A victory for the trolls

Posted Apr 25, 2011 23:43 UTC (Mon) by ajb (subscriber, #9694)
Parent article: A victory for the trolls

We know software patents are a bad idea. How do we convince non-experts?

There is an example which we can learn from, which is how Hernando de Soto persuaded the Peruvian government to reform their (real) property system. De Soto set up an experiment, in which his team of lawyers attempted to operate a small business to the letter of the law. This brought the absurdities of the official system into sharp relief: For example, to wade through the red tape necessary to register the business, took 289 days.

This sort of concrete example is much more persuasive than the kind of statistics or abstract reasoning which come from academic reformists, for all their good will.

Can this be applied to software patents? I think it can, but it would require some investment from the larger companies affected. An outline of such an experiment would be:

Obtain a moribund piece of proprietary software which no-one cares about, eg, from a bankrupt company (other than SCO..). [1]
Measure how long it takes to find out what patents it might infringe. Ideally, produce a graph of lawyer-hours versus survey accuracy.
Try to figure out how much it would cost to licence all the patents, if it would even be possible.

This won't be quite as convincing as De Soto's experiment, for various reasons, but I think it's well worth doing.

I go into more detail about this in my submission to the UK IP review. Which also notes an unexpected (to me, at least) connection between software patents and the halting problem...

[1] We probably don't want to do this exercise on software we care about, because it might cause legal problems for users.

to post comments

A victory for the trolls

Posted Apr 26, 2011 1:49 UTC (Tue) by wahern (subscriber, #37304) [Link] (6 responses)

This might be possible to do without having a huge wallet. The IP assets of your typical run-of-the-mill IT company are almost worthless. Larger companies don't buy moribund software companies for their software or patents, but usually for their employees. So an IT company that no longer has any employees is almost completely worthless, with no hope of paying off creditors.

And if we target a company whose primary business wasn't even software, but where software was merely collateral to their primary "business innovation", then it should be even easier to acquire, and more effective an example. Then we're showing that you can't do *anything* without paying these exorbitant fees.

One problem with this approach, though, is that it presumes the law expects compliance. To the contrary, most legal scholars would readily admit that as regards patent and copyright, most behavior is "unlawful". This doesn't just apply to patent and copyright. We commit torts all the time. Imagine how expensive it would be to ensure that we never illegally trespassed, or never committed a battery? It's implicit that we all make cost+benefit analyses regarding our behavior. This is ultimately the difference between civil and criminal law. It's not inherently wrong to "illegally" use a patent, or to break copyright law. But *if* you get caught, the law allows a transfer of assets. So you modulate your activity based on your expected risk of getting caught, and the law doesn't frown on this balancing act.

The only question then isn't whether you're forced to break the law. The question is whether the transactional costs--e.g. erroneous discounting of your liability (i.e. poor calculation of `damages x chance of getting caught')--are so enormous that the cost is unjustified. This is the case with the property regime in much of Latin America and elsewhere, according to de Soto. (IIRC The Mystery of Capital correctly.)

It's not clear this is the case with software patents. Or rather, given the presumed [if erroneous] benefits that policy makers believe software patents create it's not clear that the transactional costs are disproportional. Arguing that most businesses ignore software patents undercuts our argument. And even if we show high costs then the easiest solution, practically speaking, would be to make enforcement for software patents more difficult and not necessarily to make software patents more difficult to acquire, or scrapping them altogether.

So ultimately we're still left with showing that the absolute benefits created by software patents are illusory.

A victory for the trolls

Posted Apr 26, 2011 2:36 UTC (Tue) by jone (guest, #62596) [Link]

I guess you could illustrate the issue closer to home with something like recipe patents .. then open a restaurant, throw a party, or put together a recipe book that deals solely with infringed recipes. The temporal nature of food consumption seems sufficient to perhaps help illustrate the illusory nature around "intellectual property" patents.

What I find humorous in much of this is the almost deliberate taunting of the system. For example the Microsoft case a couple years ago against i4i (eye for an eye?) that claims infringement on an integrated XML editor in Microsoft Word but can't seem to even successfully use XML on their company website. And now Bedrock (Flinstone's reference?) that took some old patents from Bell Labs to try and cash in on a few foundational concepts in computing (funny how they reference Knuth 1973 - but nobody cries prior art.) Almost like suing a construction company for their use of a quarry.

