
Steven D'Aprano <st...@pearwood.info> wrote:
> > The period of the 1980s through to the 2020s (at least!) will
> > be a future dark age to historians. So many of our historical
> > records are *already* unreadable, after a mere decade or so.
> > Paper, vellum and papyrus lasts for centuries when treated
> > well, electronic records become obsolete and unreadable
> > before you can say "what do you mean we don't have a computer
> > capable of running the only application that can read the
> > data file?".
> Most applications that handle data that can sensibly be edited
> in another application at least offer an option to export the
> data in a relatively common format. For instance most CAD
> packages offer dxf export/import. The problem is that many
> applications have data that only makes sense if it is handled
> by that application. For instance my CAM software stores
> heavily processed CAD drawings. There is no point making the
> data format open because there is no other application that
> could make any sense of it. If my code was open source you
> could then use the data but we have already discussed why that
> is not practical.
>
> IMHO the greatest threat to long term storage of data is
> failure to back it up.
Although you're surely right that failure to back up data is a
big threat to long term survivability, I think Steven's right
about formats and devices.
There was an article in Scientific American some years ago about
this arguing, if I remember correctly, that modern information
created on computers has an average readable lifespan of no more
than about 30 years, after which it is effectively gone. They
compared that, again IIRC, to 200+ years for paper records.
One of the problems with record storage is that we don't know
what will be used in the future. It can be prohibitively
expensive to load every single Microsoft Word 3.0 or WordPerfect
2.1 document found on 5.25 inch floppy disks in the basement,
convert them to a later format, and store them again. Soon there
will be no computers that can read the data, and none to which
you can even attach an ancient 5.25 inch disk drive since the
hardware interfaces and drivers no longer exist. And if you
could get over all that, you'd still have to have a program to do
the conversion.
None of that applies to paper. You put the stuff in big boxes
and shove it into storage. If you need it again, you look for
the box, then paw through it to find your document. You know
when you've found it because you can read it directly - something
you can't do with a floppy disk or mag tape.
Many of the very most critical documents will be preserved. Some
only in paper copies. Some of the very most critical documents
will be lost and many of the less critical docs will be lost. I
suspect that the percentage of survivals will be lower than it
was 50 years ago in the paper era.
And that's just the docs stored on disbursed media. Online is
another story. Where, for example, will all of Usenet be if
Google decides to drop it? Who will pay to preserve it even if
Google offers it free to a good home?
--
Alan Meyer
amey...@yahoo.com
_______________________________________________
Pan-users mailing list
Pan-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/pan-users
