A Copying Culture

Version 0.8 - April, 1999.
Author: Andrew Scott


This article investigates how the Internet will impact on traditional and emerging businesses that rely on control of copyright restrictions. Such restrictions are a critical part of the business models used by the distributors of intellectual work, such as media empires. However, because this is so critical, it exposes a weakness that is particularly exploitable on the Internet, since copying can be easily carried out and copyright cannot be restricted.

In looking at how this weakness has arisen, this article discusses how some businesses on the Internet manage to avoid it. Examples from the worlds of MP3, shareware, and Open Source are investigated, providing insights into how a business can be `Internet-proof'. The basic method for achieving this is to adopt one of several business models which do not rely on charging for easily copyable products.

The two classes of model are the `Exploit the Opportunity' and `Exploit the Relationship' models. For each of these models, a business vulnerable to exploitation of the copyright weakness is sited, and reworked as an example. Media-based businesses must follow such models or face losing revenue to businesses that are run illegally, or overseas where copyright restrictions simply cannot be enforced.

Table of Contents


This article is written for people who are in an industry that involves `intellectual products'. The distribution of such products is governed by copyright, and in fact businesses that are based on this distribution now depend on copyright.

Such businesses have been around for years, but now the landscape in which these businesses operate has changed. As we will see, this change will have quite drastic implications for traditional businesses, who may have to develop entirely new business models in order to cope. The change is the spread of modern technology.

There are two particular aspects of modern technology that concern us in this article. The first is the widespread use of technologies that allow intellectual products to be copied. The other is the world-wide growth of the Internet. As a result of the spread of these technologies, modern culture has changed, and now readily accepts the practise of copying, and retrieval of copies from anywhere in the world.

In the following section, Copyright, we will see how businesses have pushed the boundaries of copyright as they have become more global, but how copyright is not enough to completely control the copying of intellectual products in a global arena. The section on Motivations looks at the extent to which people copy products, and why they do so. In Products and Services, we will see descriptions of typical businesses in the intellectual products industry, and see how the different components of their business models based on control of copyright.

The section on Internet Distribution examines how businesses on the Internet are faced with problem of widespread copying. Several of them are coping well despite this, and we examine them in the section on Internet Models. These examples suggest two main business models that resist the traditional weakness of dependence on copyright, and these are examined in Exploit the Opportunity and Exploit the Relationship. Finally, we draw the various threads of the argument together and state the Conclusions that follow from our analysis.


In order to see how control of copyright has become essential to the running of certain businesses, firstly we need to see exactly what it is, and how it has evolved.

Copyright is the legal restriction on making copies of intellectual works, such as books, films, software, and music. Control over copyright is usually automatically given to the author of an intellectual work, but may be sold to other entities, such as publishers, in return for money or the privilege of having the work published. It is only one of many legal mechanisms for protecting intellectual works, other common ones are trademarks and patents. Its advantages are that it is pretty much applicable world-wide, is automatically bestowed, allows for `fair use', and accepts the occurrence of independent invention. One of its chief purposes is to encourage creativity, by giving authors the opportunity to exploit their own work.

The media-based industry obviously benefits also from the existance of copyright. With businesses based on material that can be copied, copyright laws that restrict the ability of their competitors from duplicating that material are welcomed by the industry. We shall see how such businesses have come to rely on these copyright restrictions.

The software industry has its own particular slant on copyright, which is particularly interesting since it is the only form of work that was created into an electronic world. When software is installed and run, many copies of different parts of it are created. When copyright was invented, such a self-copying creature was never imagined. Hence, software is distributed with a license to make more copies of it, for the purpose of running it. It still isn't clear exactly how legally binding a software licence is. It is more complex than the idea of copyright, as it gives additional rights to the providers of the software. However, we can think of a licence as a particular variant of copyright.

We have had copyright for quite some time now, although its exact implementation has changed over the years. In Britain, copyright appeared in 1709 soon after the printing press, primarily as a censorship strategy. In the United States, the first copyright act dates from 1790, when copyright applied for a maximum of 28 years.

The Berne Convention is a global agreement on copyright between most modern nations, and provides the form of our present copyright system. It was formed in September 1886, although notably the United States didn't join until March 1989.

This was not the United States' first involvement in world copyright, and it has been part of the Universal Copyright Convention (UCC) since September 1955. The main difference between the UCC form of copyright and that of the Berne Convention is that under the UCC, copyright is not automatic.