A victory for the trolls

Posted Apr 26, 2011 6:40 UTC (Tue) by ajb (subscriber, #9694) [Link] (4 responses)

The huge wallet is to pay all the lawyers to read patents.

Part of the point of this experiment is that even if it were cheap to license all the patents, the cost of actually figuring out which patents you probably infringe is enormous. This is actually a significant difference between software and, for example, medicines. Medicines have a natural etymology - the chemicals that make them up. That means that keyword searching is a reliable means of narrowing down from the whole patent database to a short list of possibles to read. Keyword searching is much harder in software. A program's logic is in its structure, not the names of its variables. There are frequently several English words which can be used for the same concept (find, get, look up, retrieve, etc). Different communities often reinvent the same technique and call it by different names. Areas which have been around longer, such as automotive, settle on the same terminology over the years.

In principle, you could set up a system whereby pharmaceutical patents were described in a machine readable form, and you could search it to see if your new invention is already there, without paying any lawyers at all. Try to do the same for software, however, and I think you run into the halting problem (or more accurately, Rice's theorem: http://en.wikipedia.org/wiki/Rice%27s_theorem)

The fundamental question is whether the patent system, seen as a datastructure, *scales*. The real property system scales as the number of properties increases, because in the event of disputes, you only have to consider neighbouring properties, not potentially all other properties in the registration system.

A victory for the trolls

Posted Apr 26, 2011 8:58 UTC (Tue) by wahern (subscriber, #37304) [Link] (3 responses)

You have to discount the costs by the chance of enforcement. Discounting by risk is then the prospective cost, excluding transactional costs. If your potential liability is $5m for breaking the patent, but the chance of enforcement is only 1 in 10,000, then the prospective economic cost is $500. If executing a patent search costs more than $500, then just don't do the search. You might say that individuals don't think this way, but we do at least behave this way; and venture capitalists, with dozens or hundreds of investments do think in terms of probabilities and risk.

How many times have you turned around on a side street by poking the front-end of your car into someone's driveway? This is trespass, and the owner could sue you. But a court would probably only ever award a one-time offender $1 in damages. If a chance of suit is 1 in 1,000,000, then your cost is infinitesimal, and taking the risk is absolutely worth it. That explains why you never think about it. If the potential repercussion was death, you'd be doing a heck of a lot more curb-side three point turns.

The fact that you could be sued, and that it would be absurd in the grand scheme of things to actually be sued, doesn't mean we should modify the property regime to take this into account. There's no need to write in an exception for using driveways to turn around. The fact that enforcement is not absolute, and the limit of damages proportional to some real economic value, solves the problem for us. The homeowner will only sue when his potential return is worth the effort; and the fact that he's oblivious to most driveway incursions is significant as well--it's manifestly evident that he's only going to pursue a conspicuous offender deriving significant benefit from infringing his right.

So, when you speak with a policy maker about the "costs" of patents, you can't simply point to the costs of fastidiously avoiding infringement; how the garage inventor is oppressed by the specter of a corporation suing him in Federal Court. That's because the whole regime is predicated on the notion that enforcement is lax, and on the whole only people for who it's economically worthwhile to seek a license will do so, and conversely patent holders will only seek enforcement on a small subset of infringers capable of paying. The risk of that garage inventor being sued is miniscule, and he knows it; the same way nobody in this thread realistically believes he's going to be personally sued by Bedrock for using Linux.

Patent holders rarely sue small start-ups because it makes sense to wait for them to be gobbled up by a bigger fish, and *then* sue the bigger fish. We tend to think this is an absurdity and evidence of how illogical the system is, but in fact this is exactly how things are expected to work. If the small fish understand this, then it means there's not as much reduction in innovation as we might think. The small fish will infringe, and the bigger fish will expect to be taxed if and when they eat the small fish.

