Let us say that 35,000 CDs have been released every year since 1980, and will continue to be released.
Let us also say that a CD can generally be represented as a 100MiB MP3 file at an acceptable bit rate.
From the ‘back of envelope’ table below we can estimate that the cost of storing all the CDs ever released on a hard disk drive will fall to about $100 in 2015.
I suggest that the next file sharing application won’t be one that lets people pick and choose which CDs to share or audition. It will simply replicate and distribute EVERYTHING. There won’t even be any point in deleting all the CDs one doesn’t like. The problem will be entirely one of deciding what the heck to listen to.
Even so, once we have discovered the musicians we like there will still be the problem of how to persuade them to make more great music. Even all the music ever released can pale next to one more album from a favourite artist.
Don’t worry. That’s the problem I’m working on – enabling a musician’s fans to exchange their money for the musician’s production of music. It’s not rocket science. You just have to bear in mind that it’s not about enabling CD manufacturers to sell copies, but about enabling musicians to sell their music to their audience – directly instead of via record labels and CD manufacturers.
When you’re selling music instead of digital copies you don’t benefit from a monopoly; on the contrary, you want your music to spread far and wide.
| Year |
Price of 1TiB HDD |
CDs |
MP3 TiB |
Storage cost |
| 1980 |
$1,336,434,513.25 |
35,000 |
3 |
$4,460,831,447.94 |
| 1981 |
$735,038,982.29 |
70,000 |
7 |
$4,906,914,592.74 |
| 1982 |
$404,271,440.26 |
105,000 |
10 |
$4,048,204,539.01 |
| 1983 |
$222,349,292.14 |
140,000 |
13 |
$2,968,683,328.61 |
| 1984 |
$122,292,110.68 |
175,000 |
17 |
$2,040,969,788.42 |
| 1985 |
$67,260,660.87 |
210,000 |
20 |
$1,347,040,060.36 |
| 1986 |
$36,993,363.48 |
245,000 |
23 |
$864,350,705.39 |
| 1987 |
$20,346,349.91 |
280,000 |
27 |
$543,306,157.68 |
| 1988 |
$11,190,492.45 |
315,000 |
30 |
$336,170,685.06 |
| 1989 |
$6,154,770.85 |
350,000 |
33 |
$205,437,640.87 |
| 1990 |
$3,385,123.97 |
385,000 |
37 |
$124,289,772.73 |
| 1991 |
$1,861,818.18 |
420,000 |
40 |
$74,573,863.64 |
| 1992 |
$1,024,000.00 |
455,000 |
43 |
$44,433,593.75 |
| 1993 |
$563,200.00 |
490,000 |
47 |
$26,318,359.38 |
| 1994 |
$309,760.00 |
525,000 |
50 |
$15,509,033.20 |
| 1995 |
$170,368.00 |
560,000 |
53 |
$9,098,632.81 |
| 1996 |
$93,702.40 |
595,000 |
57 |
$5,317,013.55 |
| 1997 |
$51,536.32 |
630,000 |
60 |
$3,096,378.63 |
| 1998 |
$28,344.98 |
665,000 |
63 |
$1,797,619.69 |
| 1999 |
$15,589.73 |
700,000 |
67 |
$1,040,727.05 |
| 2000 |
$8,574.35 |
735,000 |
70 |
$601,019.69 |
| 2001 |
$4,715.90 |
770,000 |
73 |
$346,302.24 |
| 2002 |
$2,593.74 |
805,000 |
77 |
$199,123.51 |
| 2003 |
$1,426.55 |
840,000 |
80 |
$114,279.38 |
| 2004 |
$784.61 |
875,000 |
83 |
$65,472.90 |
| 2005 |
$431.53 |
910,000 |
87 |
$37,450.41 |
| 2006 |
$237.34 |
945,000 |
90 |
$21,389.85 |
| 2007 |
$130.54 |
980,000 |
93 |
$12,200.23 |
| 2008 |
$71.79 |
1,015,000 |
97 |
$6,949.38 |
| 2009 |
$39.49 |
1,050,000 |
100 |
$3,953.91 |
| 2010 |
$21.72 |
1,085,000 |
103 |
$2,247.35 |
| 2011 |
$11.95 |
1,120,000 |
107 |
$1,275.91 |
| 2012 |
$6.57 |
1,155,000 |
110 |
$723.68 |
| 2013 |
$3.61 |
1,190,000 |
113 |
$410.09 |
| 2014 |
$1.99 |
1,225,000 |
117 |
$232.18 |
| 2015 |
$1.09 |
1,260,000 |
120 |
$131.35 |
| 2016 |
$0.60 |
1,295,000 |
124 |
$74.25 |
| 2017 |
$0.33 |
1,330,000 |
127 |
$41.94 |
| 2018 |
$0.18 |
1,365,000 |
130 |
$23.67 |
| 2019 |
$0.10 |
1,400,000 |
134 |
$13.35 |
| 2020 |
$0.06 |
1,435,000 |
137 |
$7.53 |
So, if you’re hoping to fill that hard disk you’d probably better get started today.
