[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [ARSCLIST] Long-term/preservation audio
Chris,
"permanent digital archive" is a shorthand for a managed archive. Sorry, I
should have been more specific. Permanence is acheived by migration when
necessary.
When I refer to this, I refer to files in a managed, long-term storage
situation. Some third-party suppliers are starting to provide this as a
service.
I doubt we'll have much trouble reading WAV files in 100 years. They may be
laughed at, but it won't be a problem, I suspect.
Also, when I was referring to CDs, I was referring to red book audio CDs.
That is the pervasive, common denominator.
I'm seeing archives of oral history waste away--I'm managing to personally
help salvage two--I wonder how many more are out there.
One of them had the funding to also transfer to reel (38 hours) and I am
doing that. The other, with a worse problem, is struggling. They have 3000
hours.
I see how hard it is for many people today to get good playback from analog
tapes. Yes, someone like you and me will be around in 100 years to struggle
through playing analog tapes, but it won't be the norm, and it won't be
worth the cost in many instances.
Perhaps my view is skewed, but many of the archives worth saving are, at
the best, good cassette quality. Certainly there are some recordings which
deserve more than red book CD, but many of the archives out there that
provide a snapshot of our world--our voices, our sounds, our thoughts, our
inflections--are seriously not even Nakamichi cassette quality.
We're not talking Metropolitan Opera or Chicago Symphony for most of these
archives, but the cultural significance is no less important.
Perhaps people will want to transfer red book CDs into a permanent digital
archive or some other storage medium in 25-50 years and that is fine. I
don't think the 3,000 tape archive would be around then.
By the way, to put this in perspective. The record machines for the vast
bulk of the 3,000 reels were Wollensaks at 3.75 in/s. Please, will someone
tell me how this audio quality will be hurt at 44.1ks/s, 16 bit? We're
playing these 3,000 tapes on a Studer A807 and recording them onto two Sony
CDR-W33s.
Yes, there is a broad spectrum, but with the generational loss of analog
recordings, we can't go too many generations without serious degradation.
Capturing the sound to digital freezes the losses now.
With the challenges of getting good people to play analog tapes, I cannot
imagine the challenges of getting them to record them as well as possible.
I know some very good analog tape people. I don't see many around.
I'm not advocating destruction of master reels. If a better
playback/encoding scheme is around in the future, the transfers can be done
then.
I see overlapping curves as analog reproduction quality falls and digital
capture quality increases. The trick is to pick a time where the analog
reproduction quality hasn't fallen substantially and the digital capture is
starting to plateau out.
I think you're aware of a major west coast archive on a fishing expedition
to see how much it would cost to transfer 4000 or so assets to files. In
fact, in your day job, I suspect you bid on it, and I wish you well. They
were not requesting analog safeties as I recall it. I think their head is
in the right place.
As I said it's my opinion. You certainly are entitled to yours. I hope
we're not getting into an analog/digital debate, but to me the path is so
clear for the vast majority of archives...but in case I'm wrong, keep the
original reel masters, too.
Cheers,
Richard
At 12:33 AM 6/26/2003 -0400, chris2 wrote:
Richard,
First, I have to say that I have no interest in arguing analog vs.
digital at all, and that is not my intention here. The issue is too
complex to be so black and white. Criteria for proper choice of format
should be dictated by the
ability to meet preservation goals and objectives of the collection or
institution. Having said that I have to say that I disagree with your
statement. Use of the term "Permanent digital collection" is a bit
misleading. Where does the assumption of permanence come from? Digital
archives need an infrastructure to support them and proper maintenance
and upkeep. Not every institution has the hardware or IT knowledge to
support a digital archive. Sure, you may not need an IT department, but
you need, at very least, easy access to someone with significant
understanding and working knowledge of IT and digital file storage and
access. Maybe hard drives are getting cheap, but digital preservation
does not consist of finding the cheapest hard drive available to store
files on. What about initial and ongoing media and data integrity
checks? These are not things to take lightly and not things that a cheap
terabyte drive offers. What about metadata specifications, storage and
access? What about a migration plan? There has to be policies and
guidelines in place. Yes, digital archives are a current possibility and
are actual reality at this point, but they are only an appropriate
choice if you have the right people, knowledge and systems in place. It
is certainly too harsh and simplistic to say that someone who chooses
open reel tape is being "Financially irresponsible".