This slow march to global copyright laws is paralleled by the steady globalisation of media-based industry. This globalisation can be charted by the steady advance in broadcast technologies, although the industry also encompasses other media, such as film and print. In 1923, the UK firm Reuters began for the first time to transmit news stories internationally, using radio. In the late 1930s, regular, broadcast, black and white television appeared. In the 60s, colour television broadcasting began, and this present decade has ushered in digital TV broadcasts.

We now live in a world dominated by media that is controlled by a small number of large players. These include Disney, Time-Warner, News Corp, and Microsoft (who for our purposes we can consider a media empire). Global players such as these rely on global copyright laws for their business to continue. In almost every case, the copyright of intellectual product distributed by these companies is owned by these companies.

Authors of film, music, writings, and software have a hard time distributing their products globally themselves using conventional distribution channels. The easiest way to get your work into circulation is to sell the copyright to a big player and let them do the distribution. Sometimes you will get a small fraction of the royalties, often you will get a once-off payment, it depends on the agreement you can come to with the publisher. The profits from sales of these works go almost wholly to the big player, for which copyright is critical to ensure continued sales.

Additionally, it is understandable that copyright and many other legal devices can be used to best advantage by a large company. Individual authors would find it very hard to hunt down and prosecute violators of their copyright, especially as in general they wouldn't have enough money to take them all through court. An individual author might use the services of a larger body, such as APRA in Australia, to police usage and collect revenue. The disadvantages to the author in this case are that they may have to agree to the fee collection policies used by that body, they must trust the body to police their rights properly, and by retaining their copyright, they may not be able to easily use the traditional distribution networks operated by the media industry.

Microsoft and the large software companies use the services of organisations such as the Business Software Alliance (BSA) to protect their profits through enforcement of copyright restrictions. In fact, it has been reported that after the BSA discovered certain copyright violations, copyright lawsuits were not pursued when the violators signed software licences with Microsoft exclusively.

So important is copyright to the business of international media empires that they have recently pushed for, and gained, an increase in the term of copyright in both Europe and the US. As a result, the copyright term in the US is now 95 years after the creator's death, and correspondingly 70 years in Europe. (One humanitarian and apt exception is the perpetual copyright bestowed on the childrens' story Peter Pan. An act of parliament in Britain ensures that royalties will always continue to flow to the Great Ormond Street (London) Hospital for Sick Children.)

The length of the copyright term has been steadily increased, presumably due to pressure from the global media community, as this most recent increase has been. This community, and the publishers within it, must be highly dependent on copyright in order to motivate such relentless pressure. The trouble for publishers is that copyright is reliant on the particular local laws, and always runs out (except in the individual case of Peter Pan). As soon as a valuable intellectual work enters a country with lax copyright laws, or the copyright term runs out, control over that work is lost. Assuming that the copyright term will not be continually extended, copyright is never going to be completely adequate for the needs of the media empires.


Copyright aims to prevent works being copied. But copies are still likely to be made of a work when the perceived benefit of making a copy is greater than the penalty of the fine, taking into account the risk of being caught. In fact, in our culture, the copying of copyrighted works is almost a mundane occurrence. We shall investigate why this might be so.

Let's look at a couple of intellectual objects, and start by considering the benefits and disadvantages of someone copying information about the price of a certain stock that I own. Knowing the most recent information about the stock's price is of high value to me, and as the information becomes older and older, the value decreases. Knowing the price of the stock as of last week is of no real value to me. In fact, US stock information delayed between 15 and 20 minutes is available for free on the Internet. If someone gained access to recent stock information, then because it would be valuable to me, they might sell it to me. Hence making a copy of the information is beneficial to them. But copying may cause them to be disadvantaged if they take too long performing the copying, as the information will become worthless, and they will have wasted their time.

The lesson from copying time critical information, such as stock price information, is that there is only a small window of opportunity in which the copying can be benefited from. This makes it hard to benefit from, although the benefits may be quite high due to the exclusive nature of the information.

A compact disc is a more tangible intellectual object. I listened to a CD last night and enjoyed it. However, I could have listened to any particular format of it and still received the same enjoyment. If I'd had a copy of the CD on cassette, and I couldn't find the CD, then I would've listened to the cassette instead, being only slightly disadvantaged by the lower sound quality. Having a `backup' copy ensures that my investment in the CD is preserved.

