Mass Market Busking - The Inevitable Economics of Software
"Mass market busking" means providing some product or service to
everyone who wants it, regardless of ability or willingness to pay,
but requesting that the users pay whatever they think appropriate.
The payoff for the users who pay is not direct, but it is material,
not just moral. They don't immediately get a product that would
otherwise be withheld from them, but because they establish that they
will pay someone who creates things they like, it will then become
a good business decision to create such things. It is a simple matter
of supply and paying demand: when people are willing to pay for
a product, there will be a proportional willingness to expend money
and effort in supplying the product.
Buskware is a subset of mass market busking, which means producers
of digital information products (everything from computer programs
to recorded music and movies) allowing their audience to freely
distribute these products in return for the audience freely paying
whatever price they think appropriate.
Buskware is the simplest and most efficient way for producers and
their products' users to exchange control over what is produced (and how
it is used) for money.
The 5 fundamental principles of buskware:
- People generally act in their own best interests, in so far
as they understand the consequences of their actions.
- Users want control over the producers.
- Producers want to be paid by the users.
- Computer data is eternal and practically costless to reproduce
and distribute, so individual copies needn't be treated like physical
objects which each have a cost to the producer.
- Copying and distribution of data on other people's computers
can't be controlled or prevented by the producer, any more than
the originator of an idea can control who learns it.
Mass market busking has the advantage for the producer of lowering
costs by eliminating the need for advertising and other means of
convincing customers to pay for something they have not yet seen,
as well as removing the need for the whole traditional distribution
system, from wholesale to retail.
It is good for the users because the producers can have no intention
except to make products that the user will enjoy and/or find useful;
not to make products that merely appear valuable to the user, nor to
make the least expensive thing that will produce advertising revenue.
It makes the whole process honest, and aligns the self-interest of
each party with the benefit of the other.
While adequate tools for selling software over the internet don't yet
(and might never, depending on how you define "adequate") exist, tools
for sending donations are entirely adequate. E-Gold is the one I use as
an existence proof (there may well be others out there, and certainly
will be in the future, but I haven't found them).
In these days of massive unauthorized copying and unfunded
development, the traditional income model of withholding
access to software for lack of payment is inefficient and
unrealistic. Legal restrictions on copying can't be enforced
by any government that respects privacy and private property.
I've had friends who like to copy the FBI warnings at the
start of VHS tapes, and laugh at the scary punishments. The
warnings could say "If you even think of copying this, your
family will be enslaved for 10 generations and you will be
broken on the wheel." and they'd just laugh harder. Nobody
actually faces those penalties, at least not a significant
percentage. They don't care about the penalties for the same
reason they don't care what the grand prize in the lottery
is - it's less likely than getting struck by lightning out of
a clear sky.
Regardless, today's consumers buy more recorded music,
video tapes, video games, etc. - and generally at higher prices!
- than in the past. Piracy is rampant, yet sales have never
been better... how can this be? People aren't irresponsible,
they do ask themselves "what if everybody did that?" They
know perfectly well that if nobody paid for these things, they
Oh, they might not express it that way. They're more likely to say,
"That would be stealing." than "I'm protecting my future supply," but
the end result is the same.
"Buskware" and Other Words
I'm going to stretch the term "software" a bit to
include anything that can be stored as computer data:
not just computer programs, but web pages, movies, music, novels,
textbooks, etc. It is all "soft" in the sense that it has no
independent physical existence. I used to refer to this class of
things as "intellectual property", but this label is inaccurate, as
it is not always property. This definition of "software" is the
"ware" in buskware.
Software works by different economic rules than physical
products: once created, the inherent cost of reproduction and
distribution to the producer is nil; only when the
producer seeks to control reproduction and distribution does
it cost him anything. It logically follows that "distribution"
costs are actually control costs. The traditional term
"consumer" is not appropriate, because the end-user does not
consume the product, he only uses it; accordingly, I will use
the terms "producer" and "user" rather than "producer" and
Why You Should Pay
Very simply, those groups which pay will get better stuff.
