The author proposes a tried-and-tested method for providing compensation -- the offering and acceptance of tips -- and provides an example of a successful internet business using this model.
Legal cases such as RIAA v Napster have focused media attention on the problem of widespread distribution and replication of intellectual property on the Internet. The legal technicalities of the case are beyond the scope of this document, but I believe that it boils down to this: while Napster may not be directly infringing upon the copyrights of RIAA members, they are maintaining a database that is, effectively, a "list of places where Napster users can go to make copies of copyrighted materials." Whether or not this is copyright infringement is what lawyers call "an interesting question..." 
However, what the RIAA does not seem to grasp (or perhaps, is mortally afraid to acknowledge) is that regardless of the merits of their case, suing Napster into oblivion is merely the first battle in an escalating techological war that they cannot win . Napster variants such as Gnutella disperse not only the copyrighted materials but also the database that ties them together into a diffuse cloud of servers owned by a multitude of individuals. The result is an environment where there is no central database, and thus no central target of opportunity for legal action.
Even worse (from their perspective), recent proposals for cryptologically disseminated distributed databases will soon result in an environment where nobody can determine whether or not a particular server is hosting fragments of a particular copyrighted work. Such a service also has a clear "substantially non-infringing use:"  to avoid censorship and promote freedom of speech. In such an environment, a lawsuit against the owner of a particular server (whose collectible assets may not be enough to cover the cost of the lawsuit) will be both difficult and pointless. Squashing one ant in the colony accomplishes nothing.
Thus, groups such as the RIAA are fighting the wrong war. Instead of making a doomed effort to prevent their intellectual property from being distributed, they should instead be devoting their efforts to creating an environment in which users of their property feel obliged to pay for the use of it.
Fortunately, there is a long-standing business model that fits their needs. It is called waiting tables, and the payment method is called "tipping."
Tipping is totally voluntary; there is no legal requirement to tip a waiter. Yet, assuming the service provided by the waiter is acceptable, almost everyone tips (I speak from a North American perspective here; tipping is not a universal custom, of course).
The interesting thing about tipping is that tips are paid even when the customer is never again going to patronize the restaurant (for example, a restaurant in a city that the customer will likely never revisit). Why don't people "free ride" and skip out on paying a tip in this circumstance? It's not like stiffing the waiter is going to have any consequences - what's she going to do, wrestle you to the ground and pump your stomach?
The reason why people tip in these circumstances is, in my opinion, because (1) it's expected of them by society, and (2) people don't like other people to think badly of them -- even waiters they are never going to see again. These forces are extremely powerful motivators. If the ethics of tipping could be extended from cuisine to intellectual property, then the problem of internet distribution of intellectual property changes from one of restricting distribution to ensuring that it is as widely distributed as possible.
An important thing to note about tipping is that it is a face-to-face transaction between two individuals, as opposed to a face-to-faceless transaction between an individual and a corporation. This distinction has important consequences.
One of the essential aspects of tipping is that it is the customer, not the waiter, who determines the amount of the tip. While there are general guidelines, the customer is free to tip more (to reward perceived excellence) or less (to signal below-average performance) as he or she desires.
It is this choice that, quite non-intuitively, is the key to maximizing the income of the intellectual property owner. For I will now argue that setting a price for intellectual property is exactly the wrong thing to do.
For the last 2 1/2 years, I have been running http://selfpromotion.com/, a URL registration/site promotion resource. SelfPromotion.com basically provides tutorials on proper site promotion methods as well as a sophisticated submission "power tool" that minimizes the labor of submitting urls to large numbers of search engines. While I do make suggestions about appropriate contribution amounts (what I think is a reasonable price), and provide some extra goodies (some cute tools, and keeping all the data they entered online for future use) for those who do contribute a token amount, the choice to pay up and the amount is entirely up to the users of the site. Note that the site isn't a pure intellectual property example, because there are some incremental expenses involved in supporting more users (mostly, answering "fan mail"). These are, however, minor.
The results are illuminating; while about 10% of those who create an account on the site pay up, the payment percentage for those users that use the site to do more than submit to the top search engines (which can be done for free at hundreds of places on the net) is approximately 40%. Furthermore, the average "tip" I receive is about $22.50, over twice the amount required to get the extra goodies. Contributions of $50 or more are very common.
By letting the users set the price, rather than setting it myself, I more than doubled my income. This is because I captured tips both from people who could not afford to pay what I thought was fair, as well as those who would willingly pay more. If you set a price for a service before the service is provided, you will only get income from those customers who were willing to pay that price (or higher); you don't get any money from those who think your service is worth less than you do, and you lose extra income from those who would pay more.
A major factor in setting prices in "brick and mortar" commerce is the marginal cost of goods and manufacture. A car manufacturer would quickly go broke if it let the customer decide on the price of a car; the cost of each "free rider" (no pun intended ) who decides he doesn't want to pay for his car would be enormous. In addition, the manufacturer can only make so many cars, so he wants to try and find a price that maximizes his total profit with this limitation in mind. But intellectual property is different, because the incremental cost of providing it to an extra customer is rapidly approaching zero, and you can supply it to all the customers who want it; there are no real production limitations. In the past, there has always been a physical medium, such as a record, CD or book, on which the intellectual propery was embodied, and this imposed a need for a price.
But IP delivered via TCP/IP is a whole different animal, and it requires a shift in perspective. Intellectual Property delivered over the internet is not a product; it is a service.
A major concern of intellectual property owners is that they believe that if they depend on the honesty of their users, the vast majority will just "free ride" and not pay. Thus they feel that they must set and enforce a price. But as both I and Stephen King have found out, this is not the case.
