Re: arsclist Digital knowledge preservation

["George Brock-Nannestad" <pattac@xxxxxxxx>]

From: Patent Tactics, George Brock-Nannestad

Hello,

fortunately this thread has not stopped, because as long as it is 
live, it generates awareness, and we need that.

Our problems are caused by audio recordings being really audio 
systems - we cannot reproduce anything without proper equipment.

Some storage means are so simple that watchmakers would at 
any time in the future be able to manufacture replay equipment. 
The commercially manufactured type of this kind of media also 
turns out to be very long-lasting in storage. 

Some storage means are so complex that we need very skilled 
factories cooperating to provide replay equipment. Analog storage 
may be much more difficult to handle than digital - think of Dolby 
SR or even Dolby C. However, even primitive, non-signal processing 
replay of such material will provide a good impression of what is on 
the tape, whereas digital output from digital storage may be 
complete gibberish without proper decoding. 

The equipment we use for digital replay consist of two parts: the 
readout device providing raw data of some kind of digital (0s and 
1s) output, and the decoder which is frequently just one component 
- hardware - but which may be emulated by software. In particular 
because PCs are so fast now, decoding by software may be 
relevant. This removes some of the pressure on maintaining well-
functioning replay equipment, we only have to have well-functioning 
readout. That is a formidable task in itself. However, as long as 
there is some kind of digital representation, it will be possible to 
copy it to other storage means, and decoding can take place. 
Obviously the coding used and the software used for PC decoding 
must be open source.

There is no doubt that spreading the physical locations where 
material is stored increases its chance of long time survival. 
However, most material will not benefit from this approach. 
UNESCO estimates that 200.000.000 hours of audio material may 
be in need of preservation (try the conversion to years of material 
for a perspective!). In my view, only a minor part of that constitutes 
the total commercial output, the remainder is audio history, i.e. 
sources to the many activities which it was possible to document 
by audio. Apart from the tremendous radio broadcast and TV sound 
output, they would be tapes of meetings in local councils, oral 
history, local visits by celebrities to schools and many, many 
routine recordings that may contain answers future citizens (not 
just researchers or collectors) will want to ask to it, which would 
never benefit from a transfer programme. I am reminded of Edward 
Tatnall Canby's description in the 1970s (Audio Magazine) of the 
Columbia Oral History Project, where the tapes were reused after 
transcribing. Even the lousiest quality tape copy would have given 
more to today's and future users!!

Hence, even storing on data-reduced formats is better than no 
storage at all - the criterion must surely be that it must be easy to 
transfer before the dedicated replay machine (in particular the 
readout part of it) breaks down. The MiniDisc 5-hour format is 
formidable in terms of ear-quality and storage compactness, but I 
do not think it fulfils the open source requirement yet. Incidentally, 
have you considered that heavily data-reduced formats are more 
tamper-proof than linear formats: editing and filtering will create 
audible artefacts!

So a preservation programme for society would be: establish an 
open-source data reduced format, copy as quickly as possible, but 
without too many attempts to obtain "the ideal replay" the audio 
content to a data file on your computer, and then burn a CD-R with 
the data file. This will not be an audio-coded file, but a computer file 
that your open-source programme may convert back to ear-
stimulus. The error-correction code for file-saving on the CD-R will 
protect the information for some time, so before it is too challenged 
by the errors, do re-copy it to some other carrier. Metadata will 
enter as indexing and cataloguing data; it should follow the audio 
data file, but even here a little is better than nothing. Full indexing 
just means using time now rather than at the time of searching for 
the information.

Undoubtedly there will be material which we need all our restorative 
skills to extract before it is being saved in an information-rich form. 
Fortunately we are very good, but the time factor goes from 1:1 to 
100:1. And no doubt, future users will curse us for not having used 
our skills on the material they only have the remains of (data-
reduced) in front of them. However without the mass conversion 
today, there would be NOTHING for them.

Finally, it is so ironic, that now that commercial output is not worn 
down by reproducing it, the providers are really frustrated because 
they lose sales. They could have stuck with vinyl!!

Kind regards,


George Brock-Nannestad
Preservation Tactics


-
For subscription instructions, see the ARSC home page
http://www.arsc-audio.org/arsclist.html
Copyright of individual posting is owned by the author of the posting and
permission to re-transmit or publish a post must be secured
from the author of the post.