The disadvantage of getting a backup copy is that one needs to be bought, or copied onto a blank one. Preserving the backup copy requires a possibly long-term commitment. Making the copy will always have some cost associated with it, and it may never be required. The risk of being caught violating copyright restrictions in this way is very low. Hence dubbing is quite a common activity. Alternatively, the copy might be sold or given away. This is considerably more risky, but no more expensive. The benefit here is that there will be an incentive to sell the copy.

Interestingly, the current generation of people in western nations has integrated the act of copying into the culture. The spread of technology has not just benefited media empires, and today we have access to photocopiers, VCRs, integrated CD/cassette decks, fax machines, computers, scanners, CD-R drives, music samplers, and other devices that allow people to create illegal copies of intellectual work very easily.

Primarily, it is this ease that has allowed copying to become part of the culture, despite any legal objections that there might be. The practises of creating `party tapes' from existing music collections, recording programmes from TV for later viewing, photocopying newspaper articles to give to friends, or wearing fake designer watches and clothing are extremely widespread. Some people have drawn this behaviour into the philosophical ideas of post-modernism, which is not something I wish to discuss here due to its contentious nature, although it is worth a mention. People make copies of things every day and think nothing of it. The value to them of having a copy is enough to outweigh the risk of a fine, and so copying has become habit.

An OECD report on international software trade reports that pirating of software has become endemic in some countries. Some examples of figures from the report are that in 1996, 27% of software installed in the US were illegal copies. Similarly, 41% of software in Japan, and 34% of software in the UK were illegal copies. From this it is clear that in western nations, around a third of people would copy software if they thought they could get away with it. In most other nations, these figures are much higher.

People are told not to make copies of intellectual works, and not just because making copies is illegal. The usual argument against copying is the same as the argument for copyright: copying removes the incentives for people to produce the works in the first place, so in the long term we all suffer from the works not being made. This gives us a moral reason (as well as the legal reason) for why people should not make their own copies, but is it a compelling reason?

There are three straightforward objections to this reasoning:

  1. This is an argument to a group, and not an individual. It doesn't explain why some person in particular shouldn't make copies.
  2. Sometimes copying doesn't prevent the creator receiving incentive, and actually gives them more incentive.
  3. Other perfectly legal actions are just as likely to prevent the creator receiving incentive, but we think that those actions are acceptable.
Let's look at these in turn.

In social psychology, it is well known that within a group, there is a diffusion of responsibility. This means that any argument that is put to a group will be diluted because the members of the group do not always feel personally responsible for the conclusions. In our case, we are examining an argument of the form "we should all respect copyright, otherwise we will all eventually suffer". However, if we don't all respect copyright, and instead only 99% of people do, then that 1% of people will not actually suffer. These 1% are parasites on the rest of the system, but the argument cannot persuade someone to be in the 99% rather than the 1%, since it is actually a disadvantage for them to be so. The parasites get the benefit of copies as well as the benefit of continued innovation.

The second objection given was that there are examples of when the opposite of the claim in the argument is apparently true. For example, in the world of music, the band The Grateful Dead credit the bootleg industry created by their fans as one of the factors enabling them to become one of the highest grossing acts in history. In the world of software, the concept of shareware is that a potential user can try the software before they pay for it. Shareware creators encourage users to create copies and give them to their friends and workmates, since each recipient of a copy is a new potential user. In both The Grateful Dead and shareware examples, copying of the work increases the incentive that the creators receive.

The third objection was that other legal activities are as likely to prevent the creator receiving incentive, but we allow them, and people continue to create new works. These are activities where after one person purchases the work, multiple people then proceed to benefit from that work without any more money going to the creator (or their publisher). Hiring, borrowing, and buying something second-hand are all legal examples of this type of activity. In fact, as with the examples of bands and shareware above, often the creator does benefit from these activities in the form of increased publicity and exposure. Exposure then translates into money from follow-up business and increased demand for the work.

We can draw some conclusions from all of this. Firstly, copying is often beneficial to the people who do it, mostly because the risk of being caught is small and the reward is high. Secondly, due to the increasing availability of copy technologies, copying has become a common part of our culture. Thirdly, the argument that people should respect copyright because it encourages new work is not strong enough to be convincing. From this, it is apparent that people are being raised in today's culture are going to expect to be able to continue their habits of copying. In fact, this practise is likely to become even more common.

