The Sound of a Fermi Gamma-ray Burst
June 22, 2012 10:02 AM Subscribe
A gamma-ray burst, the most energetic explosions in the universe, converted to music. What does the universe look like at high energies? Thanks to the Fermi Large Area Telescope (LAT), we can extend our sense of sight to "see" the universe in gamma rays. But humans not only have a sense of sight, we also have a sense of sound. If we could listen to the high-energy universe, what would we hear? What does the universe sound like?
My God! It's full of sitars!
posted by It's Raining Florence Henderson at 10:25 AM on June 22, 2012 [3 favorites]
posted by It's Raining Florence Henderson at 10:25 AM on June 22, 2012 [3 favorites]
What does the universe sound like?
For a very different (and eerier) take: Symphonies of of the Planets.
More from NASA.
posted by ryanshepard at 10:49 AM on June 22, 2012
For a very different (and eerier) take: Symphonies of of the Planets.
More from NASA.
posted by ryanshepard at 10:49 AM on June 22, 2012
Related: Fiorella Tirenzi's Music From The Galaxies.
posted by Johnny Wallflower at 10:53 AM on June 22, 2012
posted by Johnny Wallflower at 10:53 AM on June 22, 2012
I find it quite soothing to listen to. Be interesting to score this for an orchestra.
posted by pdxpogo at 11:00 AM on June 22, 2012
posted by pdxpogo at 11:00 AM on June 22, 2012
Sonification has its uses, but in many cases the result has a lot more to do with arbitrary aesthetic choices than with the source of the data.
Choosing to arrange for harp, cello, and piano in a particular key carries more than enough cultural baggage to overwhelm the data. If you used the same instrumentation to present stock market data, you would encourage people to say "wow, gamma ray bursts sound like the stock market". Or if you present it the same data with sine waves instead, people associate the sound with a theremin and think it sounds like a 60's science fiction movie. The result may be interesting, but it's not what the universe sounds like - it's the aural equivalent of what Tufte calls "chartjunk".
posted by nixt at 11:06 AM on June 22, 2012 [4 favorites]
Choosing to arrange for harp, cello, and piano in a particular key carries more than enough cultural baggage to overwhelm the data. If you used the same instrumentation to present stock market data, you would encourage people to say "wow, gamma ray bursts sound like the stock market". Or if you present it the same data with sine waves instead, people associate the sound with a theremin and think it sounds like a 60's science fiction movie. The result may be interesting, but it's not what the universe sounds like - it's the aural equivalent of what Tufte calls "chartjunk".
posted by nixt at 11:06 AM on June 22, 2012 [4 favorites]
So they have pianos and orchestra's in space too? Who's up there playing them? moon men? this is the most artificial "Sonification" I've ever heard. totally meaningless.
posted by mary8nne at 11:40 AM on June 22, 2012
posted by mary8nne at 11:40 AM on June 22, 2012
Who's up there playing them?
Major Tom.
posted by It's Raining Florence Henderson at 11:48 AM on June 22, 2012 [1 favorite]
Major Tom.
posted by It's Raining Florence Henderson at 11:48 AM on June 22, 2012 [1 favorite]
So they have pianos and orchestra's in space too? Who's up there playing them?
Aliens from Planet Literal.
posted by Marisa Stole the Precious Thing at 11:51 AM on June 22, 2012 [1 favorite]
Aliens from Planet Literal.
posted by Marisa Stole the Precious Thing at 11:51 AM on June 22, 2012 [1 favorite]
I think that this is a vastly better analogy for what a gamma ray burst sounds like.
posted by "Elbows" O'Donoghue at 12:27 PM on June 22, 2012
posted by "Elbows" O'Donoghue at 12:27 PM on June 22, 2012
Great, can ISCAP* be far behind?
----
*Intergalactic Society of Composers, Authors and Publishers
posted by mazola at 12:58 PM on June 22, 2012
----
*Intergalactic Society of Composers, Authors and Publishers
posted by mazola at 12:58 PM on June 22, 2012
*Slaps DRM on Milky Way*
posted by It's Raining Florence Henderson at 12:59 PM on June 22, 2012
posted by It's Raining Florence Henderson at 12:59 PM on June 22, 2012
Yo, I got a sweet bootleg of PSR B1257+12 , msg me for a download link, peace.
posted by Marisa Stole the Precious Thing at 1:08 PM on June 22, 2012 [2 favorites]
posted by Marisa Stole the Precious Thing at 1:08 PM on June 22, 2012 [2 favorites]
In translating the gamma-ray measurements into musical notes we assigned the photons to be "played" by different instruments (harp, cello, or piano)
Sine waves or go home. Using specific instruments is totally bogus and worthless.
posted by anigbrowl at 1:14 PM on June 22, 2012
Sine waves or go home. Using specific instruments is totally bogus and worthless.
posted by anigbrowl at 1:14 PM on June 22, 2012
Sine waves or go home. Using specific instruments is totally bogus and worthless
Let them have their link bait, they aren't publishing their compositions to A&A.
posted by TwelveTwo at 2:12 PM on June 22, 2012
Let them have their link bait, they aren't publishing their compositions to A&A.
posted by TwelveTwo at 2:12 PM on June 22, 2012
That was nice, but yes, a bit too subjective for my taste. Now pulsar sounds, on the other hand, are pretty close to what the telescope actually receives - we're correcting for the sweep of the signal in frequency due to the interstellar plasma, and then converting the time series to a sound waveform with an arbitrary scaling, but that's it.
