Over a million developers have joined DZone.

On False Analogies

DZone's Guide to

On False Analogies

· Java Zone
Free Resource

Microservices! They are everywhere, or at least, the term is. When should you use a microservice architecture? What factors should be considered when making that decision? Do the benefits outweigh the costs? Why is everyone so excited about them, anyway?  Brought to you in partnership with IBM.

A fair number of things have intersected over the last few weeks and it's time for me to do a quick exploration of them. There's the Aaron Swartz persecution, robots, RMS and free software, zero bounds and trillion dollar coins, the state is losing control of money, and Kirtsaeng.

I'm a lawyer by training and I worked in a high tech law firm briefly. The law is a set of creeping analogies and we're at this interesting intersection of how to deal with the radically reduced transaction costs that the Internet and computers/logistics have brought to our doorstep.

Craig's List and FSF

Starting with the Econ 101 premise that marginal price should equal marginal cost and because the marginal price of another copy of software is zero, the marginal price should be zero. This is the technical basis for arguing that software should be free. There's a moral basis that has to do with individual freedom because RMS knew that software would eat the world long before Marc Andreesen was coding a browser. RMS understood that keeping software free avoids the pitfalls of ecosystems that we can't leave.

The challenge with free software is recapturing the fixed costs associated with producing the software (not the marginal cost). But Open Source software is alive and well and seems to be proving that software can be free, people will produce it, and the quality of open source may be higher than the quality of proprietary software.

Craigslist is another example of pricing something at marginal cost. Craigslist decimated newspaper revenues by breaking the oligopoly pricing of local newspapers.

Both free software and Craigslist have had a significant negative impact on existing players by reducing profits (okay, perhaps Apple has something to do with Microsoft's profits), but if we look at the overall social good, we have better software and more people selling more stuff (more transactions) when the pricing is closer to the Econ 101 ideal… and with the Internet, we are able to reduce the transaction costs of communicating about stuff (software development, availability of goods and services, etc.) and drive prices to their ideal level.

Sony vs. Universal (the Betamax case)

We saw the same thing happen in the 1980s with the Betamax case. Sony used new technology to give consumers more control over the way they consumed media.

The existing media companies were worried because the way they priced and distributed their products was under threat… if people could record shows and watch them at a different time (time shifting) and maybe even skips ads, then the existing business models would not work.

The Supreme Court in a very good for all concerned decision chose consumer freedom over existing business models.

Turns out that having consumer freedom increased media consumption. Increased media consumption led to more ways for media aggregators to package and distribute media. This led to the most excellent outcome of more media being produced, more media being consumed, and more revenue and profits for those involved in producing and distributing media.

Please go read Universal's briefs from the case. They claim gloom and doom for the media industry. Turns out that they were dead wrong. Turns out that allowing more consumer choice and reduced transaction costs involved with consumption led to increased consumption and when you've got increased consumption of something that has near zero marginal cost of production (it costs almost nothing more to distribute a TV show to 1 person or 50 people), everybody wins.

Corporate Profits should be near zero

Just a side note… if we have perfect competition in our marketplaces… if our markets are operating at near utopian Econ 101 ways, corporate profits should be near zero. So, when we measure quality of economic outcomes by corporate profits, we're saying that we want distortions in our markets because only distortions lead to high corporate profits.

We need a better measure of social welfare and overall economic good. We've seen in the examples above that existing corporate profit mechanisms may be destroyed when we societally allow new technology to disrupt (crap… I didn't want to use that word) existing pricing and distribution mechanisms. But at the same time, businesses find ways of generating revenues and profits and jobs in the alternative scenarios… the scenarios where technology does in fact drive prices towards zero because when consumption goes up (this is especially nice when the marginal damage to our planet of consumption is near zero), there exist business opportunities.

Hacking vs. Trespass

Back in 1991 when I was a young and arrogant, newly minted lawyer, I got into an excellent flamewar with the EFF's first lawyer.

As a side-note… those of you who think I'm arrogant now would not want to have met me in 1991-1995. These days my arrogant goes to 11… in those days it went to 12 or maybe even 14… and it's log-scale. But I digress.

