Re: [AV Media Matters] Digitizing Audio and Video-ReallyBigPost

[Richard Hess <arclists@richardhess.com>]

Scott and Jim,

I'm sorry I used the word "promoting" I wasn't trying to be negative or 
offensive. It is true that the work that I'm currently doing in digital 
archives is being funded by a major media corporation and they see the 
value of this particular archive as a long-term resource.

I apologize for not be inclusive when I talked about the smaller
archives. 
Yes, all archives that are non-commercial seem to be underfunded.
Perhaps 
even the commercial ones are also underfunded.

Jim, I don't think we're picking on you! Scott had lots of smilies in
his 
post. I, for one, appreciate your insight and enthusiasm for this.
You're 
not wrong where you're headed. I just don't want to lose sight of some
of 
the other issues, and the people I deal with don't yet see the benefits
of 
regional repositories.

We're frustrated at the change of paradigm from stick-it-on-the-shelf to

active management.

Certainly Eddie Zwaneveldt (What ever happened to him, he seems to have 
dropped off AMIA-L) has had some harsh things to say about the
constantly 
migrating plan that we seem to be headed for.

In another post on this list, someone asked about how long before the 
migration. I'd like to address some of the mechanisms that trigger
migration.

(1) Migration is required when the format is declared "end of life" by
the 
format vendor(s). This does not mean that in a roadmap-based growth path

(such as LTO, DLT, AIT, and S-AIT) that you have to do it each
generation, 
as long as the roadmap includes backwards compatibility for read.

(2) Migration is a good idea at the steps in the roadmap as these steps
are 
usually doubling the storage. Therefore, the robotic (or shelf system) 
storage capacity is doubled each time you migrate at a roadmap
milestone.

(3) On tapes that are used, they are monitored for bit error rate and
other 
performance parameters. When the error rate exceeds a certain threshold,

then the tape is cloned, or migrated.

(4) Tapes that are not used should be read at some interval and the
rules 
of (3), above, applied during that read.

(5) There should be an absolute number of load/unload and read cycles
that 
generate a clone/migration process.

In any event, I THINK current wisdom is check all your tapes every five
years.

As an aside, someone brought me an old NAB broadcast cartridge that had
3.5 
minutes of audio on it. Assuming it was stereo and CD quality (that was 
generous on both accounts), when we laid it on top of the S-AIT-1
cartridge 
sample I have on my desk, I did a quick calculation and, giving the NAB 
cartridge the very generous bit density of CD audio, in approximately
the 
same volume we had a 12,500 increase in storage. The same cartridge can 
hold 8 hours of broadcast quality video. It's 500GB ( 0.5TB) in an 
LTO/DLT/SAIT sized cartridge now with a roadmap to 4.0TB in about six
years.

Cheers,

Richard

==^================================================================
This email was sent to: waiscool@cool.conservation-us.org

EASY UNSUBSCRIBE click here: http://topica.com/u/?aaP8EW.b4jznQ.d2Fpc2Nv
Or send an email to: AV-Media-Matters-unsubscribe@topica.com

TOPICA - Start your own email discussion group. FREE!
http://www.topica.com/partner/tag02/create/index2.html
==^================================================================