Products and Services

There are two main categories of business that relate to intellectual works, such as books, CDs, software, and films. These are businesses that are involved in selling products, and businesses that sell services. The obvious question is: what is the essential difference between a product and a service? Simply put, a product is something that can be given away, but a service is something that can be taken away.

Control over the continued possession of a service is in the hands of the provider, while control over the continued possession of a product is in the hands of the receiver. This is because while the receiver can deprive the recipient of a service, once the recipient gets their hands on the product, the provider cannot ever be sure of depriving them of it. To make it a totally clear, let's look at some examples of products and services in the world of intellectual works.

A radio station provides the listeners with a service: the listeners are provided with recent music by good, or well-known artists, and commentary by amusing disk-jockeys. Of course, some stations provide more or less than this, but this is usually the essence of it. A radio station also provides a service to advertising agencies or community groups, allowing messages from them to be broadcast to the station's audience. These are both services, as the radio station could stop them at any time.

A bookstore provides products and services. Book buyers are provided with the service of having a collection of easily browsable publications in a single place. A store also retails a product which is something that can be taken away and read on the loo, or curled up in bed, or wherever you like to read. Interestingly, I suspect that the majority of a store's day-to-day expenses are involved in providing the service (staff wages, leasing shop-front, etc.) while all of the income is from sales of the product. Once again, the service is something that the store can do away with at any time, but the product is yours to read however and whenever you like.

A software publisher provides products and services as well. The product is the software that you run on your computer to entertain you or increase your productivity. The service is writing this software to meet your requirements, in enough time for you to use it. Sometimes the service is predicting your requirements so that the software is there when you need it. There are often follow-up services such as providing you with support once you start using the software, and modifying it as your needs change.

A newspaper publisher resembles an interesting mix of the services of the radio station and product of the bookstore. One service the newspaper provides is to make an up-to-date product available to you and to many other people that you know. The product is, of course, the newspaper itself. As an aside, newspapers never used to provide up-to-date news. In the pre-telegraph era, news would slowly disperse from newspaper to newspaper across the country, and could be many months old by the time you read it. However, these days most people wouldn't bother to read a month old newspaper. The product is not so important to them, but the service is.

In a couple of these examples, the service and the product were tightly bound, for example the service consisted of distributing the product. In Geoffrey Moore's book Crossing The Chasm, the term "the whole product" is used to describe this conglomeration of services together with a product. Without a whole product to buy, consumers are unlikely to commit themselves to a purchase.

In these cases the business model is often to charge for the product, and not for the service. This is often because it is easier to charge for the product than for the service: customers understand the policy of paying for a product; the flow of the product can be easily regulated to ensure that only paying customers receive the associated service. So why is this process so product-focussed?

Copyright comes in once again. Copyright restrictions are used to ensure that other people or organisations cannot distribute the same product. As long as the only way that the product can be received is via the service that is provided, then it is only necessary to regulate the product to regulate the service. Control of the copyright and control of the service go hand in hand to ensure that a premium can be charged for the product.

This once again indicates how much the media empires rely on the restriction of copyright to maintain their business. Without copyright restrictions, anyone could distribute the product through a competing service. In this case, as no income is being received by the business through sales of product, the service is no longer viable. The use of copyright in this way ensures a `monopoly' over the product, controlling supply, hence increasing demand, and meaning that people are willing to pay for it, to get the service as well.

Internet Distribution

Academia has a history of being rather lax when it comes to respecting copyright. It is common to photocopy large numbers of articles and claim fair use, although my experience is that often this would not be legally justified. Also, in publishing a paper, the copyright is sold to the publisher, but often the writer or writers continue to give out copies of it. This atmosphere of sharing knowledge for the good of the discipline has probably had a very positive effect on research overall, and this is the atmosphere that the Internet was born into.

The Internet has evolved from those early days, although it still retains the influence of its cultural origins. For instance, when advertising first appeared on the Internet, it was regarded as some sort of perversion. At that time, making money from the Internet was still thought of as wrong. This has changed, but unsolicited commercial email, known as spam, is still regarded as offensive to many people (and hopefully it will remain an unsanctioned use of the Internet for many more years.)

Regardless of the Internet culture, violations of copyright would still occur on the Internet. This is for the simple reason that to effectively restrict copyright, the restrictions must be policeable and enforceable. The Internet is a loosely controlled network, widely distributed over many varying legal environments. This combination dashes any hope of either global policing or global enforcement. The only real alternative is that people will choose not to violate copyright restrictions, and this is not likely to happen, as we have seen, due to the personal benefits of copies and the lack of any strong moral incentive to restrict copying.

