User:Woozle/piracy prevention

From Lessig Wiki
< User:Woozle
Revision as of 23:22, 23 December 2008 by Lessig (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

This is what seemed to me, back in maybe 2000 or so, as the best way to minimize the damage done by music piracy while making positive use of the new tools which previously enabled that piracy.

The basic idea is to encourage a culture of artist compensation by drawing attention to each new track's licensing status (is this a legal copy? if not, would you like to make it legal for 25 cents?), and making it easy to compensate artists (or, for those who think of all copying as piracy, making it easier to be honest) and helping the audience think of themselves as being in partnership with the artists to fund -- and thereby help create -- more excellent music (which is true even now, but the disconnect is so great it's very difficult to believe that one's decision to buy or not buy makes even the slightest difference to the artist).

The solution is largely a matter of raising consciousness over an issue to change the way people think about it -- and I don't think the desired change is as radical as the media industry would have us believe; most music-lovers want to help the artists whose music they enjoy. The industry only makes it more and more difficult to do this, with less and less indication that the artists are actually getting their just due; this proposal would turn that completely on its head by bringing the income streams directly to the artists and creating an open accounting of those streams.

Right now, if I torrent a CD but decide I really only like 2 tracks, there's no way to compensate the artist appropriately; I have to buy the whole CD or nothing. I don't have $10-$15 I want to spend on just those two tracks, so never mind. Or maybe I like the whole CD, but I'm busy and don't have time to buy a copy -- since I've already got the music and don't really feel a burning need to own a physical copy. But if my player offered to buy it for me (including printable cover art, perhaps) for something like $5-8 just for clicking "yes", I might well do it.

Components

It has two three essential components:

  1. a set of guidelines for all internet-connected media players and CD burner applications to follow. I'll get specific about this below, but the goals are:
    • to encourage paying traffic
    • to encourage quantifiable feedback from users on every track and collection
    • to maximize the correlation between {how well a work or artist is liked} and {how much money the artist makes}.
  2. a logo/brand/seal (call it "Creative Compensation" for now) to indicate that software or hardware complies with the guidelines.
    • Players, burners, and any other appropriate media equipment may display this logo if they conform to the guidelines -- with the understanding that if they display the logo while actually violating the guidelines, there will be penalties (in correlation with the popularity of the sofware/hardware in question; we don't want to stifle innovation by making it unfeasible for a Linux player written and distributed for free to support the standard).
    • Perhaps there would be a different logo to indicate that the author thinks the software/hardware conforms and is seeking any indication that it is not; after a suitable review period with no violations, the higher-grade "conforms" logo could be applied.
  3. free public data aggregated from Creative-Compensation-compliant players plus input from the artists, allowing users to see view popularity of all releases, how much CrComp income is going to each artist, how much each artist spent creating their works, touring & promoting, etc. (and therefore how much loss or profit they're making)

Music-lovers will prefer Creative-Compensation-branded products (players, burners) because they know those products help the artists (while other products do not).

Authors of music-player software will want their players to support Creative Compensation because they know it's a far better option than Digital Rights Management (and doesn't require any encryption or certificates) -- although technically it does manage digital rights, just not in an ugly way.

Lower-tier artists will support Creative Compensation for obvious reasons; reactions will probably be mixed among upper-tier artists, especially at first... but I think eventually things will balance out. There will be far less stigma attached to an artist of the calibre of, say, Pink Floyd, making millions on a release versus, say, Britney Spears making the same amount. (And often, I suspect, the actual artists end up making far less than we think they do -- or even owing money, under the traditional system.)

Record labels... may hate it at first, until they figure out how to revamp their business model to make money from it. Ideally, they should go back to the business of promotion, with the artists as boss rather than the other way around -- this model connects artist and audience/customers much more directly, taking away whatever might be left of the industry's stranglehold over distribution and compensation.

About Those Guidelines

I don't think it's necessary to understand this bit in order to understand the basic plan, though it may give you a better idea of the picture I'm trying to draw. The details here are not essential, and will probably need a little trial-and-error anyway just to see what works.

Also it's important not to make the guidelines so complicated that they'll be difficult (i.e. expensive, in programmer-time, to implement), but here's what comes to mind:

First of all, assume the existence of a coordinated (but non-centralized) way of collecting "spin" data -- what was played and when. I'll get into the details of how this might be done below.

Next, here's what player software needs to do (in order to conform with the guidelines):

