[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [ARSCLIST] Audio storage on external hard drives
At 05:12 AM 2/25/2005, George Brock-Nannestad wrote:
I think that each person who is occupied with preserving knowledge about
sound phenomena for the future (and far future) as well as the sounds
themselves should realize which part they wish to play in that game.
This is a very, very important consideration. It is not limited to
individuals, but needs thorough review by institutional departments and
entire institutions as well.
Richard L. Hess [23 Feb.] wrote:
"As Scott [Philips, 22 Feb.] said earlier, it's a shame that we all need to
become IT professionals, but that is, I fear, indeed the case. Now, I need to
go back to rearranging my online storage systems."
Like in the case of mass deacidifation, it is only a question of money, and
it is only you cheapskates who cannot afford a mainframe system with a
service contract that need to become amateur IT handlers.
My interest in my redundant, geographically separated (well, my neighbor's
home) online storage system is driven as much by my digital photography as
by anything else. It's an opportunity to store and access images in a way
we've never been able to do previously, and I'm excited by it. My clients
get to worry about the long-term preservation of the audio that I transfer
for them. I hold the audio typically for six months after delivery and then
delete the project--all copies of it. I'm not selling the service of being
an archival storage depot. But someone should.
I still remember reading Edward Tatnall Canby writing in Audio Magazine
sometime in the 1970s about the Columbia Oral History project. He scorned the
project for having reused the tapes after having transcribed them. There was
not enough money to spend a tape on each person, so they used them the best
they could. However, it is quite likely that the content of the tapes would
have disappeared anyway in some funding cut "no refreshment this year", and
then what? So, we have the written transcriptions and that is still
infinitely better than either "nothing" or "fewer interviewees".
This brings up the story of the California State University at Fullerton
Center for Oral and Public History collection. They have about 5000 tapes.
To be blunt, they couldn't afford (for a variety of reasons) proper storage
and were not a big enough client of the university central plant system to
stop them from using an economizer cycle that brought in moist outside air
overnight--peak humidity during June/July was over 70%. The tape binder was
breaking down.
I suggested this would be a good project for CSU to set up a centralized
media store with large server farms and/or robotic tape archives at three
locations (after all, California is prone to earthquakes, and three
physically separated archives ensure the survival of two after the worst
case earthquake scenario). I said two would also be good, but why not plan
for ultimate redundancy to begin with?
The need to transfer the tapes was more pressing than getting a project
like this off the ground, and I found no one interested in sponsoring it.
They ended up making two Mitsui Gold CD-R copies of each and every tape
(due to complete in about a month), and one set is being stored off-site.
They are keeping the original tapes.
This was a system that was very easy for volunteers to run. We set up reel
and cassette transfer stations. At one point, I think they had three
Nakamichi Dragons, a ReVox A77, two Studer A807s, and assorted other reel
decks. They had four or five pairs of Sony CDR-W33 CD writers and burned
two copies of each tape in real time.
This is a good interim solution, but they could have done so much more if
the support and infrastructure had been there.
I challenged the Society of South West Archivists last year to see if they
could think of a regional repository that would be a central store for
these assets.
I believe that the cost curve is an interesting one:
It's relatively inexpensive to deal with a few (less than five, say)
Terabytes of data on your own.
After that, you start needing an IT department. At that point, the cost per
Terabyte skyrockets. From there, as you add Terabytes, the price falls,
until you need to make another step up in infrastructure, then the price
per Terabyte jumps again. It looks like a series of sawteeth.
Finally, however, a really, really large data archive can provide a
reasonably low cost of storage, I believe. I'm not sure it gets as low as
the one person with less than five TB, but it gets low per TB (or PB).
However, this needs to be a redundant store with geographic diversity IMHO
and a steady stream of funding. Managed stores are not as forgiving to
interruptions in funding, although we're currently seeing the damage that
is being done by the interruption of electricity to a film storage facility
in India, so any high-tech storage system, be it data or refrigeration,
does not suffer budget or infrastructure degradation well.
Cheers,
Richard
Richard L. Hess email: richard@xxxxxxxxxxxxxxx
Vignettes
Media web: http://www.richardhess.com/tape/
Aurora, Ontario, Canada (905) 713 6733 1-877-TAPE-FIX