That article sounds like it supports a corrollary of rayiner's theory, that men are sick of having to "earn it". The article certainly seems to support that it is indeed women who control sex, and anytime they "settle for less" they feel like they're "giving it up".
Little do they know that even in marriage the dynamics don't change much, only the number of people involved (married people stereotype joke :-))
"Detecting Money Laundering and Terrorism Financing Activity in Second Life and World of Warcraft", Angela S M Irwin, Jill Slay - University of South Australia
Heh, I'm surprised there have been no New Yorkers yet chiming in to say "good riddance". Most New Yorkers -- in my experience -- seem to share the sentiment that there are too many tourists already.
Oh boy, that paper is a huge exercise in setting up a strawman and knocking it down. Lemley has done some good work, but this is not it.
1. He goes on and on about "patent theory", and it's worlds apart from how patent systems work in practice. His feeble connection is that practice is based on theory.
2. He cites a paper that directly contradict what he's saying.
2. a) He cites a paper to ostensibly support his theory that patent offices are pro-individual, but the reference actually says, without paraphrasing, "The patent laws do very little to prompt this view." It actually concludes that the lone-inventor rhetoric "has done little to stave off the increasingly anti-individual-inventor changes in substantive patent law."
2. b) He perpetuates the same "Watt stopped Steam" myth that Boldrin and Levine make. And to support this, he cites the Turner/Selgin paper that asserts precisely the opposite!
3. Further, regarding steam, he insinuates patent law didn't work as expected because Watt's patents promoted steam research by forcing workarounds. In fact, innovation through forced workaround has long been a known (though possibly post-facto) rationalization of the patent system.
4. All the examples he gives of "multiple independent invention" are actually examples of multiple individuals independently inventing different inventions. And he further proves it by saying that many of those inventors secured patents for their own inventions as well!
5. All the other parties working on the "genius" inventions did contribute to overall knowledge, but a) their work was not wasted since they did things differently, exploring the problem space further, and frequently got their own patents, and b) those inventions that ended up being remembered as genius were actually the commercially more successful ones, typically because they were either technically superior or their inventors were sharper businessmen.
6. He implies accidental inventions are less deserving of protection, completely omitting that those inventions all happened in labs or research environments by people who were actively experimenting.
Throughout this paper, Lemley consistently makes two mistakes:
* He conflates the broad “idea” of something with the specific implementation that makes that something work;
* He conflates popular narrative about patents with the patent system;
And based on these two, he enumerates flaws in popular narrative where most major inventions were not invented only by those that got all the credit.
And then he completely fails to show how this flawed narrative has anything to do with the patent system, which, by his very own examples, does reward individual and incremental contributions.
My problem with that book is that the authors start it with a lie [1], and changed to different lies when being called out on it [2]. They also made some misrepresentations of how software and open source works.
Just from that, it was difficult for me to take anything else they say at face value. To top it off, I just came across a reference [3] suggesting Boldrin and Levine's analysis of the dye market was inaccurate, and that patents actually played a big role in German dominance in the market.
The one thing I've realized from reading a lot of studies, anyone that makes a broad claim that IP if "all good" or "all bad" is wrong.
Did someone say "research" and "patents" in the same sentence? :-)
I was going to rifle through my vast collection of references to studies about patents, but I just came across a pretty good review of the research of economics of patents:
"Recent Research on the Economics of Patents" - Bronwyn H. Hall, Dietmar Harhoff (Google for PDF)
I'm still going through it, but it will give you an idea of what the current research looks like.
Here's the deal: Whether patents promote innovation, and whether their benefits outweigh the harms, are extremely difficult questions to answer. I mean, how do you even measure "innovation"? You could say by "counting patents", but that's simply a circular argument! Not to mention the complaints that many patents are low quality, or that companies like Twitter are regularly called "innovative".
So in absence of any direct indicators, what the vast majority of studies do is roughly this:
1. Pick a few metrics that act as proxies for whatever they measure (innovation, patent quality, economic benefit, economic harm, productivity, R&D efforts, etc.);
2. Gather data from which these metrics can be gleaned, typically constrained along many dimensions such as time, industry sectors, sources, etc. to make gathering it feasible. (Sometimes this step is actually optional and author outright just run simulations on what they think are "reasonable" approximations of data. Sometimes they don't even do that. In the paper above, they're referred to as "Theoretical Evidence");
3. Present a hypothesis;
4. Construct a model;
5. And test the model to see if their hypotheses stand up to the data or not.
And at each step they provide varying degrees of explanations of their methodologies, assumptions, controls, potential confounding factors, flaws in their data, and so on.