Profitable enterprises will give the producer-recipient a
motive and the resources to produce more, and encourage
productive competition for your payments.
It's like choosing to vote. Your choice, as an individual,
is rarely important, but your choice as a member of a group
of like-minded people is extremely important. In a
sense, when you choose to vote or to pay buskware producers,
you are making the choice for everyone who shares your opinions
relevant to the choice.
The group has resources, desires, and a reputation. The
pool of producers will respond to the desires of the groups
which build good reputations through generous payment from
their own resources.
How Much You Should Pay
It is a clear absurdity for anyone to be "unable to afford"
something that can be produced for essentially no cost.
Equally absurd is to charge less than a person is willing
and able to pay, yet this is practically the only circumstance
under which commercial software is sold (if the price is
higher than an individual person is willing and able to pay,
then there is no sale; the case where the price is the exact
maximum that the buyer will pay is rare to the point of
insignificance). Shareware is as unsatisfactory a solution
to the payment problem as commercial software because of this
(although it does improve the distribution).
I can't give you a simple formula because there are too many
variables: wealth of the user, value to the user, likelihood
and desirability of related future development, funds available
to the producer, and so forth.
Clearly, people with more money should pay more, not in some
egalitarian-socialist way "because they can afford it," but
because it's in their best interests as it makes them more
important to the producers. Give a hundred times as much as the
typical user, and you will have a hundred times as much influence;
ten people who spend $10000/year on buskware will be listened to
as carefully as a thousand people who each average $100/year.
Do you want something specific from the producer? Send a note
about what you want, with a generous donation. The more influence
you want, the more you should send.
As a guideline, though, you will probably have to spend at
least 1/3 of what you spend on software now, carefully distributed,
if you want buskware that replaces traditional software. I would
very roughly but conservatively estimate that 2/3 of software price
are made up of retailer markup, middleman markup, distribution costs,
reproduction costs, promotion costs, intellectual property policing,
and all the other things that don't benefit the user in the slightest
way. Associated practices, such as component sharing, could further
reduce the cost of development and increase the quality.
When You Shouldn't Pay
The most obvious and best reason is: you didn't enjoy the works
of the producer, or find them useful, and you don't see any potential
for them to develop into something worthwhile if only they were
better funded. You certainly don't want to encourage bad work.
<p"They have enough money" is a valid reason, sometimes. Remember,
the buskware concept is based on self-interest. There's no need
to make a billionaire of the author of your favorite amusement when
making him a multi-millionaire will do and there are other producers
who may be a bit less interesting but still worth supporting, and in
greater need. At the same time, being too stingy and just allowing
producers to scrape by will kill the whole buskware concept and push
us back into the less efficient systems. If the most talented
people can't get rich from buskware, they'll go do something else.
Besides, useful companies should grow; if 3 people build a great
game in their garage, don't you want to see what they can produce
with ten million dollars and a large staff of artists? Plenty
of great companies were founded on a shoestring, but none is likely
to hang around on one for long.
Be very careful not to go too far down this path. There have to
be disproportionate rewards in the mass market, because it's such
a harsh and risky place. Even with the open distribution allowed
by the internet, not one in ten mass market entrepreneurs will turn
a profit, and unless one in a thousand gets rich, there won't be
any reason for talented individuals to participate. It should
be a harsh place, you shouldn't waste a million people's time with
anything less than the best, but it should also offer great rewards
to those who can offer the best.
In general, you should not pay before you've thoroughly examined
and tested the product through use and it has demonstrated its value
to you. When you are very enthusiastic about a new product, exercise
some restraint and just give a little, if long-term use shows it to
be as useful as you thought it was, you can always give more. If
it turns out to just look useful, you can't decide to grab
your money back later.