In regard to access in 100 years, based on the playback mechanism, I'll
take the open reel deck in a bet. If you can wrap wire around a chunk of
iron and move the tape at a constant speed you can get the data off and
recover something useful from tape. Simple mechanisms and systems that
can be easily reconstructed beat complex ones.
The chances of being able to build a machine from scratch in 100 years
that can play analog tape is much greater than being able to build a
computer, operating system and drive mechanism that can access a CD and
then play back the file format stored on the CD. If obsolescence is an
issue in the analog domain, it's a crisis in the digital domain. Look at
the rapid changing of technology in the digital domain just over the
past years. How much data do you think has been lost and is stranded on
obsolete computers, operating systems and media? There may be more CD
players than open reel decks that have entered into the market, but that
is only part of the equation. After you have a drive mechanism, computer
and operating system there are many more complexities. First, are we
talking about a data CD or audio CD? If it's a data CD, how is it
formatted and what is the file system? Is it a CD-R or CD-RW? 650 MB or
700 MB? What is the file format of the data on the CD? What is the bit
stream encoding method, sample rate and bit depth? Is there a container
or wrapper involved? Is the metadata stored internally or externally and
in what format? How many CDs is the data split over and what is the
consistency of metadata between them.
I really say all this to demonstrate the inherent complexities in
digital storage, and this is only scratching the surface. My description
of tape is also an oversimplification to some extent, but I'm just
trying to make a point. Truly, chances are that future archivists and
librarians will be confronted with a lot of WAVE and PCM data and will
be able to fairly easily interpret this data. This implies many things
though that can not easily be glossed over. Also, while Jerome Hartke
gives good input on CDs, CD may not truly present the best option to
many institutions and can not so easily be given credit as the best
stopgap before digital archives are available to the "average archive".
Depending on the goals and objectives (preservation, interim storage
medium, access) there may be other viable options beside CD. The
decision of preservation/destination formats is a big one and worthy of
careful consideration, particularly for larger collections. There is no
right or wrong answer at this stage in the game. It's not about if a
Macintosh or PC is better. It's about which tool is right for the job.
That's my .02.
Chris Lacinak
-----Original Message-----
From: Association for Recorded Sound Discussion List
[mailto:ARSCLIST@xxxxxxx] On Behalf Of Richard L. Hess
Sent: Wednesday, June 25, 2003 5:47 PM
To: ARSCLIST@xxxxxxxxxxxx
Subject: Re: [ARSCLIST] Long-term/preservation audio
Hi, Lance,
Since the quality of reel-to-reel playback is very machine and
technician dependent I have serious doubts about the playability at
highest quality of an analog magnetic tape in 100 years time.
By then, the technology will be stone-age. Nobody will be around to
understand it. No one will care. It's dead technology now. It started
dying in the 1980s with Ampex leaving the business. Studer is now
essentially out of the tape recorder business.
What will happen when people like Jay McKnight of Magnetic Reference Lab
no longer make test tapes?
Can we even buy good blank tape today? For how long? Quantegy and Emtec
are still in business--the latter is or was in a bankruptcy-type
restructuring as I understand it.
Few highest-quality recorders are available new today and those that are
are end-of-life products being kept in production for archival
migration. The Otari MTR-15 is a special order product but the MX5050
BIII is apparently available from stock. Studer is currently NOT listing
the A807 on their Web sites and did a "last call" for this machine two
years ago.
I personally feel it is financially irresponsible and a waste of scarce
capital to fund transfers to reel tape today. The money spent doing this
could be better spent transferring to a permanent digital archive.
The use of CDs is the best stopgap we have before permanent digital
archives become pervasive, affordable, and robust for the average
archive user (i.e. not requiring a dedicated IT department). CD players
because of their huge penetration--more by far than reel tapes ever
were, I suspect--will be around for years to come.
These are my opinions only.
Cheers,
Richard
At 04:41 PM 6/25/2003 -0400, Watsky, Lance wrote:
>The question actually posted specifically asked about long term
>preservation. The problem is not whether or not CD's and DVD's will
>last for posterity, but if the players will still be around in the
>future. Although digital and optical media is wonderful for providing
>access, I believe that the Library of Congress still promotes utilizing
>reel-to-reels to serve as their preservation copies. Can someone please
>correct me, if I am wrong.