A new commerce model for the software market?
Submitted
by
tambo
tambo writes "I've been thinking a lot recently about how software is sold. I've felt that the pricing of software has been pretty crazy for a few decades, but I haven't been able to specify the basis of my feelings.
Today, the crux of the problem finally occurred to me: Standard models of capitalism are unsuitable in the absence of scarcity.
As we all learn in Economics 101, standard models of commerce are based on the laws of supply and demand — based on the idea of achieving an optimal price point for a particular article. Both of these curves are critically based on scarcity — the concept that the transaction is about a particular, physical article. The supply side is critically driven by the cost of creating that particular article — the labor and materials needed to manufacture it, the shipping of that product to a store or customer, and associated costs (the costs of building the factory, maintaining a retail store, etc.) And the demand side is critically driven by the idea that only (n) number of articles are available, and that the price should be set to the highest value that (n) customers are willing to pay. In a nutshell, the point is to maximize the utility of the scarce number of goods by providing them to the consumers who want them the most.
None of these concepts apply to a market without scarcity... like software.
Software is increasingly deployed electronically. Consumer-level storage capacities continue to grow at an astounding rate (1 terabyte drives for $100?!), and bandwidth and server costs continue to drop. As a result, the per-item cost of deploying a piece of software to a customer is essentially zero. (Sure, deployment servers and bandwidth cost money, but it's trivial on a per-software-deployment basis — especially with economies of scale for consolidated libraries like Steam and the App Store.) Additionally, the traditional model is deeply economically classist: only rich people can afford the best software, and even though computers are increasingly affordable, poorer users can't actually afford the software. (Are the copyright lawsuits inherently economically classist? Do they disproportionately target poorer computer users, who may resort to piracy to obtain software that they can't afford? Should we regard this as a socioeconomic inequity?)
Viewed from a pure utility perspective, the market should not be organized to ration the deployment of the software, but to deploy the software to everyone who wants it.
This doesn't mean that supply and demand aren't relevant to this market. They are still relevant — they dictate choices about what products may be developed — but they need to be reformulated in the absence of scarcity.
Things that won't work:
* Free-as-in-beer software. Professional development and quality cost money. There has to be some sort of payment mechanism.
* Pay-what-you-want software. These models never work well, because there is no incentive for customers to pay more than the absolute minimum.
* Centrally sponsored and dictated software, where the owner of the distribution system declares what software is to be created next. Again, these models never work, because centrally predicting demand is always inaccurate, if not outright corrupt.
Here's my idea: A software repository with a flat monthly user fee, where developer royalties are apportioned based on popularity.
Here are the details:
* For a flat periodic fee, each customer gets total access to the entire software library. They can install whatever they want — no quotas, no holds barred. Their choices determine demand, and their demand is limited only by their time and interest. Better still, because there is no incremental cost for new software, there is no reason for a user *not* to try any particular product.
* The owner of the distribution platform takes a cut to cover its administrative expenses, but the vast majority of the collected funds go back to developers *in proportion with the popularity of the product*. That is, the subscription revenue collected from customers each month is apportioned to developers based on the use of their products for that month.
Benefits of this system:
* High utility through unlimited deployment. Everyone who wants a piece of software gets it.
* Economic equality. Again, everyone who wants a piece of software gets it — regardless of their socioeconomic status.
* A tight coupling of developer payment to demand — and to *continued* demand. If users keep using the software, the developer keeps getting paid, even though the customers don't need to keep paying *specifically* for the software.
* A more efficient reward system that is not pointlessly limited by inapplicable concepts of scarcity. Instead of losing sales to customers who are priced out of the market, developers benefit from the broadest possible deployment.
* An end to piracy. There is no need to pirate the software when it is free for all subscribing customers. Demand can be better gauged than conventional sales numbers that do not reflect pirated deployments.
* An end to the secondhand software market. Used software markets are inefficient, but are necessary to offset the pointless dependency of the current market on scarcity. (Poorer users who can't afford the retail price can use the software cheaper, but later, and only when retail-purchasing customers are done with the software.) This inefficiency is eliminated in this market — every deployment directly benefits the developer.
* A focus on value. Developers do not need to choose projects based on sales figures. Rather, developers strive to produce software that will achieve the broadest deployment — i.e., the highest popularity, based on the best value presented to customers. Accordingly, this model kills many "features" that benefit the developer at the expense of the user, such as paid-for ad-based sponsorship, sales of private user data, paid-for downloadable content (which is another form of socioeconomic inequity), and unwarranted "leased software" schemes (which often extract ongoing money from a customer without a return of value).
The biggest problem is cheating: a central repository is a potential source of corruption, as we've seen in Apple's draconian control of the availability of products on the App Store. The central repository might also lie about demand in order to skew royalty payments and control the market or sell private user data (a la Facebook). This could be offset by making the central repository an unaffiliated, open organization — maybe a nonprofit. Even better, its administrative costs could be totally sponsored by a government, thereby eliminating the motivation to seek additional funds.