Don't give money to people who don't ask for it. Remember that
you're buying control; if they don't indicate their willingness
to respond to these payments and record them for their potential
competitors to learn about, you aren't getting anything for your
money. Some - strange creatures! - would even be offended.
Remember that influencing one particular producer is not as
important as producing the general condition that anyone who makes
a product that you like gets rewarded. This is influence over the
How To Make Money With Buskware
Give your product away, placing few or no restrictions
on redistribution, and ask for whatever payment the user thinks
is appropriate. Ignore all comments, requests, complaints, etc.
that are not accompanied by money, and pay close attention to
the ones that are. Look at the bottom of this essay for an
example of a buskware notice.
I prefer the simplicity of releasing my products into the public
domain. The existence of a central website from which I make all
my new releases is enough guarantee of credit for me, and I have
no objection to people making any kind derivative works. It is in
the best interests of people who make derivative works for them to
encourage me to produce more, just as it is for the end-users.
Annoyingly, there are large amounts of valuable code available
under mutually incompatible "free" licenses, making a lot of useful
If you just want to get your toes wet, try releasing some of
your older products as buskware, ones that you don't distribute
or don't bring in a profit anymore. It won't cost you anything,
and sensible users would pay you to encourage you to release the
rest of your old software and your new releases as buskware.
If you are already a free software producer, you truly have nothing
to lose by making it buskware. Why not give it a try?
You can't just hang back and wait for buskware to get going.
Major software producers aren't likely to just jump right in
because someone mentions the idea. First, the user-donors must
establish the general belief that there's money in it. This
might involve breaking a few of the rules that will make sense
For the user-donor: set aside a fixed sum of money that you
will spend on buskware each month; it doesn't have to be a large
amount. Spend it on the most worthwhile buskware you can find,
but spend it all, even if it doesn't seem entirely worth supporting.
In this early stage, your goal is to establish that buskware pays,
establishing that only good buskware pays will be easy enough later
However, it should not be hard to find worthy buskware producers,
if you break another rule and pester people who produce freely
available software as a public service to take the money you're
trying to shove into their hands.
There are also methods for the producer-recipients...
The fundamental action of capitalism is: capital is invested
to create product, which is offered for sale, hopefully resulting
in profit. However, these simple transactions are not always the
most attractive. Variations on these basic themes are offered:
renting, leasing, financed purchase, timeshares, etc. can substitute
for sales, and there is a similar diversity of methods for sharing
the risk of investment.
Similarly, the fundamental actions of mass-market busking are:
product is created and shared freely to the public in the hope of
attracting donations, and the public donates freely to those
producers whose work they enjoy in the hope of encouraging further
production. However, this might be a hard sell, unless and until
the idea is firmly embedded in our culture. To get the ball rolling,
there are things you can do to sell control more directly.
Street Performer Protocol is one such method. In this one,
the product is released in installments, and the threat is made that
the next installment won't be released unless a certain quota of
donations is met. The only problem here is that payments are likely
to drop off sharply once the quota is met. Stephen King is trying
something similar to this, though he is making the mistake of fixing
the price per user. It's a very good idea for things that are easily
seperated into installments.
The FlyingRat service, when
used as the only means by which
the producer can be communicated with and influenced, is another.
It only allows you to send email if you pay a certain price.
If someone wants to complain about a bug or request a new feature,
they have to pay you for the privilege.
This method can be improved by having different priority levels with
different prices ($0.50 - "it'll get to me", $5.00 - "I'll read it",
$50 - "I'll reply", $500 - "I'll do definitely something about it").
A more elaborate variant of the same method would be to have
a discussion group, where people can spend money to promote one
message or thread over another, to show you which areas you should
be concentrating on.
The problem with all of these methods is that they draw attention
away from what should be the primary focus (influencing the whole
industry) on to just considering the effect on this particular
producer. However, this might be a good tradeoff in some cases.
