ELL: Fwd: more on how to archive your language etc.

Brian Levy xernaut at YAHOO.COM
Tue Oct 24 19:03:21 UTC 2000

>Here are some thoughts on an approach to your problems.
>Here and Now. I think the sad fact of life is that when it comes to
>permanent archiving of anything over tens of years, it is simply not
>possible to copy it somehow to some support and then put it in a vault and
>forget about it. Tape and current storage technologies do not permit this
>level of security and I doubt if anything is coming in the next 20 years
>that will adequately address the problem (I hope that I am proved wrong).
>That's the bottom line.
>Talk about archive formats addresses important issues of data formats and
>supports: the general objective is to find a stable support which will last
>a long time and not to unacceptably compromise your quality by choosing a
>good method of recording the data onto that support.
>This has to be the starting point, and archive preservation becomes as a
>consequence not only a technical matter but 50% also a management strategy,
>requiring regular and planned intervention, and anticipating rather than
>reacting to problems.
>It must take into account the need to migrate to new supports regularly, say
>every 10 years (longer with luck).
>It must use the most appropriate technology available now within your
>budget, and meeting your quality criteria. Don't wait for tomorrow.
>It must take into account the possibility of physical damage by fire,
>flooding or machine malfunction.
>It must also take into account that no single support can be guaranteed to
>last indefinitely - accelerated testing and manufacturers claims apart -,
>there is the additional danger of unforeseen modes of deterioration.
>There must be planned continuous quality control, on the preservation
>copies, with appropriate measurements of error rates or drop outs, so that a
>good percentage of the copies are checked every (say) 3 years. (There are
>books and methods in management and systems courses to plan this - it need
>not be heavy to be effective).
>Chose good storage systems.
>Keep a thorough database.
>Use separate work copies for routine access and avoid touching the
>preservation copies.
>There must be a regular planned assessment of the technologies being
>actually used and assessments of the advantages of any new technologies.
>If you have got material now, start now, using the best techniques available
>now. Don't plan on the speculation and hopes of others becoming reality.
>Part of any preservation strategy then must be to hold two preservation
>copies, each on a different support, and that these be stored in separate
>Migration in digital is more transparent and technically easier: the problem
>with digital is being sure that the players are around when you want to
>migrate. This you address in your management planning.
>Make clear as well the distinction between how you encode the audio and
>video, and how you then store it. In the IT world, you can change storage
>medium without changing the encoding - transparently.
>Since you are originating video at 'domestic' quality, I would suggest that
>digitising your video using a good quality capture card in a PC and bit
>rates of a few Mbps - or buy the systems for transferring the DV stream from
>the camera tape direct to disc: this will preserve essentially the original
>quality of the video from a domestic DV camera you seem to be using. Perhaps
>a higher bit rate is appropriate for SP and other 'professional' recordings.
>The hardware is not too expensive. If you use one of the recognized subsets
>of the MPEG2 formats, (see ebu/smpte report on harmonization) as used in DVB
>and ATSC, then you should be reasonably future proofed for a while.
>This has got your video as far as a hard disc. Now look to how you are going
>to store it.
>Burning a DVD either as a DVD-Video or DVD-R is one option. I have my doubts
>about the longetivity of recordable CDRs, and the JITS2000 archive
>conference in Paris earlier this year went into a lot of detail about the
>problems with it (I have copies if you want). I imagine DVD will be much the
>same. Nevertheless it is a good option for one of the preservation copies. A
>second copy kept as computer files on (say) a local server with RAID discs
>and/or on storage space rented over the internet seems a good option and
>reasonably secure. The details depend on the local market in your part of
>the world.
>Having got this far, additional storage on audio or video tape is a step
>backwards, and I do not recommend it. Another problem with this is that you
>have to go to the professional video formats which are expensive machines
>for the quality of video you are now originating. The IT solutions suggested
>are correct for domestic digital cameras.
>A similar approach can be used for audio: a CDR (as now) and server and/or
>internet storage is also interesting.
>An advantage of the IT approach is that your material is already there on a
>server if you are planning putting it on the web.
>IBM and Sony (amongst others I am sure) also have services for storing and
>accessing audio and video via a private or public web type front end. You
>are then paying for a service rather than having to manage the storage
>technology yourself. I cannot say if this is more economic.
>Concerning the debate about analogue versus digital: I accept that an
>analogue machine for audio or video is probably more easily re-constructed
>and mad-to-work than a digital machine. However - if you get your management
>strategy in place, you are addressing the problem of obsolescence by
>regularly assessing the state of the technological market and state of your
>recordings. It is a proactive strategy.
>I hope that this is of some help.
>Best Wishes
>Tony Gardner
>European Commission AV Service (Europe)
>All opinions are my own.

Endangered-Languages-L Forum: endangered-languages-l at carmen.murdoch.edu.au
Web pages http://carmen.murdoch.edu.au/lists/endangered-languages-l/
Subscribe/unsubscribe and other commands: majordomo at carmen.murdoch.edu.au

More information about the Endangered-languages-l mailing list