Slashdot Mirror
The Grammy In Mathematics
An anonymous reader writes "A mathematician will receive a Grammy award for restoring the only known recording of a live Woody Guthrie performance — a bootleg someone made in 1949 using a wire recorder. Guthrie's daughter, who had never heard her father perform in front of a live audience, oversaw the restoration. The article links very cool before and after clips."
To all those who like to argue against the ongoing use of analog recording mediums for original masters, let this be a lesson to you.
Always record your originals in analog and immediately transfer to digital, and one day you may find that more of the original sonic environment can be recovered from that master than you ever thought possible through the progression of physics, chemistry and math.
-1 Uncomfortable Truth
because featuring two aif's on slashdot is clearly not going to go well.
I work for the Department of Redundancy Department.
Congratulations, "A Mathematician"!!
How awesome is that, to do some really interesting work, and finally get some world-wide recognition and even get your name on the front page of Slashdot!
Oh, wait...
Common people, let's give credit where credit is due. Thanks. The guy's name isn't even mentioned until the 11th paragraph of the story! Somehow when it's something cool like this it's enough to say, "mathematics did it!", as if this restoration technique of identifying the hum of a 1949 power supply to help guide a dynamic warping and interpolation technique just dropped out of thin air.
(It's Kevin Short by the way, although if I understand the article, this sound engineer Jamie Howarth played a large part as well.)
See also this famous thread
The link points to where Steve Albini enters, the next few pages attempt to hammer the long term storage argument home. For those who can't be bothered to read it, SA's contention is that the cost of maintaining digital archives is prohibitively expensive when compared to the cost of storing tape.
1) Things that we believe we can't distinguish now, we may demonstrate that we can distinguish in the future. Just because you can't tell the difference consciously when you listen to two samples doesn't mean that some subconscious part of your brain can't determine a difference. We cannot rule out subsonics, subliminal effects, and so on.
... here it comes ... uplifted dolphins? (This is really just a sensationalist version of #2: "applications we haven't thought of yet".)
2) There are technologies that would benefit from having more information available. Imagine being able to extract enough information from a recording to simulate that vocalist singing something else. Heck, for an example of a technology that benefits from much fancier recordings than some people ever thought they would need, consider the game "Rock Band". You can't (today) use a master recording in Rock Band unless each drum in the drum kit has a separate recording track. This is why the old Rush songs in the game are covers and not masters. Almost nobody imagined they'd actually have a need for those more detailed recordings, but now we do. (I say "you can't today" because the software to de-mix the drums isn't advanced enough yet. Once it is advanced enough... we may determine that common digital recordings aren't as good for this purpose as straight-up analog recordings!)
3) This is the far-out one -- go ahead and warm up your mockery engines... what about superhuman hearing? Are you sure that, by technology (biotech, cybernetics, whatever), human hearing won't ever be improved? What about
What format is this data going to be in; raw PCM, big-endian, little-endian? On what tape format will it be stored and how do you interface with a computer 75 years from now?
Look how many of the digital tape formats are dead and then go compare with the obsolete analog formats that can still be successfully replayed/transferred. You might want to read the thread I linked and think about the problem in real world terms, just as people who own professional recording facilities do.
I am intrigued by the fact that this was a bootleg recording. It just goes to show how short-sighted modern performers are. They actively try to prevent what may later be considered a valuable part of America's music heritage.
There is no problem with aliasing on correctly mastered CDs, as they are supposed to be low pass filtered at 20KHz. The extra 2KHz overhead is so you can have a shallow enough rolloff that no perceptible distortion is introduced.