If users of SelfPromotion.com were simply pursuing their own narrow self-interest, those who wanted to use the full power of the site would simply pay the minimum amount they could get away with to keep their account alive . Yet not only do they not do this, but when their circumstances are such that they can only afford to pay the minimum amount, the payments are often accompanied by notes mentioning how sorry they are that it isn't more.
I believe that the reason for this is that the site is very personal in nature. Users understand they are not dealing with a faceless corporation, but with an actual person. It's easy to rationalize stiffing MegaCorp, but much harder to screw the hard-working guy the top of whose head they can see on the webcam answering their emails . Note that this means that the tipping model may be best suited for intellectual property whose authors are clearly definable individuals -- the best examples of which are, appropriately enough, book authors and musicians.
This is the great fear of the RIAA, that the internet will "disintermediate" them from their position as mediators between the artists and their fans. At the same time, it is their greatest opportunity. Someone is going to build a huge business helping artists and their fans connect in a personal way, and enabling the fans to tip the artists. Record labels can exploit huge "first mover" advantage because of their existing business relationships with artists.
They have a choice; like King Canute, they can declare war on the oncoming tidal wave of change, or they can try and surf the net.
This summer, Stephen King began a laudable experiment of a variant of the tipping model. He published the first part of a new novel, "The Plant" on the internet, and asked readers to pay $1 if they liked it. If enough readers paid up, he promised to release further installments. So far, he reports that about 76% of the people who downloaded the first installment have paid up.
Notice how many parallels to tipping there are in Mr. King's model:
However, my experience leads me to believe that Mr. King may have made a mistake in setting the price at a fixed $1, instead of letting the readers decide how much to pay. Numerous readers have, for example, offered to pay a bit extra to make up for the "free riders". The Plant will have 11 or 12 installments, and clearly there are people out there for whom a Stephen King novel is worth more than $12, just as there are some for whom it is worth less.
However, the Plant Experiment is a great first step, and I hope that Mr. King and others build upon it and experiment with variants of the pricing model. More experiments are needed in order to determine what works, and doesn't work.
The Street Performer Protocol and Digital Copyrights is a great paper about a variant of the tipping model which involves the digital equivalent (for the lack of a better term) of bribery! It's very similar to The Plant Model.
I believe that tipping can be the basis for viable intellectual property-based businesses, and may be a major part of the solution to the recording industry's current battle with the internet. Furthermore, because tipping encourages both socially responsible behavior (from the tipper) and better service (from the tippee), it is a positive feedback mechanism of great potential.
With respect to the music industry, both artists and record labels have much to gain by adopting such a model; the challenge will be in developing methods to make it as easy as possible for the consumers to tip their favorite artists.
Finally, with respect to the issue of use of the tipping model on the internet, the "tipping business method" has been in use at SelfPromotion.com for over 2.5 years. I have specifically and deliberately not applied for a patent or any other intellectual-property protection for this business method, and have placed it in the public domain for all to use without charge.
Unless, of course, you want to send some tips ["7] !
I would like to acknowledge the generous assistance of Andrew Greenberg in the creation of this document. Andy, who started out as a hacker, became a lawyer, and then went back to hacking because he wanted to be respectable again, was the first person to comment on the obvious parallels between the SelfPromotion.com business model and tipping.
In addition, both my wife and mother  played essential roles in the formation of the SelfPromotion.com business model. I had originally intended to make the service free (I was annoyed at all those "We can promote your site to the top 1,000,000 search engines for only $49.95" spams), and they threatened me with death or worse unless I charged for it. The voluntary payment ("ShareService") system was my way of placating them -- I never actually thought people would pay!
 The unsaid end of this statement is "...and finding the answer will cost you a lot of money."
 There are obvious parallels to early technological arms races, such as copy-protection of computer games; as the author of one of the more widely pirated computer games of the era, I had strong views on the matter at the time. I got a second taste of it as the author of one of the earliest Macintosh anti-virus programs, Interferon (later Virex).
 Way back when the movie companies sued Sony (now a movie company) because of the BetaMax VCR, the courts held that VCRs were legitimate because, even though they could be used to infringe upon copyright, they had a "substantial non-infringing use:" recording broadcast TV for later viewing (aka Time-Shifting).
 This is the only intentional falsehood in the article. Of course I intended that pun. And I'm not ashamed of it, either.
 I recognise that my model is not a 100% pure tipping model, but it's pretty close. Users can preserve their account by paying any amount (even 1 cent!), and a major reason I ask for a payment is to allow me to clean out inactive accounts before they clog up the server. The extra goodies can be accessed by anyone who contributes $10 or more, and the main reason I do this is to discourage people from abusing (either by accident or on purpose) the most powerful tool, a search engine deep-submission system.
 The infamous baldspotcam that once appeared near the bottom of the SelfPromotion.com homepage. One side-effect of having a camera pointed at you is that you quickly learn to pick your nose only in the bathroom.
 And by the way, "tip" is not an acronym for "to insure promptness", nor is "tips" an acronym for "to insure prompt service." All those cute stories you've heard repeated so many times are bogus. According to the alt.usage.english FAQ:
"Tip", in the sense of "gratuity", does NOT stand for "to insure [i.e., ensure] politeness/promptness" or "to improve performance". It may derive from "tip" in the sense "to tap, to strike lightly" or in the sense "extremity", both of which have cognates in other Germanic languages. Or it may be a shortening of "stipend".
 In the first draft of this article, this sentence started "In addition, my wife and mother", until Andrew's wife pointed out that people might construe this to mean that I was entirely too attached to Mom. Mom, I love you, just not in that way.
To see how tipping works in practice, visit SelfPromotion.com