On Wed, Mar 08, 2006 at 02:12:23PM -0800, Ken Durling wrote:
> Not magnetic tapes - the adhesive that holds the particles to the
> substrate deteriorates in 15 years or so. I know this from bitter
> experience. It also matters greatly if the tapes are stored head or
> tail out. The print-through resulting from long-term storage head
> out can obliterate material on neighboring layers. Cassettes - well
> they don't last long at all.
Some do, some don't. Some 9-track tapes had particularly flimsy oxide layers.
QIC cartridges were much better. My co-worker recently extracted a bunch of
them as part of an archiving project. I asked him what his experience was.
"I found that qic-02 tapes were actually very good over the long haul. I read
many, many tapes from the late 80's that had a successful extraction average
of around 98%"
DDS tapes had a terrible reputation, perhaps deserved, due to their helical
recording. Still, they were inexpensive, so that's what I did my backups on
from 1994-2000. I always kept a couple of the more recent tapes in a fire
safe, but the others were stored on a shelf in the open. I recently extracted
all of them to build a file history database (disk space being rather cheaper
now). Out of ~60 tapes, mostly 120m (very thin media) DDS-2 cartridges and
read on a different-generation, different-manufacturer drive, I had just two
that exhibited any read errors at all. I'm sure it helped that I verified them
all after writing, and of course reran backups that failed verification.
I think this notion that old media is almost guaranteed to fail is really
reflecting a notion that any significant *chance* of failure is unacceptable.
John
--
John DuBois [EMAIL PROTECTED] KC6QKZ/AE http://www.armory.com/~spcecdt/
*
****
*******
***********************************************************
* For list instructions, including unsubscribe, see:
* http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************