The simple act of browsing the web causes the web browser to create copies of material, without asking anyone's permission. The web is a special case too, as sometimes copying apparently happens, but really doesn't. This happens when a web page on one server contains a link to content on another server, eg. an image. When the page is rendered, that content is seamlessly rendered as well, making it appear as if the other server's content had been copied onto the first server, when in fact it was the web browser simply doing its normal copying activity. It is not clear yet whether this sort of thing is legal or not.

One response to the growth in copying of images on the Internet has been the corresponding growth in watermarking. This is a technology that allows a `digital signature' to be invisibly written into the image in a way that will be permanent, even though the image may be rescaled or slightly altered. If a watermarked image is discovered, then it can be immediately identified as a copy of the original, `unsigned' image. This is a clever technology, and it indicates that people are reacting to the need for policing copying, but it only works on certain unencrypted images, and it doesn't solve the problem of enforcement.

To highlight this problem, let's look at the fictional country of Photocopia. Photocopia is part of the global communications network, including the Internet, but is not a signatory of any global copyright treaty. Mr Earnest Dodge sets up an Internet site that contains a whole stack of copyright material, pulled straight off the Internet, and sells it for one dollar a megabyte. Any subscriber can point their web browser at this site and get the content. It seems that the copying that happens as part of using a web browser is considered `fair use', so the guilty party here is Mr Dodge. However, because he is based in Photocopia, there is no way that the copyright restrictions can be enforced.

Mr Dodge might not charge for his site. Perhaps he aggregates a lot of content and sells advertising on it, replacing any advertising that was on the initial content (and presumably paying for the development of that content). Perhaps he doesn't actually store the content himself, and simply puts up links that go directly to the content that he's aggregating. This latter model is the way that the very successful site Slashdot works.

These sort of models can be successful because the Internet is a new form of distribution, one that is not controlled by any media empire. As we saw before, the media empires control the traditional distribution network, and it would be enormously expensive for a competitor to emerge and compete with them by creating a duplicate one. The empires protect themselves by owning the copyright and hence forcing those that want the a particular work to get it from that media empire. Now the Internet can be used to distribute works. An alternative now exists to using media empires for distribution.

Internet Models

I want to make something quite clear. I am not advocating the destruction of the existing media empires. I am not suggesting that we all go out and violate copyright. If you thought that these were my intentions, then I could understand, as there are currently strong feelings like these on the Internet today. However, I am trying to show that violations of copyright are now part of our culture, particularly on the Internet, and that those who work in businesses that depend on protection of works via copyright should consider this carefully and take it into account in their business models. In fact, as David Niemi points out, it is not even necessary to violate copyright in some cases to lose what you were trying to protect, good software can always be re-written by good programmers, given enough resources.

Some may think that there cannot be any successful business model that deals with intellectual work on the Internet. That is not true, and by looking at some of the activities on the present-day Internet, we can piece together some possible Internet-proof business models.

Shareware is the busking of the Internet, and is the single most successful software business model for independent authors. Although, it does depend on how you define `successful'. Many authors write shareware as a hobby, spend very little money distributing or marketing it, then sit back and receive money from the users of the shareware. When a user of shareware pays for the software, it is known as registration. Copying is actively encouraged by the authors of shareware, since each new user of the software is another potential registrant.

Most shareware relies on the honesty of the users. Figures quoted by the Association of Shareware Professionals (ASP) are that between 2% and 50% of users register. A 1994 shareware survey estimated that between 5% to 10% of users of professional quality shareware register.

However, if the shareware author doesn't rely on the honesty of the users, but implements a crippled product that is repaired when registration occurs, then an order of magnitude more users register. This is what Colin Messitt discovered when he performed a large scale experiment. From this, it is pretty clear that over 80% of users will copy software and not pay for it if given the chance. However, due to the large number of users exposed to the software by allowing copies to be readily distributed, shareware generates surprising revenue. For example, Colin's experiment earned him over US$30,000 in one year.

Project Gutenberg was started by Michael Hart in 1971 in order to make a large number of important, out-of-copyright, written works available to the population at large. This was done by storing everything from The Bible to Alice In Wonderland in simple text files, and placing them on bulletin boards and web sites. Volunteers offer their time and their flat-bed scanners to the project, and the library has continued to grow to around 2,000 works.

