I thought the earlier responses you received pretty much covered the gamut,
which is why you likely didn't receive more.

The primary guidance most people refer to the the NIST Special Publication,
500-252, which gives numbers for media longevity under optimal conditions
and also how you determine whether or not your media (and more importantly,
the data recorded on it) remains viable.

There are numerous caveats though, most of them related to the initial media
type you're using, the storage and handling conditions it's subjected to
(environmental factors, including moisture, dust, heat,vibration/shock and
light) and also decisions made regarding labeling.  All of these factors
weigh heavily into the longevity of media.

There are a few things not mentioned in the publication though, such as the
inability to determine what forces and conditions the media may have been
subjected to PRIOR to your taking ownership of it.  Storage in warehouses
with extreme heat, transportation in vehicles lacking climate control,
exposure to electrical fields, shocks, bumps, and bangs while being
transported from point to point all weigh into the condition the media may
be in when initially received.  

The reason some of these are critical is the potential for damage to the
dyes and also the edge seals of the discs, which when breached allow dust,
moisture and other elements into the area where your data is recorded.  Some
of the earliest reports of damage and/or loss of data on CDs (late 1980s)
were from the West Indies, and it was determined that the media had been
subjected to severe handling, including being dropped and stored in high
heat conditions, which caused the seals to split on the edges and moisture
seeped in and attacked the recordable surface/s of the disc. 

These articles address the phenomena known as "CD Rot", which was widely
discussed in the early 2000s primarily by music enthusiasts.

One of the most authoritative reports on the subject of media degradation
and the causes was published in Scientific American in 1995 (and revised in
1999) by Jeff Rothenberg

The other source of guidance, and this applies primarily to Federal Agencies
and their Contractors generating permanent retention Federal records for
transfer to NARA is found in 36CFR, which states:

"Before the media are 10 years old, agencies must copy permanent or
unscheduled data on magnetic records storage media onto tested and verified
new electronic media."

36CFR also provides the following as guidance for sampling of TAPE, but does
not give any similar guidance for other media, but this could be used as a
means of setting a threshold for testing.

"Agencies must annually read a statistical sample of all magnetic computer
tape media containing permanent and unscheduled records to identify any loss
of data and to discover and correct the causes of data loss. In magnetic
computer tape libraries with 1800 or fewer tape media, a 20% sample or a
sample size of 50 media, whichever is larger, should be read. In magnetic
computer tape libraries with more than 1800 media, a sample of 384 media
should be read."

Hopefully, this provides the level of detail you were seeking.

[log in to unmask]

List archives at
Contact [log in to unmask] for assistance
To unsubscribe from this list, click the below link. If not already present, place UNSUBSCRIBE RECMGMT-L or UNSUB RECMGMT-L in the body of the message.
mailto:[log in to unmask]