June 24, 2024

Infinite Storage for Music

Last week I spoke on a panel called “The Paradise of Infinite Storage”, at the “Pop [Music] and Policy” conference at McGill University in Montreal. The panel’s title referred to an interesting fact: sometime in the next decade, we’ll see a $100 device that fits in your pocket and holds all of the music ever recorded by humanity.

This is a simple consequence of Moore’s Law which, in one of its variants, holds that the amount of data storage available at a fixed size and price roughly doubles every eighteen months. Extrapolate that trend and, depending on your precise assumptions, you’ll find the magic date falls somewhere between 2011 and 2019. From then on, storage capacity might as well be infinite, at least as far as music is concerned.

This has at least two important consequences. First, it strains even further the economics of the traditional music business. The gap between the number of songs you might want to listen to, and the number you’re willing and able to pay a dollar each to buy, is growing ever wider. In a world of infinite storage you’ll be able to keep around a huge amount of music that is potentially interesting but not worth a dollar (or even a dime) to you yet. So why not pay a flat fee to buy access to everything?

Second, infinite storage will enable new ways of building filesharing technologies, which will be much harder for copyright owners to fight. For example, today’s filesharing systems typically have users search for a desired song by contacting strangers who might have the song, or who might have information about where the song can be found. Copyright owners’ technical attacks against filesharing often target this search feature, trying to disrupt it or to exploit the fact that it involves communication with strangers.

But in a world of infinite storage, no searching is needed, and filesharers need only communicate with their friends. If a user has a new song, it will be passed on immediately to his friends, who will pass it on to their friends, and so on. Songs will “flood” through the population this way, reaching all of the P2P system’s participants within a few hours – with no search, and no communication with strangers. Copyright owners will be hard pressed to fight such a system.

Just as today, many people will refuse to use such technologies. But pressure on today’s copyright-based business models will continue to intensify. Will we see new legal structures? New business models? Or new public attitudes? Something has to change.


  1. A lot of spam on this blog post; you should get busy deleting it…

    [I’ve done some cleaning. See the following post about this new type of spam. — Ed]

  2. The future has to be subcription “all you can eat” busniess models. As most of you have pointed out there is no going back to the buy as you go model for most. I saw a cartoon last week that showed a guy holding a iPod and the caption hold 10,000 songs…@ a buck a piece. When was the last time you carried $10K in your pants…The logical alternative is to license all the music and try and track who deserves what. Models have been made just for this purpose and people have generally been open to the idea. If some pay 15 bucks a month ofr WOW, music lovers should be willing to do the same. the problem is DRM and how restrictive themusic ends up. If the music fees was levied at the ISP level and became a $5.00 per month charge across the board everyone would have to pay but no one would worry about virus laden files. One issue with this whole piece is how software is treated the same way as music now. Office online apps and open source may yet be the only viable way to the future.

  3. Money for recommendations creates a motive to recommend material other than “because the material is good”, unfortunately.

    And what’s this “monthly subscription fee” BS anyway? I’m not paying a fee just to access data I already have. Never. No way. Maybe for automatic receipt of new material, so if I quit paying I stop receiving new stuff but nothing happens to what I already have…

  4. Interested Reader/Listener says

    I’m curious about why you would need propagation at all for anything other than updates. In ten years time, the four major labels will likely be smaller in number, probably just one of the existing “majors” plus an aggregator of independent labels. Why wouldn’t this mega-label partner with the device maker to sell devices pre-loaded with every song ever recorded. This is effectively equivalent to a universal jukebox, except without relying on the availability of streaming.

    In either case, the revenue model switches from a sale to a recommendation. If you listen to a song because it was recommended (either via your friend’s device or in an advertisement or via a blog post somewhere) a portion of the monthly subscription fee is paid as a royalty to the artist and a smaller affiliate fee to the person who recommended it to you.

    This would provide a monetary incentive for those making the recommendations and would discourage the type of generic consensus taste-making we have on radio today. Many of us have someone who influenced our tastes in music (for many nerds this may have been Dr. Demento). How great would it be for someone like that to be able to make a living, or augment their existing living, based on their ability to find and share great music?

