X-Message-Number: 5663
Subject: archival storage
Date: Thu, 25 Jan 1996 11:37:50 -0500
From: "Perry E. Metzger" <>

> From: "Keith F. Lynch" <>
> 
> In #5642, "Perry E. Metzger" <>
> > That isn't saying much. Many forms of tape backup won't last more
> > than a few years.
> 
> Does anyone know how long DSHD 5 inch diskettes last?  I have about a
> thousand of them, up to ten years old.  I store them in my clean, dry,
> but very warm (85F or 30C) apartment.

Unclear. You won't have print-through problems, but someday you may
get problems with the oxide detaching from the substrate or other
things one can't even anticipate. I'd say you are probably okay for a
while longer, but who knows?

> The web page mentioned in Leonard N. Zubkoff's message #5636,
> http://www.cd-info.com, contains claims from various manufacturers
> that CD-R discs stored under normal conditions will last 70, 100+,
> or 217 years.  Even longer, given better CD-R readers.

I'd say that any such claim is necessarily crap. There is no way to
actually age a CD-R 217 years, so you can't find out what sort of slow
chemical reactions might occur. Also, many things will last far longer
under ideal or freak conditions than under normal conditions --
corpses, for example.

> > Lastly, I am quite unsure as to whether or not you are going to
> > find an ISO-9660 CD Reader lying around in 100 or 300 years.
> 
> I can't believe that a technology capable of repairing brain cells
> would be unable to read a CD-R disc, even in the very unlikely case
> that all information on how data are stored on them is lost.

The issue is whether you want to put out data and hope that someone
can reconstruct it, or whether you want to do the best job of storage
you can do. This is much like when people doing cryonics don't simply
flash freeze people and hope that the technology will be good enough
someday -- they try to do the best possible preservation so as to
minimize the work people have to do in the future and maximize odds
that the person can be fixed. Well, are you trying to maximize the
odds that your data will go through, or not? Something that survives
to the future without the need for reconstruction has far better odds
of being read than something that requires reconstruction.

Perry


Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=5663