The EFF was defending a bunch of hackers that broke into a system, wandered around, and did little or no actual damage (except to the egos of the folks that claimed the systems were secure.) The EFF was trying to invent all kinds of theories to deal with the case. My suggestion was very simple: treat it as trespass to information. Treat the hackers just like you'd treat someone who trespassed onto a farm. Yeah… if there's actual and measurable damage, then it's a problem. But for the kids who are just out virtual cow tipping, treat them the same way you'd treat obnoxious teenagers who cut across farmer Brown's farm.

But more importantly, choosing the right analogy, especially for the law, is super important.

The Law and analogy

The law is a beautiful thing. It's the collective thinking of some of the greatest minds the world has ever known. It's hundreds of years of evolution of thinking about balancing the needs and wants of an ever increasing population in the face of changing technology.

The law is active sociology with a most excellent feedback circuit.

The law is driven by analogy… this new thing is like old thing A but not like old thing B, so we should start evaluating the rules about this new thing using the rules we apply to thing A and then tune them based on the differences between thing A and this new thing.

But getting those analogies right is very important because the analogies drive the direction of thought for the way the law treats the new technology.

Property is just a social construct

When I was in law school, I got in an interesting argument with my dad about the nature of real estate. My claim was the real estate (houses, condos, etc.) were simply social constructs and entries into ledgers at city hall. There was no intrinsic concept of real property ownership… it was a social convenience that had a long and storied history.

Having a dog, I realize that it's part of our nature to mark our territory. Having kids, I realize that "mine" vs. "yours" is wired into our brains.

When done right, the way we allocate and mark property as mine vs. not mine can radically reduce social and transactional costs.

Put another way, of the choices of buying a piece of property and paying a small amount of taxes on the property to have the state insure that I have the benefits of my property vs. having to hire 10 thugs to guard my property and insure its mine (the libertarian utopia), I'll go with the former. It costs a whole lot less and allows me a lot more freedom in the use of my money/resources.

Mine, Mine, Mine… it's MY THING

The problem is that we as a society don't always get the balance right.

This is becoming increasingly apparent related to intellectual property.

We have lots of time-tested constructs for dealing with tangible goods. Goods that have a physical manifestation. Goods that cost money to produce (although piracy of the bible was commonplace in the 1500s), so we've rarely had to deal with the zero marginal cost problem. Goods need to be physically transported and there generally needs to be some intent to posses a physical good (one does not accidentally walk into Good Vibrations (oh, I thought it was a record store that sells mostly Beach Boys) and have a sex toy "cached" in their pocket but one can accidentally go to a web site and have obscene material cached on their hard drive).

We have a lot of people that mistake cash flow for the promotion of the arts and sciences.

We have people who measure the harm of file sharing at thousand of dollars per file. This is absurd. Aaron Swartz was prosecuted to death for "stealing" things that were costless anyway. Further, the prosecutor tried to frame Aaron's actions as causing lots of actual harm by analogizing the copying of academic articles to physically taking possession of a valuable tangible good.

Property is a valuable social construct. But we are analogizing things that have zero marginal cost to duplicate and zero actual harm in the act of duplication (there may be other harm which I'll get to in a minute) to the physical world where someone taking possession of my computer denies me access to my computer.

Measuring the harm

But copying has zero marginal cost and does not deprive the owner of the use of the good. Where copying does cause harm is if the copying denies the owner of the revenue that the owner might otherwise have obtained had the owner been able to exchange money for the grant of permission to make the copy.

The convenient way to measure that harm is by taking the retail price of the permission grant and multiply it by the number of copies that were made. But this analogy doesn't work.

How to price things

The problem is how to price things. What price would a specific buyer pay for a specific good. It's a really hard problem. It's a hard problem because each buyer has a different marginal utility curve for a given good.

The issue raised in Kirtsaeng is one of the right to marginal pricing vs. using a legal loophole to control a good after the initial sale.

In Kirtsaeng, the publishers want to price good differently based on the marginal utility curve of the likely purchasers. The publishers use geography as a nice proxy for marginal utility because students in the US will be higher on the marginal utility curve because they are wealthier than students in other countries.

The problem in Kirtsaeng is the cost of transport and the transaction costs of sales for goods has dropped radically because of computers/logistics and the Internet. So, either the publishers lose a tool for marginal pricing or everyone who makes goods overseas gets to control the disposition of those goods beyond the first retail sale.

Measuring harm and computing marginal pricing is the same issue.