  5. @Kaleberg

    So then the RIAA will arrest kids for talking about the type of music they enjoy and the court will be told that some undocumented Copyright violation occured (or might have occured, or there was a tangible possibility that the music might have been available to be copied).

  6. Wow, that’s neat. You’ve come up with a whole new rationale for Facebook and MySpace. Not only can you collect friend-oids, you can find out who has the music you want. Next stop is the MP3 to MPEG converter that turns your face music into a “visualization” based on a picture of your face or your dog’s face or whatever. It’s such a pity, at least for their stockholders, the media companies value control over money.

    P.S. You don’t even need infinite storage for this. For practical purposes, most people already have infinite storage for music.

  7. Of course, rather like vulcanized rubber (or pick your own efficiency-enhancing invention), infinite local storage (conjoined with near-infinite local processing) also offers the promise of enormously increased revenues for people who approach the problem right. The amount of music I might like is huge compared to the fraction of music I can reliably find out about in any usual way

  8. It’s not like the music industry doesn’t know it’s ship is sinking. It’s just that they still don’t see a good alternative to bailing water. Once someone shows them a lifeboat, they’ll abandon ship.

  9. I think the question of “storage size” is largely superfluous already.

    Most heavy music listener listen to 3-4 hours of music, and probably own about 500 or so albums. at 256kbps a 30-60GB players is more than enough to carry lifetime worth of albums. In that player a person can listen to music without a repeat for weeks on end.

    In the future people don’t want “quantity”. People simply want “good” music they like all the time. Perfectly matches the mood. No filler and no junk. The novelty is not carrying unlimited amount of music, but to have perfect music for each special moment all the time.

    Unfortunately that is not a trivial question to answer. Music appreciation is artistic effort. Any small PC can now automatically search the net, collect music and accumulate it in a hardrive. It’s a trivial exercise to catalog several TB worth of popular tunes. (P2P tracker search essentially is this catalog)

    In the near future when wimax is everywhere, the notion of owning every music available on the net in a pocket is very much true.

    so taken together, albeit with little effort, the question of “able to obtain any music on the net” is already here. (trust me, I run an mp3 blog) It’s not hypothetical.

    The challenge is to answer somebody who ask “know any good music?” This one is very hard to do. Machine assisted suggestion can only go so far.

  10. It’s an interesting observation, but with wifi enable devices now coming on the market the technical possibility of having an infinity media collection (and that’s audio and video) in a single device exists today. However even though wifi has been around for 5 years it is only now that its power and potency is starting to be realized.

  11. This is a cool idea, but it leaves open the question of how you get these devices filed with music in the first place. Until some critical mass of people has a filled device, searching to locate the song will still be needed. How to establishing such critical mass without prompting a reaction from the RIAA is a mystery to me.

    Moving to a different but still on-topic idea, something similar to this occurred with chess games. For a long time master and grandmaster games would sell in magazines and books. A niche business thrived around that because top level players need to keep current with new ideas that came out in recently played games. Every so many month, new books would be released with all the games played during that time and every major tournament released a book of the games played. In the 80’s I had friends with bookcases full of such books.

    Now all the games ever played at master and grandmaster level can fit on a single optical disk. You can now buy databases containing millions of games with subscription services to automatically update your database with the newly played games when they are published.

  12. This post & the subsequent comments bring 2 thoughts to mind… one that is not new, and is ongoing, and the other being a very recent (read: today) development.

    First – is everyone here familiar with “The Big Hack”, and OFF (Owner Free Filesystem). It’s a pretty interesting concept, and I’m sure Ed (and others here) can add some insight into it’s viability. The big idea being that at their core, all of these media & products are just bits, and who owns the bits?

    For example, let’s take an MP3 of a Madonna/Beach Boys/Megadeath/Nora Jones song. It is a bunch of 1’s and 0s. Let’s say it’s 5 MB in size. Let’s say I don’t anywhere on my hard drive have those 5 million bytes strung together in an MP3 file. I just have a bunch of smaller blocks of bits. But somehow, I am able to put them together “at run time” in a player to make music. Maybe there is a meta-file somewhere with indexes into this table of bit-blocks, maybe it is done in another way… Where is the copyright infringement?

