I've just read the new CLIR report, "Capturing Analog Sound for Digital Preservation." You can find it at
You can print it out.
I've not yet completed the appendices, but gotta get my big typing finger going some comments on the report proper...
I agree with almost all of it. However,
p. 8. Omitted . Initial play of tape at slow speed to discharge surface magnetic accumulation which, I was told, will reduce print-thru. At the same time, the tape should be checked for old splices.
p. 9. 1st para. "In the absence of a lead-in groove.." Play the end of the record backwards, invert the soundfile in the computer, and edit to its proper place in the full file. This is sometimes an effective technique when the groove is deeply scratched.
p. 10. 1st bullet. Warning! Do not use dubs of alignment tapes.
p. 12. Para 2. Automated transfers. This is in conflict with the recommended best practice that all damaged splices be repaired before transfer (p. 6, last bullet.) The only way the p. 12 automated transfer can work is if you are willing to accept a certain loss of tape information when the spice pulls rather than breaks, or risk loosing tape when the splice spearates and the tape recorder slaps the spae around while it comes to a stop.
p. 14. Research and development. Baking is said to be destructive. Is there any real (reel?) proof of this?
p. 15. First bullet. Gerald Gibson prepared a huge glossary of recordiing-relared terms in many languages for IASA when he was at the Library of Congress. I don't know if it ever saw publication, but I have a late draft. It's terrific!
p. 16. Add to bulleted list: training of specialists in the repair and maintainence of obsolete audio formats. Build a library of service manuals and, get the necessary permissions, and distribute it on CD-R or DVD.
They now recommentd 24/96 since this seems to be an emeging standard for DVD audio encoding, now endorsed by IASA. I'd like to know more about this.
It's a fun read. Enjoy.