Now, that's the general model. Once you crunch the numbers, of course, most of us would agree that on the whole it's not worth it. Also transactional and collateral social costs (e.g. people overestimating their risk of suit and reducing economic activity more than what is warranted) can overwhelm the social benefit of the system, but it's not as easy to show as one might think.

This entire analysis applies equally to the real property regime. Once you start to break real property interests into all the different kinds of estates, like life and leaseholds, and interests, like tenancies and contingents, then the complexity explodes from the trivial model you're implying. And, again, it works because perfection (solving the halting problem, if you will) is not the goal. If it's difficult to determine or enforce a property right, then it will indeed be enforced much more rarely and lightly, and the value and potential costs of that right will be diminished accordingly. Consider that if you sell a house with a warranty deed you can charge a higher price than if you sell a house with a quit claim deed (where you don't warrant clean title). The value of the right to both seller and buyer is diminished by the risk of enforceability of the right.

A victory for the trolls

Posted Apr 26, 2011 17:29 UTC (Tue) by ajb (subscriber, #9694) [Link]

Low probability of enforcement is evidence that the system is not working correctly, for a couple of reasons:

You can't simply factor out the probability of enforcement, because uncertainty also has a cost. And it's a cost to the system as a whole, not just a transfer between parties, because uncertainty causes friction. (It may seem peculiar to say that uncertainty is a problem for entrepreneurs - aren't they supposed to be all about taking risks? But we want them to spend their risk-taking capacity on innovation, not working round legal uncertainty).

This degree of uncertainty over property ownership is not seen in other areas of property. Yes, discounting by risk occurs in real property, but the vast majority of real properties - at least in the developed west - have much more certain ownership; with discounting only used to get rid of residual risk. We have seen from the recent mortgage debacle that it causes huge stress if too much uncertainty is allowed into the system. In developing countries, uncertainty over property ownership is a huge drag on the economy. Even in patents, in other industries, such as pharmaceuticals, small companies rely much less on assuming that enforcement will not occur, and actually do licence. That is why the issue in pharma is whether there is a 'patent anticommons'.

You make the same argument that Paul Graham makes (http://www.paulgraham.com/softwarepatents.html), that trolls are content to wait for the bigger fish to buy up the smaller ones. We can assume that he knows what he is talking about. But we can't assume that it will continue that way. Not all smaller companies are bought by bigger ones; in fact, small companies make up a large proportion of the economy. Trolls will want to monetise them if they can, and we can expect them to become more efficient at doing so over time (eg, via exchanges such as http://www.ipxi.com, which is implementing the ideas in Lemley and Myhrvold's "How to make a Patent Market (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1012726)). Also, it is not necessarily the case that infringing a patent simply incurs some percentage 'tax' on profits. In a perfectly efficent world, each troll would negotiate separately with each infringer, but that can't happen, so we see fixed prices which prevent certain uses, such as in the MPEG pool. So the liability created by patents can make some businesses uneconomic, and this will become a larger effect once smaller companies are drawn into the system.

Secondly, the uncertainty is not just due to the low probability of enforcement, but due to the inaccuracy of determining whether you are infringing a patent. This inaccuracy dilutes the incentive effect that is supposed to be the reason for patents.
You are right that we can't obtain perfect accuracy, but the interesting question is how much does it const to obtain enough certainty for the system to actually work. That's why I suggest measuring how much information you can get for a given cost.

Also, from a political point of view, it's worth challenging the simple narrative that proponents of software patents push - in which software patents are a 'no-brainer' benefit to the economy. Exposing the real situation is well worth doing.

A victory for the trolls

Posted Apr 26, 2011 18:07 UTC (Tue) by copsewood (subscriber, #199) [Link]

Bankruptcy administrators are obliged to monetise the bankrupt's assets including patents. This might most conveniently occur through sale of the patent to a third party, but if no-one wants to buy it the administrator is obliged to enforce the patent against potential licensees. If the startup offered a license can afford neither to license it nor to litigate it, the administrator is likely, if it takes the matter to court, to obtain the startup business and it's assets as it stands. In this situation the only strategy available to the startup owner to avoid losing the business is to threaten to spend all assets attempting to defend the suit until the business is worthless, in which event the administrator will see a negative asset value due to unrecoverable legal costs. But this is a dangerous bluff if considered as such and called. I've heard of once such case where a US small software business owner took this line and lost everything in the process.