Component Sharing: the Icing on the Cake
Much has been written on the value of open source computer software,
so I won't go into that, except to say that there is clearly more value
in software for which the sources are available, and in mass-market
busking, more value for the user should translate into more income for
What I really want to talk about is extending this to include all
information products. This means unmixed tracks from music, the model
files from the CGI special effects in movies, storyboards, sound effects,
scripts, etc. Everything, release it all into the public domain.
If you are creating your own work from scratch, you aren't competing
with people who cobble together their own Frankenstein's monster from
bits of other people's work. They will be recognized and treated as
the second-rate works they are, like the fanfics and parodies of today.
In any case, it is the users' responsibility to ensure their money is
going to promote the producers that benefit them; there's no reason
they would pay more to the group that put the pieces together than to
the group that made the pieces.
The same direct benefit applies: more value means more income,
and it really doesn't cost anything. The real benefit, though, is in
developing the industry. It will become much easier for newcomers to
learn the ropes when they can see how everything was done, and get
some practice fooling around with good quality components.
Eventually, your audience will come to expect you, and all your
competitors, to release components. When somebody doesn't, there'll
be complaints and protests and boycotts, and it'll all add up to less
income for him.
Who Really Creates Value?
Consider operating systems: why does the mass market favour one
operating system over another? It isn't for superior features of the
operating system, those are mostly the concern of the software
developers, not the users. The users choose an operating system for
the applications available on it. Even the applications are often
chosen for compatibility.
Clearly, the main value of an operating system is not in the operating
system, but in the software available for it, and much of the value of
the software is not inherent either, but in who is using it.
This sets up the economically damaging situation of one entity
controlling access to value it doesn't create, like a corrupt gatekeeper
who only lets people go through when bribed.
The customers know it: they hate paying the gatekeeper to get
at what someone made on the other side, especially when the gatekeeper
recognizes that demand and charges whatever the market will bear.
Direct competition from one product can't fight it, a large segment
of the industry must simultaneously make the effort to move to
the new platform.
This phenomenon is not limited to computer software, though.
Everything from game systems to popular music derives much, if not
most, of its value from its popularity.
While it is very profitable to become a successful gatekeeper, it
is something that happens by luck, not by design or special skill.
The great Microsoft empire was founded on clumsy legal work by IBM,
when they commisioned and did a lot of the work of creating MS-DOS but
didn't secure any exclusivity agreements.
You can, however, take away the gatekeepers' advantage, so you can
compete with them and avoid paying them, by releasing freely
redistributable products which can never be used to extort money for
the privilege of benefitting from the efforts of 3rd parties. People
prefer to produce the added value when they know that nobody is going to
stand between them and their users demanding a toll. Look at Linux,
for example: there's much more free software for Linux than there is
for Windows (and much of it portable to any Unix system), partly because
the developers want people to be able to access their product without
having to pay Microsoft.
Big as the advantages might seem to try to get rich by controlling
access to the work of others, there is no reason for the consumers to
allow you to do so. A buskware product designed for the same role will
be preferred for the obvious reason that users hate gatekeepers, and
the redistributable nature of buskware makes it impossible to use it
to take this role.
The Corruption Inherent To Advertising
Advertising is manipulation. When a product is paid for by
advertising included within it, the producers are no longer working
for the users - they are working for the advertisers. Advertising
has its strongest effect when it is more interesting than the
content it is embedded in. Television is at its most profitable
when it is just barely interesting enough to keep people watching,
often through some promised revelation ("Tonight on the news: A
terrible new threat to our nation's children. What is it? We'll
tell you some time in the next hour!"), so the ads are a welcome
break from the show.
Advertisments rarely provide objective, useful information.
They frequently mislead and often outright lie. While a rational,
wary mind can easily ignore the ridiculous claims of advertisements,
and recognize their hostile, manipulative intent, advertisments are
usually targeted at people who are trying to relax and be entertained
when their minds are least likely to be carefully filtering out bad
information (even worse is when it invades spaces where it is not
expected, to catch the eye and mind of the unwary).