It's a hard issue, but I think at the end of the day, we should not look at lost revenue based on retail pricing. I think we need to look at harm based on societal harm… lost incentives to produce useful arts and sciences.

This cuts an entirely different way. Nobody cares about someone who pirates a song from an unknown band. But pirating a song from an unknown band is likely to deprive that band of a lot more marginally useful revenue than the loss to the Rolling Stones of pirating a copy of Sympathy for the Devil.

Rise of near-zero marginal cost

I think we need to help courts and prosecutors and Congress focus on the core issue: how can we create incentives for people to create more arts and sciences?

How do we create a compensation system based on near-zero marginal cost of distribution (even for goods)?

The way our founders viewed the issue was to create a structure of temporary monopolies for the creators of the arts and sciences so that they could extract higher than marginal cost rents for the goods and this in turn would create incentives to create more arts and sciences. Yay!

But the way we've perverted the approach to the monopoly rents is to measure the value in terms of corporate profits rather than some broader social good. We've also perverted the measure of harm for breaching the laws and the way we've perverted the measure has led to a very tragic incident and many social harms.

It's an allocation problem

But, at the end of the day, it's an allocation problem.

How do we allocate enough resources to make it economically feasible for someone to create a work of art or make a new scientific discovery?

The issue is exacerbated by the whole robot issue. If the marginal value of many traditional types of human labor is dropping because that labor can be done by machine, how are we going to price the value of human endeavors such that there's a good allocation of wealth such that there's a broad base of consumers and a broad base of people who have enough resources to satisfy the bottom two levels of Maslov's Hierarchy of Needs?

We use money as a social convenience and a nice way to bridge between the current delivery of value and future consumption. Money is a valuable way to measure and store excess current value so that it can be converted into something of value in the future.

Pricing money is a fun sport and the price of money really does vary over time. We measure the gross variation as inflation, but inflation is an aggregation and a way of reallocating actual value among stake-holders: more unexpected inflation is a boon for debtors (the actual value of the debt decreases in real adjusted terms) and less unexpected inflation is a boon to the rentiers (the actual value of the debt is higher in real adjusted terms.)

We play games with money pricing in terms of manipulating inflation, government debt vs. GDP, etc. But at the end of the day, it's just another tool for allocation of wealth among folks that trade current cash flow for long term assets (people who collect pay checks and take out mortgages and car loans) and folks who trade blocks of cash for cash flow over time with a premium for turning cash flow into blocks of cash.

Part and Parcel

But the whole issue of the law dealing with this new near-zero marginal cost world and dealing with the marginal value of many forms of labor getting radically reduced and the way we price money is all part and parcel of the same issue.

We are at a brink of something potentially wonderful in the world. We are at a point where the amount we can produce is radically higher than it ever has been. We are at a point where the harm to the world of producing many of these things is much lower because the goods are increasingly "virtual". We are at a point where state is losing control of money because BitCoin has demonstrated that cryptography can substitute as a trusted source of limited value stores in much the same way that governments and treasuries have in the past.

Okay… I've laid out the landscape and I can imagine a world where we have figured out how to allocate the vast increases in GDP while at the same time not spending so much effort protecting the monopolies that control intellectual property.

Sadly, I'm struggling on how to get to that place. I'm struggling because there are so many intrenched players that will come out as losers if we move to a more "remix oriented" approach to content creation. I'm struggling because there's a huge bias against the kind of moderate inflation that could bring us out of the zero lower bound and help get the unemployment level back to normal.

The last time were were at a crossroads like this, it took a world going to war and the detonation of a few nuclear bombs to get the allocation issue under control.

I am hoping the Aaron Swartz's death will be one of the worst points in the transition that we're going through as a society. It would break my heart if we needed a world war… death on a massive scale… to once again restore balance to the allocation of GDP and to help folks see beyond the pettiness of "mine mine mine" and see the societal value of an open source approach to virtual goods and governing.


Discover how the Watson team is further developing SDKs in Java, Node.js, Python, iOS, and Android to access these services and make programming easy. Brought to you in partnership with IBM.


Published at DZone with permission of David Pollak. See the original article here.

Opinions expressed by DZone contributors are their own.


Dev Resources & Solutions Straight to Your Inbox

Thanks for subscribing!

Awesome! Check your inbox to verify your email so you can start receiving the latest in tech news and resources.


{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}