Strictly, it isn't a business, and hence doesn't have a business model. However, what it does may be copied by others, and they may all be competitors to your business. The project talks about "replicator technology" where a single electronic copy can be replicated as many times as is desired. Hence each electronic book is immediately free to anyone who cares to read it - the act of experiencing that book is parallelisable.

The products that Project Gutenberg produces are not charged for. Basically, the books are given away for free. Project Gutenberg is not a profit-making exercise, but they don't seem to care if the mechanism that is used to get at the free books is being charged for. In this case, someone (the ISP) is offering the service of getting access to literature, just like a bookstore. Although in this case, the customer who gets the book pays for the service, but not for the product.

A more recent phenomenon than the spread of free books is the spread of free music, stored in the MP3 format. The format itself is widely popular, challenging the traditional WAV and RealAudio formats - even CNN is using it on its site. However its most famous, or infamous use, is to distribute popular music. Sites such as MP3.com allow musicians to place their music for free, for people to download it for free, and advertising pays for it. Musicians get free exposure, increasing consumer awareness, and the possibility of offers of tours and other gigs. Sites such as Goodnoise select the music that is available on their site, giving some of it away for free, and selling some of it for around US$1 per track. There are also many pirate tunes available on the Internet in MP3 form at `leech sites', many which can be found via a search engine written to locate MP3s, such as Lycos.

In the case of MP3s, very few sites are charging money for the download of music. Given the availability of free pirated music, competitive pressures are going to be influential here. Since no site can compete with free product, few do. Goodnoise sells very cheap product, but they also sell advertising, and no doubt sell the opportunity to retail music through them. They sell very cheap product and make money selling services.

The Open Source movement has risen to some prominence with the enormous growth of the Linux operating system software, and the number of `respectable' organisations using open source software as part of their business. The Internet itself is built upon software developed as part of open source-like development practises, which have their roots in academia and the peer-review process. Software written under open source arrangements contains a licensing agreement that places unique copyright restrictions upon the user. Source code for the software must be made available, and any changes to the source code must also be made available. This restriction actually prevents the software from ever being restricted to a closed group - such as those who pay a price for it. Several thriving businesses have appeared in the open source scene. These include companies like Red Hat and Cygnus who sell the service of distribution and support of the free product Linux.

In these Internet examples we have seen the shareware concept, project Gutenberg, MP3 distribution, and open source software. They all involve the spread of copies of intellectual objects, but there are two themes that emerge from these examples. The theme from shareware is that each copy provides another chance to gain a registration, and the more copies that exist, the more potential registrations. This model is discussed in the next section. In the other successful examples, the products are given away for little or nothing, but the associated services are charged for. This model is discussed in the section titled Exploit the Relationship. Together these models appear to cover the successful, emerging businesses on the Internet. It was inevitable that successful models such as these would appear once the Internet appeared in a society with a copying culture.

Exploit the Opportunity

When a person receives a copy of a product for the first time, this provides an opportunity for the person who created the product to generate income, but only if they were prepared to exploit it. Shareware works because it exploits this opportunity - the product displays messages asking the user to register, it may tell the user about other products by the same author, it can demonstrate the benefits of registering without providing all the benefits of registering. In essence, the shareware product is a small marketing machine that strives to convince every person that uses it that they should send the creator money.

Shareware is not used only for distributing people's dinky home software creations, but there are many big name software programmes that have used the shareware distribution method. Here is a small list of shareware examples:

The shareware concept is chiefly used by small operators who do not have the resources to run large marketing campaigns. Given this, it makes a great deal of sense to allow other people to pay for the production of copies of the single thing that they are using to market their work - that work itself. The social communication networks that exist between people facilitates the spread of the software as people who appreciate the shareware pass copies onto others that they know. On the Internet these communication networks are particularly powerful as they may be spread very far and between very large numbers of people.

There are two prominant strategies for distributing shareware. The first strategy (Strategy #1) is to have public and private versions of the software, and sell the means for turning a public version into a more enhanced private version. The second strategy (Strategy #2) is to distribute only one version of the software, and rely only on the honesty of users to pay for the software when they make use of it. Each of these strategies is most successful when used to distribute a particular type of software product.

