George,

Karl Miller originally wrote:

>I would agree, but to my question, if you had a donor, what would be the
>best use one could make of say, a gift of $5-10M?

And I responded, apparently triggering Karl's enthusiasm:

>It depends - perhaps the best thing would be to set up an archival
>store of multiple tens or a hundred TB (or expandable to that) plus
>an annuity to pay for the ongoing support plus an annuity to pay for
>incremental digitization.

Notice my two annuities in the list. I think it's imperative that
this be considered. You made a lot of good points in your recent post.

I fear that we have transcended stable media on the shelf, and I am
not sure we'll ever return.

At this point, my best guess for stable media are certain data tapes,
but they still need review...unless we decide to go back to shellac
as some wags at National Public Radio suggested a few years ago on
the first of April. Steve Smolian was even implicated in that effort.

I am not prepared to publish anything yet -- in fact, it appears that
much of the grunt work has been published already but in journals
obscure to media people -- but the more I research this, the more I
believe that tape is not forever, either.

I do tell people that digital archives do not react well to budget
interruptions. I'll go further to say that power interruptions can be
engineered around, but budget interruptions on the order of even a
few years can be catastrophic.

Karl's enthusiastic response to the first part of my conjecture is
great and we need more people talking about this. However, if we look
at current "best practices" of digitizing everything at 24/96, a TB
is roughly 1,000 track hours. Hence my call previously to consider
some bit reduction. My preference is 16/44.1, but other options are
available. 16/44.1 provides 3,000 track hours per TB and I think is
adequate for 1.88 and 3.75 in/s oral histories recorded on Wollensaks
and even Uhers, as well as cassettes recorded on most (non-top-end)
machines. So, my 100 TB store proposal would hold between 100,000 and
300,000 track hours or 50,000-150,000 hours of stereo material, minus
overhead for proxies and metadata.

Yes, they could/should be networked. My proposal for the Cal State
Fullerton project was to see three equal repositories for the state
of California, located in three geologically diverse regions,
connected by high-speed fibre optics. Replication software would be
running as well as background error checking. I still think this is a
good model for storage. I have replicated it here on a minor scale
with three stores, each 1.25 TB, and one is in a geographically
distinct area -- my neighbour's basement <smile>--connected by a
medium speed fibre optic link (100 Mb/s).

I continue to look into LTO tape, as I hear the prices have dropped
somewhat on drives. The tape cost has always been attractive.

Cheers,

Richard

At 12:36 PM 5/24/2006, George Brock-Nannestad wrote:

>----- I think that the major problem is that unless money is supplied on a
>regular basis for maintaining, migrating and also to ensure that
>_cataloguing_ is migrated, then the one-time effort is entirely worthless. It
>did not use to be like that, because paper-based materials had survivability
>as long as they were kept in "human" conditions as we know them in the
>western world and a few other places. And it was mainly an intellectual
>effort to interpret them. So, transfer staged ..... money gone! I fear that
>private funding bodies think the same - they want monuments, not a constant
>drain.
>
>It so much goes against the grain of archivists not to have permanent
>accessibility to permanent media. And I shudder to think that our cultural
>heritage hangs in such thin threads: stable power supply, stable
>manufacturing basis. As Richard Hess has shown us, even private individuals
>may duplicate the threads, but only as long as somebody manufactures them.
>
>----- it is a question of making preservation and accessibility fashionable
>in society, and to develop technical standard procedures that will permit
>every institution to trust what is done in another institution. If we are
>truly an audiovisual culture, then it will logically follow that the content
>must be preserved/made accessible as a public duty. However, we are _not_ an
>audiovisual culture, we are merely consumers, and as long as we are
>entertained, we do not make uprisings, and we do not really care at all.
>
>Not very long term optimistic, I know, but as long as I am entertained - by
>listening to my preferred, very early recordings and live music in a suitable
>mix .........................................
>

Richard L. Hess email: [log in to unmask]
Aurora, Ontario, Canada (905) 713 6733 1-877-TAPE-FIX
Detailed contact information: http://www.richardhess.com/tape/contact.htm
Quality tape transfers -- even from hard-to-play tapes.