    I have done an exceedingly poor job of illustrating the concept with my example, but I’d love to see discussion of “The Big Hack” and OFF here at some point (the Big Hack forums seem to be pointing to developments being presented towards the end of the year). Note that this model (bit-blocks without any inherent meaning on their own) applies equally to music, movies, books (well, ebooks), software, etc…

    Secondly — this just in today — Nine Inch Nails (aka NIN, aka Trent Reznor) has just been released from his record label contract, I’m sure in no small part due to his overt and extremely harsh criticism of his label, and in fact all music labels in general. He has a very large & devoted fanbase, and I’m 100% sure he will be very successful without a label behind him. To my knowledge, this is the first departure of a very major artist from the labels, in the “I’ll do it myself” direction. Other artists will watch, other artists will follow. I think Radiohead is trying something very interesting with their next album (google them if this is news to you).

    My guess is that this is the beginning of the end of the copyright cabal. Not because they will come to see the light, but because the light will be forced upon them. It will take years, and it will be a tumultuous and painful death, but it is coming.

  13. Crosbie,

    I share your hopes, but not your optimism. If the past ten years are any indication, the music industry is clinging as tightly as it can to the business models of yesterday and shows little sign of changing its attitude or actions.

    Unfortunately, I think copyright abolition is something that is very far off in the future, if not completely impossible, seeing as how our representatives in government fit nicely in the “copyright cartel’s” pocket.

  14. Danny, I have every confidence that the copyright cartel will soon give up its anachronistic notion that it can control the diffusion of published works.

    I also did not mean to imply by ‘An internet jukebox’ that this service was necessarily implemented on one machine, or by a single service provider.

    When copyright is abolished, the opened floodgates to distributed technologies (p2p) will soon dispel any notion of centralised control.

  15. Crosbie,

    I think the unspoken assumption you’re working on is that songs can flow unimpeded from one person to the next. If this were the case, you’re right, there would be no point of personal storage. But that assumption is not valid; the powers that be (RIAA, et al) would like to act as gatekeepers for this flow of information and therefore these transfers have a price/risk. If getting songs over a network carries a potential price tag of being sued into financial oblivion, the collective will to mutually transfer songs drops sharply (arguably). The corollary is the problem with your Internet jukebox idea: if some hostile entity is in control of the centralized collection of all music ever made, it’s really not a useful service.


    That’s a pretty cool vision, but I’m confused about a couple parts of it.

    The possibility of infinite storage does not necessarily mean that users will be packing these drives full of music. Just because my friends could be capable of storing every song ever recorded in a lossless format on their hard drives does not mean they would actually take the time to do so: they just might have their favorite songs on their drives, as they mostly do now: It’s rare to see someone’s hard drive 100% full of music. (Another concern is that it’s a enormous task to organize/manage a huge music library: different formats, different metadata, mislabeling by the people you got the music from, etc.) What is the point, from a user perspective, of having all that music on disk if one does not listen to it?

    So it seems that we will end up having to continue to proposition strangers for songs if friends do not have the desired music. The other development, I believe, that is needed for the situation that you described in your post is (near) infinite bandwidth. The possibility of a “flood” increases in likelihood if everyone is connected to a T3 line (or faster) instead of a 200 baud modem.

  16. Nothing will change if the RIAA and the MPAA lobby congress enough to pass laws worse than the DMCA. And I’m sure they will do this instead of embracing the new technology.

    I think the world will really change when children who grew up with this kind of technology at their disposal are in control of the corporations that are so against it right now.

  17. It’s not far off that people will be able to pop over to some copyright disrespecting country and for a mere $100 purchase a pack of 100 Blue-ray DVDs with MP3 versions of every CD ever published.

    But, the other thing to note is: Why does anyone want possession of such a collection? If there is an Internet jukebox that provides free streaming of every CD ever published, why keep the copies?

    In the future, it is not that the copy becomes free, it is that the copy actually disappears. There is no copy. There are only originals, either published or unpublished. Logically you pay for unpublished works, but published works are free to all.

  18. Perhaps people will finally start using FLACs…oh I hope so.