FADING FILES OR FAULTY FACTS
Congressional Audio Survey Leading Us Astray?
(From Radio Recall, December 2010)
A lengthy study, co-authored by Sam Brylawski and Rob Bamberger on the preservation of sound recordings in the U.S., has concluded that new digital recordings of events in U.S. history and early radio shows are at risk of being lost much faster than older ones on tape. It also found many historical recordings already have been lost or can't be accessed by the public. That includes most of radio's first decade from 1925 to 1935.
The study was mandated by Congress in a 2000 preservation law, and was recently released to the media by the Library of Congress. According to this survey, recent history — such as recordings from 9/11 or the 2008 election — are at risk because digital sound files can be corrupted, and widely used CD-R discs only last three to five years before files start to fade.
As might be expected, this audio file survey hit the OTR hobby like a massive bomb. “The sky is falling” trumpeted the print media, as the claims of Brylawski and Bamberger were circulated throughout the country.
To gauge the accuracy of the main theme and specific allegations set forth in this survey, RADIO RECALL enlisted the services of Jerry Haendiges, a sound engineer in southern California who has decades of experience in audio sound files and duplication. He is the co-founder of SPERDVAC and a specialist in archiving vintage broadcasts.
Below are Jerry’s responses and observations on the pertinent points raised in this survey (which points are printed in bold italics.)
New digital recordings of events in U.S. history and early radio shows are at risk of being lost much faster than older ones on tape and many are already gone.
I totally disagree with most of the findings made in this survey and I'm speaking as an audio engineer with over 50 years of heavy experience in both analog and later digital audio and video. I don't think any of the conclusions reached in this survey actually came from the authors’ experience, but rather from hearsay from other people. As far as I'm concerned, digital archiving is absolutely the superior method. Now when I mention digital audio, I'm referring to raw Wave file recordings, not to compressed formats such as mp3, which are far more susceptible to corruption.
Even recent history — such as recordings from 9/11 or the 2008 presidential election — are at risk because digital sound files can be corrupted, and widely used CD-R discs only last three to five years before files start to fade.
This is pure hogwash! I've heard this argument beginning some ten years ago. I began recording to CD-R in early 1997. Every single disc I recorded back then is still quite playable today. Then, a few months later, in an effort to try to save some money on the the discs that originally cost about $2 each, I started looking around for something cheaper. Like so many others, I started using bargain basement no-brand discs. My first clue that this was not a good thing should have been the rather high reject rate. But I just wrote that off as being a lower quality control process.
Fortunately it didn't take long for me to realize that the dye used was highly important and really did make a difference in the longevity of the discs. Yes, I did have a lot of discs go bad due to the dye "fading" for that period of time. So I began looking into the dyes used by the various manufactures. I would then buy several brands of CD-r discs, record them and then hold them up to a very strong flood lamp. Some of the discs started displaying the fading problem in a couple of hours, many showed the problem in several hours, and a few showed little or no effect even after several days and then months. One of the brands that displayed no effect...ever, was Ritek. So I began using Ritek exclusively in 1999 and have done so ever since. No disc that I have checked since that time has ever been bad, and that's eleven years, not three to five as stated in the article.
Shows by singers Duke Ellington and Bing Crosby, as well as the earliest sports broadcasts, are already gone. There was little financial incentive for such networks as CBS to save early sound files.
That's because they were never digitized. :-)
Digital files are a blessing and a curse. Sounds can be easily recorded and transferred and the files require less and less space. But they must be constantly maintained and backed up by audio experts as technology changes. That requires active preservation, rather than simply placing files on a shelf.
Why in the world is this a problem? Of course files need to be backed up constantly. But huge storage devices are available at dirt cheap prices. And excellent software is readily available at a low price...or even free that will do the backups automatically. I store all of my files in one and two terabyte hard drives, each with a mirrored hard drive that contains the exact same information as the original. The hardest part of setting this up was plugging in the USB cable to my ten-port USB hubs. Yes, hard drives do go bad, but using this method, all I have to do is swap the bad hard drive with its assigned backup and plug a new hard drive in it's place. Now the software backs up onto it's new drive...again, all automatically.
Although I am an engineer, this in no way requires any technical skill to do it...any one can! By the way, there is one other inaccuracy that statement. "Sounds can be easily recorded and transferred and the files require less and less space." Wave files take up exactly that same amount of space they did at their inception. The difference is that due to higher capacity storage devices, many, many more files can be stored on a single device.
Those old analog formats that remain are more physically stable and can survive about 150 years longer than current digital recordings. Still, the rapid change in technology to play back the recordings can make them obsolete.
I don't know how anyone with an ounce of experience could make that kind of irresponsible statement! In my restoration business, I deal routinely with studio master tapes and transcription discs. Most recording tape prior to 1970 was made from acetate or, prior to that, paper. As anyone using these tapes knows, you kind of hold your breath when opening an acetate tape box. First of all, you are greeted with the strong small of vinegar (acetone). Then you look to see, not whether or not the tape is curled or warped, but just how badly it's warped. And if you are lucky enough to be able to play the tape, you have to have the splicing block, tape and razor blade handy for when the inevitable break in the tape occurs. And usually, if you can get the tape to play, you need to have some way to hold the tape tight to the head, because of the warping.
Paper tape doesn't warp, but like any paper product, it dries out and becomes brittle, resulting in many breaks during playing. And for the same reasons stated above, many people would later use bargain basement tape, which started going bad in just a few years. So, best case scenario, under ideal conditions, recording tape can be expected to last maybe thirty or forty years. A far cry from the 150 years plus stated in the survey. And getting back to the loss of all the programs in the 1925 to 1935 era, could it just possibly be that all those transcription discs didn't quite make it to the 150 year mark either.
Recordings saved by historical societies and family oral histories also are at risk. Audio cassettes are just time bombs; they're just not going to be playable.
Well, finally we do totally agree on something; cassettes are probably the worst method of archiving there is. It's one of the few things you can positively know will go bad eventually.
There are few if any programs to train professional audio archivists, the study found. No universities currently offer degrees in audio preservation, though several offer related courses.
This is nonsense. Anyone really interested in professional archiving can easily find many good resources. I've trained a few myself and all top professional software companies, that I know of, offer very extensive training. It certainly does not require a university class nor a degree.
The biggest problem that I see in today’s archiving is that anyone with a computer and some sort of audio program thinks they are an expert in audio preservation and restoration. And sometimes they make it even worse. After over-processing with cheap equipment and programs, they end up "archiving" this already terrible sounding audio in low bit-rate mp3 format.