Advertisers are as much an enemy of the industries they work
for as the consumers they try to manipulate. When you're competing
against a massively advertised product, it's not enough to merely
make a better product or have a lower price, you've got to advertise
as much or more than they do. For example: look at the 1-800-collect
campaign, versus the 1-800-callatt service. 1-800-callatt could
make the simple and true claim that they were always cheaper
than 1-800-collect, and obviously nobody should have bothered with
their competitor. However, they didn't spend nearly the advertising
budget, and 1-800-collect carried on with their massive campaign,
and continued to survive despite offering an identical service for
more money, when both were available from any phone and equally easy
to use (I have no idea whether this is still the case). Like soldiers
and lawyers, you need to have as many or more advertisers than those who
oppose you, and have little other purpose for them.
While advertising tries to solve the same problem as mass-market
busking (denial of access until paid is impractical or undesirable),
it fails utterly in sustaining a direct link between the user and the
producer. Instead, it sets the user and producer at cross-purposes;
the user wishes only to view the producer's content, and the producer
needs the user to view the advertising.
What does this have to do with mass-market busking? Advertising
currently fills the role that mass-market busking was designed for,
and is often seen as the only viable tool for the job. Anywhere you
rely on advertising to support a service, mass-market busking can do
the same job without damaging the integrity of the service.
Why a Fixed Price Is Bad For Everyone
Buskware is emphatically not shareware. Shareware comes very close
to it, but still insists on holding a legal stick (theoretically) over
the users' heads, and on setting a fixed price based on a guess at
what people will be willing to pay. Crippleware is just commercial
software with a free demo and really has nothing to do with either
shareware or buskware.
The effect that fixing a price has is to exclude those unwilling or
unable to pay as much as the required price (in this case, whether they
pirate the software or just go without makes no difference to the
producer; either way, he doesn't get paid), and to limit the payments
of those who would be willing to pay more. Either way, the end result
is less income. Additionally, some people end up breaking
the law, others end up deprived of the benefit from the software, and
if there isn't enough income, all the users will be deprived of future
versions of the software.
The Free Software Island
Free Software is a nice idea, once you get past all the silliness
about it being a moral imperative. Everyone wants better software,
and we programmers want a look at the code. However, following the
practices of Free Software puts software production out on an island,
isolated from the rest of the economy. Software is produced for the
producers, and the users have nothing to contribute and no control
(unless they are also software producers). Any consideration of
the end users is an act of pure charity.
Where do the doctors, machinists, construction workers, farmers,
factory workers, and all the other people who don't produce software,
fit in? They have no influence in the Free Software world, despite
the facts that they want software produced for their needs, and that
they provide goods and services that the software people need and want.
What a step backwards in economics!
Of course, you could still hire programmers if their work is
worth enough to you to pay for it your self. This is probably
how most computer programs in daily use are produced, and nothing
is likely to change that in the near future. Custom development,
whether closed or open source, will always be important, and buskware
will have no effect on it one way or the other.
I would like to emphasize that selling software on the
honour system is not optional. You can't prevent unauthorized
copying. Right now, it is inefficient and many still have
inhibitions against it, but these inhibitions fade as people
see the endless variety of software that they can't afford,
but they can copy with minimal effort and no real probability
of being caught. Go have a look at
Napster if you won't take my word for it; as was inevitable,
internet systems have developed which wave copyrighted materials
before the people who want them, and it only takes a moment to
grab them. They will get better and more efficient as time goes
by, and make it easier, safer, and more tempting to make unauthorized
copies. The only question is, will you pretend that people can't get
away with it, or face reality?
If you are a commercial software producer, you can go around restricting
supply, suing people, and generally making "pirates" feel
justified, as if they're fighting some evil tyrant, or you can
drastically cut your overhead, deliver better products for less
cost, and still get paid while generating good karma. The
former will only work for a little while longer, while those
who follow the latter course will be following a business model
that will survive any possible change to intellectual property law.
This essay is buskware