A victory for the trolls

Posted Apr 28, 2011 8:30 UTC (Thu) by ekj (guest, #1524) [Link]

This makes sense up to a certain point, but there's huge problems with handing nuclear weapons out on the *assumption* that those getting them will generally tend to not use them against small fish often.

The *law* is supposed to decide who is punished and who isn't. (offcourse modified by the fact that the chance of getting caught ain't 100%)

But increasingly, we have laws where "Everyone is a criminal, few are prosecuted, but the penalties for those few who are -- are enormous."

This hands a lot of power to the people deciding who to prosecute. Power they're not supposed to have. A big company with many patents, say IBM, can at any time decide to destroy any small software-company they dislike by claiming patent-violation. They can do this for any reason whatsoever, though you're presumably right that most of the time economical considerations will be the most common reasoning.

It's not a good thing for competition to have an environment where any of the big can, at any time, take out any of the small. Offcourse being big they might be able to do that in any number of ways, but that's no argument in favor of giving them even -more- ammunition than they've already got.

A victory for the trolls

Posted Apr 26, 2011 4:59 UTC (Tue) by FlorianMueller (guest, #32048) [Link] (1 responses)

Sorry to say so, but this approach won't impress political decision-makers in the slightest. First of all, it's just a hypothetical thing in terms of the licensing cost. In practice, not every patent is asserted against every potential infringer (not even in the smartphone space, though there's a lot of enforcement going on there). Politicians don't care about what might theoretically happen if it doesn't happen in practice. If the likelihood of this kind of total patent enforcement scenario is so low that maybe one software company will go out of business every 150 years, they won't care. Moreover, the problem you point to could also be solved through other means than abolition, such as ensuring that the total amount of damage awards to all patent holders is only a certain percentage of a product's overall economic value.

I explained a while ago right here on LWN that politicians won't abolish software patents until they see that software companies really suffer and approach them with calls for abolition, and that those calls aren't just based on filling out a web form in five minutes but that there's a serious effort bein made. In terms of the constituency that politicians would care about, I mentioned what a staffer of the conservative group in the European Parliament told an anti-software patent activist years ago. He said: bring on those middle-aged closed-source entrepreneurs with beards, bellies and glasses.

A victory for the trolls

Posted Apr 26, 2011 7:12 UTC (Tue) by ajb (subscriber, #9694) [Link]

Indeed, but I think such an experiment would play a useful part in persuading closed-source entrepreneurs of the dangers of software patents. In the long term, it is the growth of trolls which will do this, but we need them to understand before the trolls actually win. (If patent markets such as http://www.ipxi.com/ become entrenched, it will be impossible to reform the system).

I don't think closed source people quite understand their potential exposure right now. That is what this kind of experiment can change. We need to show them that the software world is effectively living in a 'black economy' - like the third world entrepreneurs who can't register their property.

One of the things about a black economy, is that businesses can't appeal to the law to protect themselves. We are not quite to that point in the software world, but you can imagine analogous problems. For example, businesses are often concerned about the 'disgruntled employee' problem. If a disgruntled employee offers a rival company his employers trade secrets, they will usually report him to the police, because they would get into legal trouble otherwise. But suppose instead a disgruntled employee offers evidence of patent infringement. This kind of thing already happens in the copyright world, with 'shop your employer' lines for software licenses. It's not much of a stretch to imagine it happening over patents.

A victory for the trolls

Posted Apr 26, 2011 17:42 UTC (Tue) by fritsd (guest, #43411) [Link]

Sell your commercial software under two licenses: one, normal priced, with a sticker "not for sale or use in the USA and Japan", and another one, 100x more expensive, without that sticker and with lots and lots of pages of legalese about the kind of software patents infringements that the software license doesn't indemnify, and why you should be careful using software in "software-patent-land".

Once this price differential has come to the attention of corporate buyers in the USA, it won't be very long before their software patent laws suddenly are not a problem anymore :-) Because it suddenly puts a visible price tag on the problem.