Thoughts?"
Today, the crux of the problem finally occurred to me: Standard models of capitalism are unsuitable in the absence of scarcity.
As we all learn in Economics 101, standard models of commerce are based on the laws of supply and demand — based on the idea of achieving an optimal price point for a particular article. Both of these curves are critically based on scarcity — the concept that the transaction is about a particular, physical article. The supply side is critically driven by the cost of creating that particular article — the labor and materials needed to manufacture it, the shipping of that product to a store or customer, and associated costs (the costs of building the factory, maintaining a retail store, etc.) And the demand side is critically driven by the idea that only (n) number of articles are available, and that the price should be set to the highest value that (n) customers are willing to pay. In a nutshell, the point is to maximize the utility of the scarce number of goods by providing them to the consumers who want them the most.
None of these concepts apply to a market without scarcity... like software.
Software is increasingly deployed electronically. Consumer-level storage capacities continue to grow at an astounding rate (1 terabyte drives for $100?!), and bandwidth and server costs continue to drop. As a result, the per-item cost of deploying a piece of software to a customer is essentially zero. (Sure, deployment servers and bandwidth cost money, but it's trivial on a per-software-deployment basis — especially with economies of scale for consolidated libraries like Steam and the App Store.) Additionally, the traditional model is deeply economically classist: only rich people can afford the best software, and even though computers are increasingly affordable, poorer users can't actually afford the software. (Are the copyright lawsuits inherently economically classist? Do they disproportionately target poorer computer users, who may resort to piracy to obtain software that they can't afford? Should we regard this as a socioeconomic inequity?)
Viewed from a pure utility perspective, the market should not be organized to ration the deployment of the software, but to deploy the software to everyone who wants it.
This doesn't mean that supply and demand aren't relevant to this market. They are still relevant — they dictate choices about what products may be developed — but they need to be reformulated in the absence of scarcity.
Things that won't work:
* Free-as-in-beer software. Professional development and quality cost money. There has to be some sort of payment mechanism.
* Pay-what-you-want software. These models never work well, because there is no incentive for customers to pay more than the absolute minimum.
* Centrally sponsored and dictated software, where the owner of the distribution system declares what software is to be created next. Again, these models never work, because centrally predicting demand is always inaccurate, if not outright corrupt.
Here's my idea: A software repository with a flat monthly user fee, where developer royalties are apportioned based on popularity.
Here are the details:
* For a flat periodic fee, each customer gets total access to the entire software library. They can install whatever they want — no quotas, no holds barred. Their choices determine demand, and their demand is limited only by their time and interest. Better still, because there is no incremental cost for new software, there is no reason for a user *not* to try any particular product.
* The owner of the distribution platform takes a cut to cover its administrative expenses, but the vast majority of the collected funds go back to developers *in proportion with the popularity of the product*. That is, the subscription revenue collected from customers each month is apportioned to developers based on the use of their products for that month.
Benefits of this system:
* High utility through unlimited deployment. Everyone who wants a piece of software gets it.
* Economic equality. Again, everyone who wants a piece of software gets it — regardless of their socioeconomic status.
* A tight coupling of developer payment to demand — and to *continued* demand. If users keep using the software, the developer keeps getting paid, even though the customers don't need to keep paying *specifically* for the software.
* A more efficient reward system that is not pointlessly limited by inapplicable concepts of scarcity. Instead of losing sales to customers who are priced out of the market, developers benefit from the broadest possible deployment.
* An end to piracy. There is no need to pirate the software when it is free for all subscribing customers. Demand can be better gauged than conventional sales numbers that do not reflect pirated deployments.
* An end to the secondhand software market. Used software markets are inefficient, but are necessary to offset the pointless dependency of the current market on scarcity. (Poorer users who can't afford the retail price can use the software cheaper, but later, and only when retail-purchasing customers are done with the software.) This inefficiency is eliminated in this market — every deployment directly benefits the developer.
* A focus on value. Developers do not need to choose projects based on sales figures. Rather, developers strive to produce software that will achieve the broadest deployment — i.e., the highest popularity, based on the best value presented to customers. Accordingly, this model kills many "features" that benefit the developer at the expense of the user, such as paid-for ad-based sponsorship, sales of private user data, paid-for downloadable content (which is another form of socioeconomic inequity), and unwarranted "leased software" schemes (which often extract ongoing money from a customer without a return of value).
The biggest problem is cheating: a central repository is a potential source of corruption, as we've seen in Apple's draconian control of the availability of products on the App Store. The central repository might also lie about demand in order to skew royalty payments and control the market or sell private user data (a la Facebook). This could be offset by making the central repository an unaffiliated, open organization — maybe a nonprofit. Even better, its administrative costs could be totally sponsored by a government, thereby eliminating the motivation to seek additional funds.
Thoughts?"