We mentioned Geoffrey Moore above, when talking about a whole product. Another important concept that he advocates is changing your products and altering your marketing techniques when you "cross the chasm". Marketers see the population divided into different groups depending on likelihood of buying a product: innovators, early adopters, early majority, late majority, and the laggards. The majority of the population is in the early and late majorities, and Moore's chasm occurs between the early adopters and the early majority. The theory that a different type of marketing and product is required to reach the majority of your market may explain the varied success of the two types of shareware strategy.

Strategy #1 seems to be more successful in the early adopters market. These people are willing to put up with a reasonable amount of hassle to get the benefits from a product. This is consistent with the idea of having to register before the product performs to its full potential. Very few of the really big name shareware products use this strategy, although it has been known to work, as in Messitt's test described above.

Strategy #2 appeals to the mass market, since once they have decided to use the product, they only need to send out a cheque - the product already works. The ASP frowns on the type of crippling used in Messitt's test, and has pointed out that shareware products that have become the "standard of industry" use no crippling and few nagging reminders. The reduced percentage of registrations that no doubt result from less stringent enforcement of the shareware licences is easily made up by the far greater number of potential users in the mass market than in the early adopters.

Shareware is the busking of the Internet, and the strategies for successful busking are applicable with shareware distribution. Bob Yuncken has described the three main busking rules as:

Addionally, an article on balloon juggling by Larry Moss emphasises the need for These points are confirmed in the findings of the 1994 Shareware Author and User Case Study, mentioned above. In particular, if a user thought that the author was a helpful type, or was writing the code to support his kids, etc. then that user was more likely to register. This importance of projecting the appropriate personality demonstrates how branding concepts carry over into shareware too.

The business model of exploiting the opportunity is not restricted to software. An example of how social communication networks were exploited and a product used newly encounted people as opportunities is the example of Hotmail. Hotmail is a web-based email system, that attaches a little sentence at the end of every email that it sends out, mentioning itself. Recipients of email from Hotmail find out how good the system is because they've received email from someone via Hotmail. New recipients then become Hotmail subscribers by paying for an account in the form of information about themselves. If you don't think that personal information is valuable, then just think how willing you would be to give details of your suburb, age, and occupation to a complete stranger.

Hotmail encourages people to send out little advertisements for Hotmail to all their friends and associates, just like shareware. Similarly, potential Hotmail users are assured of the ability of Hotmail before they buy, just like with shareware. Hotmail has become the single largest email system, and is the dominant player in countries that they have never spent a cent marketing in. The Business 2.0 magazine reports that now Hotmail attracts more than 100,000 new sign-ups per day.

Exploit the Relationship

Once a copy of an intellectual object is obtained via digital means, there is no further need to obtain copies of that object, since perfect copies can be produced by anyone. Given that our culture is becoming one where making copies is a mundane activity, and the Internet makes it hard to know who is making copies of what, basing a business on selling intellectual products is a risky proposition. A parasitic business might appear, using your products, and selling them cheaper, or giving them away for free. How can a business expect to survive against these competitors? A more robust alternative is to base the business on selling services, since there are certain kinds of services that can't be copied by a parasitic competitor.

Often the service provided is the distribution of the product, which may not be a compelling service when the product can be copied by anyone. However, there are types of distribution service that are obviously resistant to this. These types are those that make certain guarantees about the service that are hard for others to duplicate without using (and paying for) the service themselves. The two most obvious of these are the distribution of time-sensitive information, and the distribution of personalised information. Another less obvious example is the distribution of information using environmentally sensative network providers.

The Wall Street Journal (WSJ) is one of the few big-name sites that charges for providing subscription content. This content includes information not available from the free WSJ site. It could be significant that very few other papers provide subscriptions at a charge. Perhaps in general the subscriptions aren't personalised enough, or perhaps a site needs to be a very large player before it can afford to charge for information that can be easily copied. In any case, charging for subscriptions to online information services is rare.

It could be the case that consumers do not really understand or appreciate a model where they get product for free, and yet they pay for an associated service. This is interesting, since in many cases the service is more desirable than the product that the service is based around. Maybe in general it will take a while for customers to understand this business model, but this is the very model that is being used to sell open source software. In this case, there have been very clear successes. Customer understanding must be spreading.

One question concerning this model is why would anyone produce product if they knew that it was being given away? In the case of open source, the people who produce the product aren't even receiving income from the associated services. Eric S. Raymond makes a good case for the theory that people will produce product for increased reputation alone. For example, one possible benefit of a good reputation is that it leads to future material gains in the form of interesting and well-paid job offers.