(You'll notice that the link does not include B1257+12 - an utterly outlandish pulsar with a handful of planets - and nor does the Jodrell Bank archive. For that you'll have to try Marisa Stole the Precious Thing, I suppose... )
posted by RedOrGreen at 8:01 PM on June 22, 2012
(You'll notice that the link does not include B1257+12 - an utterly outlandish pulsar with a handful of planets - and nor does the Jodrell Bank archive. For that you'll have to try Marisa Stole the Precious Thing, I suppose... )
posted by RedOrGreen at 8:01 PM on June 22, 2012
This and dawn chorus should be added to Minecraft's ambient music library.
posted by speedo at 9:21 PM on June 22, 2012
posted by speedo at 9:21 PM on June 22, 2012
Gamma ray bursts are probably best listened to at a considerable remove. They've been proposed as a resolution of the Fermi paradox.
posted by 0rison at 11:45 PM on June 22, 2012
posted by 0rison at 11:45 PM on June 22, 2012
I think we can draw analogies to graphing conventions here.
Using a specific key signature, or instrumentation could be compared to using a limited color palette in a graph, which is perfectly acceptable as long as a viewer is not expecting color to be informative in any way other than distinguishing data sources. Or using fixed width bars in a bar graph, as long as the viewer does not expect the width of the bars to be meaningful.
Here, the important data, in order of relative importance, are: (density*time), quality, frequency. In order of the number of states each variable can hold we have: (density*time), frequency, quality. So they use time and density over time quite literally with minimal quantization, they use instrumentation for a variable with only three states that we want to be easily distinguishable, and frequency for a variable with a wide continuum of states. The mapping between the amount of information to be conveyed by each variable and the amount of information conveyable by its musical counterpart is quite well done.
The real problem is that relative pitch and instrumentation are not what most listeners will notice about it. I casual listener will often just hear the tempo (totally arbitrary, though change of tempo is meaningful) and the mood of the piece (totally haphazard - the key signature is selected arbitrarily beforehand and quantizes the data, and changes of harmonic feeling are unlikely to say anything interesting about the data set).
But this is as much the listener's fault as the sonifier's. If we desire to meaningfully convey information in a musical format, we need to expect that the listener will meet the music half way. If you look at a pie chart, and your main takeaway is that the colors were clashing or complimentary, that probably because you don't understand pie graphs. The best practice with sonification is to try to understand the "key" (not the key-signature of the musical piece, but the mapping from data to sound) before letting yourself listen.
If you don't have the ear to hear the properties the key encodes (here density, time, relative pitch, instrumentation) without being distracted by properties that are fixed with no relevance to the data (amplitude, key signature, European instrumentation, MIDI instruments and all that entails), know that you simply won't derive useful information from the sonification. I think these people did a fine job of preventing the last kind of property: the ones that vary freely with no relevance to the data (composers that dabble with sonification are very bad about this, because they compulsively want to make things "more musical", I would compare that to "graph junk" where a grapher fudges data to make the graph more visually appealing, and that seems to have been mostly avoided here.
posted by idiopath at 7:35 AM on June 23, 2012 [2 favorites]
Using a specific key signature, or instrumentation could be compared to using a limited color palette in a graph, which is perfectly acceptable as long as a viewer is not expecting color to be informative in any way other than distinguishing data sources. Or using fixed width bars in a bar graph, as long as the viewer does not expect the width of the bars to be meaningful.
Here, the important data, in order of relative importance, are: (density*time), quality, frequency. In order of the number of states each variable can hold we have: (density*time), frequency, quality. So they use time and density over time quite literally with minimal quantization, they use instrumentation for a variable with only three states that we want to be easily distinguishable, and frequency for a variable with a wide continuum of states. The mapping between the amount of information to be conveyed by each variable and the amount of information conveyable by its musical counterpart is quite well done.
The real problem is that relative pitch and instrumentation are not what most listeners will notice about it. I casual listener will often just hear the tempo (totally arbitrary, though change of tempo is meaningful) and the mood of the piece (totally haphazard - the key signature is selected arbitrarily beforehand and quantizes the data, and changes of harmonic feeling are unlikely to say anything interesting about the data set).
But this is as much the listener's fault as the sonifier's. If we desire to meaningfully convey information in a musical format, we need to expect that the listener will meet the music half way. If you look at a pie chart, and your main takeaway is that the colors were clashing or complimentary, that probably because you don't understand pie graphs. The best practice with sonification is to try to understand the "key" (not the key-signature of the musical piece, but the mapping from data to sound) before letting yourself listen.
If you don't have the ear to hear the properties the key encodes (here density, time, relative pitch, instrumentation) without being distracted by properties that are fixed with no relevance to the data (amplitude, key signature, European instrumentation, MIDI instruments and all that entails), know that you simply won't derive useful information from the sonification. I think these people did a fine job of preventing the last kind of property: the ones that vary freely with no relevance to the data (composers that dabble with sonification are very bad about this, because they compulsively want to make things "more musical", I would compare that to "graph junk" where a grapher fudges data to make the graph more visually appealing, and that seems to have been mostly avoided here.
posted by idiopath at 7:35 AM on June 23, 2012 [2 favorites]
And as I consider this more, I think there is a human problem here. We expect music to be taken in on an instinctive level, and speak mainly to emotions. This clashes oddly with the task of conveying information. Whether this human problem is primarily cultural or biological is a whole other can of worms.
posted by idiopath at 7:42 AM on June 23, 2012
posted by idiopath at 7:42 AM on June 23, 2012
« Older This is what the Internet is for. | Peter Gentenaar: Sculptor in Paper Newer »
This thread has been archived and is closed to new comments
OK, this is seriously cool and I really want to see this done with other celestial phenomena. I wonder, for example, if a supernova would sound like Liz Phair.
posted by Marisa Stole the Precious Thing at 10:15 AM on June 22, 2012