Assuming a conservative 20Mbps share rate (given an efficient file-sharing system and no network contention) that works out at around 75TiB per year. In other words, all music ever released could be shared via the successor to BitTorrent within two years at such time as it became economic for everyone to store a duplicate set.
Within a decade, those who don’t share published music will be seen as a burden upon everyone else, akin to the way leechers are already perceived today.
Hi Crosbie. I too have anticipated the possibility of storing every song ever produced on disk. I have been slowly amassing music in preparation for it. The fastest way of sharing music these days is by swapping hard disks, and that’s mainly how I have come to own a very large collection of music. The thought that this practice could be merged with new sharing technologies had occurred to me. Perhaps devices connected to one another via a wireless mesh, constantly uploading and downloading to one another. The technology to do this is already possible. I imagine it is only a matter of time before somebody does.
Comment #000287 at
2009-05-29 15:30
by
Some guy
Yup, it’s just a matter of time.
I had expected FreeNet to fill this gap, but if they don’t no doubt someone else will. Perhaps even Google? Wave perhaps?
A spindle of ‘BluRay Super+’ HD-DVDs will no doubt suffice for those too impatient to download.
Comment #000288 at
2009-05-29 15:45
by
Crosbie Fitch
If the data in your table holds up to be true, and providing some people have 2.4gb/s connection speeds (optimistic, I know), then it would take little over a day to download 120tb of music in 2015. A lot of ifs. But enough to make you think.
Comment #000289 at
2009-06-01 14:39
by
Some guy
The data is cobbled together from very quick web searches and 'back of the envelope' calculations. It is intended only as food for thought. I am confident others can provide more accurate data, and I’d look forward to reading it (and their derivatives of my article).
Comment #000290 at
2009-06-01 15:33
by
Crosbie Fitch
Well, still, what’s the point of having access to that much music? Think of all the electricity power needed to do such a thing….
Comment #000317 at
2009-09-02 08:36
by
Let us say that 35,000 CDs have been released every year since 1980, and will continue to be released.
Let us also say that a CD can generally be represented as a 100MiB MP3 file at an acceptable bit rate.
From the ‘back of envelope’ table below we can estimate that the cost of storing all the CDs ever released on a hard disk drive will fall to about $100 in 2015.
I suggest that the next file sharing application won’t be one that lets people pick and choose which CDs to share or audition. It will simply replicate and distribute EVERYTHING. There won’t even be any point in deleting all the CDs one doesn’t like. The problem will be entirely one of deciding what the heck to listen to.
Even so, once we have discovered the musicians we like there will still be the problem of how to persuade them to make more great music. Even all the music ever released can pale next to one more album from a favourite artist.
Don’t worry. That’s the problem I’m working on – enabling a musician’s fans to exchange their money for the musician’s production of music. It’s not rocket science. You just have to bear in mind that it’s not about enabling CD manufacturers to sell copies, but about enabling musicians to sell their music to their audience – directly instead of via record labels and CD manufacturers.
When you’re selling music instead of digital copies you don’t benefit from a monopoly; on the contrary, you want your music to spread far and wide.
So, if you’re hoping to fill that hard disk you’d probably better get started today.
Assuming a conservative 20Mbps share rate (given an efficient file-sharing system and no network contention) that works out at around 75TiB per year. In other words, all music ever released could be shared via the successor to BitTorrent within two years at such time as it became economic for everyone to store a duplicate set.
Within a decade, those who don’t share published music will be seen as a burden upon everyone else, akin to the way leechers are already perceived today.