However, there are some businesses that fence sit on the issue of selling services not products. O'Reilly and Associates sell the development books and tools for the free utility Perl. The production and distribution of books and software tools are services to the developer that keep them up-to-date. Since books and tools are products as well, this makes them vulnerable to the weakness that we have previously identified. Perhaps O'Reilly should look at how they might be able to restructure their operations concerning Perl to avoid this weakness. However, traditional distribution methods are used to distribute these products, so they are not yet under threat by the Internet.

Another fence sitter is Goodnoise. Although Goodnoise was one of the Internet examples listed above, it doesn't closely follow the recommendation to charge for services and give away products. As a final example, let's develop a new business model for it, based on the above analysis.

At the moment, Goodnoise does not charge its downloaders for the service that it provides, instead it charges them for the music, which are products. Goodnoise does charge advertisers and music labels for services, but the exception for downloaders appears to be a particular weakness. Instead of charging US$1 for each track downloaded, it could charge for the service of collecting and providing lists of music that are personally tailored to subscribers of the service. For example, for a charge of US$10, I might be able to download up to 5 hours of music from the site, receive lists of music at the site that I would be personally interested in, know that the operators of the site will be seeking out music that I'd like to listen to, send questions to the musicians who play my favourite music, and submit fan artwork or reviews that are then displayed on the site. These services would all cost money for Goodnoise to provide, but because they are charging for them, they can't lose.

The lesson for Goodnoise is that allowing music to be downloaded is cheap to provide, anyone can do it, and due to the copying culture and the Internet, anyone does. Some businesses follow the Exploit the opportunity model and benefit from large numbers of copies being made, but many businesses do not. To make these online media business `Internet-proof', achieving profits cannot be based on the uncertain sales of a duplicable product.


Control of copyright restrictions is a critical part of the business model of the distributors of intellectual work, such as media empires. However, copyright has always been inadequate, and legislators have been pressured to extend it. A business based on copyright has a potential Achilles heel.

Due to the increase in technology, copying has become extremely easy to do. Copying is now a normal part of our society, and copyright violations of intellectual works are regular occurrences. Violators do not usually feel bad about their violations, and engage in the act of copying routinely.

Media empires often earn large revenues from sales of their products, although they incur hefty expenses in related services which are at no additional charge. In a situation where copyright cannot be restricted, the business may not be viable.

The Internet is a case where copyright cannot be restricted, due to its global presence and cultural basis. For example, a direct competitor may operate from a region where copyright is not respected nor enforced. Businesses that attempt to incorporate the Internet into their business model should be prepared to cope with this.

Two models were proposed that addressed the problem of endemic copyright infringement on the Internet. These were the models of `exploit the opportunity' and `exploit the relationship'. The former model uses each copy of a product as a marketing device that can be exploited. The latter model involves charging for services associated with a product. Of course, these models can be combined, and there may be other models that do not have a dependence on charging for copyable products.

Without the adoption of a business model such as these, an online media business cannot be considered `Internet-proof', and may fall victim to reduced sales of a duplicable product.

Further Reading

Here are some pointers (in addition to the ones in the text) to works that concern the issue of copyright on the Internet:


Thank you to Arran Stewart, Kate Parsons, Eric S. Raymond, Brad Starkie, Brendan Quinn, and Bob Yuncken for the suggestions and kind criticisms that they offered while I was preparing this article. Also, thank you to GeoCities for hosting this page.

Version History

0.1 Initial, rough and ready version. 1st March, 1999.
0.2 Added acknowledgements. Began rewriting following suggestions by Arran and Kate. 10th March, 1999.
0.3 Added Table of Contents. Made small changes at the suggestion of Brad and Eric. 12th March, 1999.
0.4 Added Introduction and information about shareware at the suggestion of Brendan. 19th March, 1999.
0.5 Made small typographic fixes and minor additions suggested by Bob. 22nd March, 1999.
0.6 Made large revisions, continuing the suggetions of Bob, Brendan, and Arran. The section on `Value' is now on `Motivations'. A couple of new sections have been created - these are still unfinished. 26th March, 1999.
0.7 Completed the new sections. Added Further Reading section. 29th March, 1999.
0.8 Fixed minor typos. Thanks Bob. Added John Perry Barlow to Further Reading. 1st April, 1999.