  • When encountering a media track with no identifying info, the player should attempt to identify the track. This could be done by asking the user to fill in the data -- or perhaps use a service like the one used to verify radio playlists (does anyone know what this is called?) which can do this automatically for the more popular tracks. Maybe there could be a standard web site to which the player could turn things over, to keep the "minimum compliant player" as simple as possible.
    • CD rippers can do this pretty reliably from CDDB data -- but CDDB is a lousy format, and has lately become encumbered with licensing issues, so is much in need of obsolescence.
    • ID3 tags would probably work for storing identity info, though I suspect we'll want to come up with a more general format.
  • All media tracks should have some kind of licensing information attached. If a track has no license attached, the player should ask the listener which of the following applies:
    • I legally downloaded or copied this file.
    • This file is a rip from a CD or other media I already own. (Yes, this will be on the honor system; I expect that fact will be the starting point of many objections, which I'll be happy to answer... but the whole point of this solution is that it is non-coercive; that's why people won't hate it.)
    • This file originates with me (i.e. I recorded this song, or I'm acting with the artist's permission) -- an easy way to get new tracks into the system
  • The player should always make it easy to buy licenses for unlicensed songs. Some reasonable nagging might be included whenever an unlicensed song is played, but care needs to be taken that this doesn't become so annoying as to cause people to abandon Creative Compensation. Experimentation is needed to find the right mix.
    • The player should also offer the option to delete unlicensed files -- and report this fact, along with any listener comments, back to the database.
    • Perhaps there should be a bulk rate for large numbers of files. I think the average music-lover would be willing to fork over a lot more if it were just one transaction and s/he knew that each artist would be compensated according to how much s/he liked that artist's work. In fact, this might become the dominant mode of compensation: download tracks over a few months; when the "NOT LICENSED" notices start to get embarrassing or annoying, fork over $50 to license them all. For this reason, maybe there should be a grace period on unlicensed tracks. (Perhaps this could even be set by the artists. If an artist starts being an asshole and setting the grace period too short, this will hurt sales -- and the data will show why. Let the market find the ideal grace period, rather than setting one by fiat.)
  • When playing any song, it should discreetly display the license under which the song was obtained. This is one place where we'll use just a tiny bit of negative reinforcement to encourage people to make their "pirated" copies legal: if they're playing a "pirated" copy, anyone looking over their shoulder will see this. If they have their songs "announced" on IM or IRC, the licensed status should be included - and everyone will see who is supporting the artists and who isn't. We want to encourage, gently but on a vast scale, the idea that playing unlicensed tracks should carry a bit of social stigma -- since it's so easy to do the right thing, support the artist, and make all your tracks legal.
  • (Oh, yeah, there's an idea) If a player "announces" songs on IM or IRC, it should include a link to the artist's page or perhaps even a page specific to that song (make it easy for others to hear exactly what you're hearing).
  • Where appropriate, the player should make it easy to:
    • Visit the artist's web site (a clickable link or button which loads a web browser would be fine)
    • Give the artist a donation directly
    • Keep track of the donations you've given to each artist (possibly by song, even... again, this could be turned over to a web site to minimize browser complexity)
    • Rate each track and collection, preferably in multiple user-maintained categories
    • Provide additional personal information (on a purely voluntary basis, knowing that it could help the artist) such as geographical location

The guidelines could also specify services (and standards for providing them) which artists could provide and which players could tap into if it seemed appropriate -- RSS feeds of artist news/blogs, merchandise, new releases... Simply by creating some standards and drawing attention to them -- by linking artist information feeds with the listening experience -- a lot more communication between artist and audience could happen.

Collecting Data

Again, understanding this bit isn't essential for understanding the proposal overall; I'm only including it for technically-minded skeptics who might be asking how this can possibly work without being centralized.

"Spin data" is basically a timestamp of whenever a song is played -- possibly including some additional info so we can identify patterns of play without identifying specific players (like what the previous song was and when it was played, and whatever other info the user is willing to release voluntarily such as location, age, etc.). This data will be used to allocate artist compensation from "pooled" income streams such as internet radio, and also to generate popularity charts (which should far more accurately reflect what people like and listen to and might buy more of than sales charts do, I should think).

What comes to mind is something like this:

First, you have a standard format (probably something XML-based) for listing "works" -- name of piece, copyright owner, license terms (price per listener for radio play, price for personal single-play, prices for downloads at various qualities...), album ID, artist ID...

Second, every artist would provide this data, at a fixed URL on their own website (like an RSS feed), for all their published works. Third parties could publish data for inactive artists; labels could publish data for all their releases. (There would need to be some process for resolving data conflicts in case the same works were documented by multiple parties.)

Third, "aggregation" sites would publish lists of all the artist data URLs. This data would be freely copyable and modifiable, but perhaps licensed under some specific terms to prevent disreputable individuals from playing fast-and-loose with the data (e.g. redirecting artist income or pointing artist URLs to their own online store). Every aggregation site would be required to add any valid artist data URL to its list, and to list all the other known aggregation sites. (Is this how torrenting works?)

Fourth, any web site could announce its availability as a "data collection" site; aggregation sites would be required to list data collection servers as part of the price of the "works" data. (Perhaps this could all be done under the same "creative compensation" logo.)

In short -- distributed model, with lots of copies of the same data so no one copy is vulnerable to manipulation and so no one entity has all the control, with a process for resolving discrepancies and a process for updating. There are obviously some details to work out, but that's the basic idea.

Other Applications

I've mainly discussed how this would work with software music players. Obviously a similar model could work for software video players, including "media-center"-style home video players which are based on a computer. TiVoed TV episodes could start earning money (how do they count those, anyway, as far as the ratings upon which ad rates are based?); torrented TV episodes and movies could start earning money. Makers of especially popular free videos (e.g. Pat Condell on YouTube) could start earning more money for their work, without even asking (much less demanding).

It's not clear that we would want to try to encourage this scheme for types of software other than players. It's probably redundant for CD rippers, since you don't get the benefit of ripped tracks until you put them in a player. It might be helpful for CD burners to be able to report which tracks you're burning to CD, but is it worth it? My take at this point is that it is not.

And then there's hardware, which is a whole different problem. Old-fashioned CD players (including MP3-CD players) are probably not worth the effort, at this point; you run into a brick wall just trying to get the "spin" data back out of the player, at least with today's technology.

Digital MP3 players, however, seem entirely fair game. I don't own one, so I can't really make any specific suggestions... but the same rules should probably apply -- though perhaps the user-interaction should take place during the stage where you're copying files over to the unit, rather than at play-time.

And I've gone on into far more detail about this than I thought I would, so I hope I haven't glazed too many eyeballs. Please feel free to comment or criticize on the discussion page.

--Woozle 19:22, 23 December 2008 (PST)