As you can imagine, it is rather difficult to make solid, generalizable conclusions. For one, the metrics may be pretty poor approximations. Or the model may be poor. Or the hypothesis may be flawed - an unfortunate problem with many studies is that they do not take into account changes in the legal environment (such as the changes wrought by the AIA and decisions like Medimmune v. Genentech) so their very premises are flawed.
Fortunately, many studies are all about finding flaws in other previous studies, so there's some semblance of balance.
Really, go through the paper above and the papers it cites, and you'll see why it's downright impossible to make broad assertions like "patents promote innovation" or "patents harm innovation".
> ... and said "I represent a concerned group of local small business people and programmers, and software patents are destroying our ability to make a living and causing us to lose jobs", that software patents would have been a solved problems years ago.
That is very true, but my very strong hunch is that this hasn't happened because software patents are not really a problem -- at least not anymore. As I've said elsewhere, the vast majority of programmers barely even know what a patent is, and tons more consider them a good thing. Those are also overwhelmingly the people that don't spend time posting on places like this.
Another indirect piece of evidence is the current uproar in Congress about patent trolls: A handful of trolls (that I'm aware of) -- like the WiFi guys, the scan-to-email guys, the vehicle-tracking guys -- send out demand letters to mom & pop businesses, and suddenly Congress is up in arms. It's possible, but I find it unlikely that, if there are so many companies involved in software patent litigation, there are not even a few that would reach out to their representatives.
Glad to finally see someone who looks at the other side of things as well! I generally avoid IPWatchDog for the same reason reasons as Groklaw: too shrill, snarky and narrow-minded.
> Judges apparently think that you just tell the computer what to do and it requires no undue experimentation.
Yes, Fonar v GE is a ridiculous one when generalized, and I do wholeheartedly agree that the "enablement" requirement is very weak currently and a ton of patents don't meet it in my eyes. Take the PageRank patent, for one, often cited as an example of a good patent. It is very lacking in implementation details, so much so that a blogger set out to implement it and ended up with a bunch of posts complaining about all the undue experimentation he had to do.
But take the average case: how much undue experimentation is required for your average patent? How much undue fiddling would be required to implement, say, Twitter's pull-to-refresh, or the iPhone rubber banding patent, or the Google doodle patent, or the MS FAT patent? I think my point still stands: how many patents couldn't you implement just by reading them? (Also, keep in mind triviality of implementation is very different from non-obviousness.)
> They apparently haven't read the newspaper...
Like I said, tech media is not only clueless, it's deliberately misleading. Newspapers are merely anecdata, and worse, potentially biased anecdata [1]. Why rely on that when we have studies that tell you the data they looked at so you can find the flaws in them rather than accepting unsubstantiated, rageview-bait at face value?
> Also, we'd best not look at the steadily increasing number of patent cases, nor all of the goings on in Marshall, TX.
Actually, studies [2, 3, 4] did look at it and found no real increase. [2, 3] found that the AIA joinder rules, where previously a single case could have multiple defendants now must be individual cases, caused an inflation in lawsuits with no statistically significant increase in the number of litigants involved. [4] actually finds EDT is better than others with respect to plaintiff win-rates.
> Interestingly, widespread reinvention of a patent is not considered empirical proof of obviousness
A few points here, since this turns up often:
1. Reinvention many years later is not proof of obviousness. It simply means an idea has permeated widely enough to become obvious to others. A big reason the way US (and other) patent offices do examination, where they must support a rejection with enough prior art references, is to avoid hindsight bias.
2. There is very little "widespread" re-invention going on; what is happening is one product/project/company re-invents something(s) and that gets widely used. (Think Android, or the iOS SDK.)
3. Simultaneous independent invention could be considered proof of obviousness... Or it could be proof that two or more really smart people were working on the same problem at the same time, which has happened in "patent races". But at least at the PTO it happened so rarely that they changed to first-to-file.
> You're confusing "contains" with "is" here.
No, you're confusing both, the nature of software and the inventions being claimed. The software is what makes a machine do useful things, and those useful things are the inventions that are patentable.
> I've also seen patent lawyers try to do a reductio ad absurdam saying that software is hardware (read some of the Patently O comments…).
What they mean is, it's "equivalent", which is perfectly accurate. Also, please explain how software can exist without hardware. What's in your mind is steps to implement or recreate software, but it cannot exist without hardware outside your mind. Just like any other invention.
> Apparently they don't realize that software is information ...
How is "Software is information" different from "machines are metal"? Doesn't seem to be my logic that's broken.
> By how much of it is actually new.
Sure, but how do you determine that a patent covers something actually new?
> We can actually enumerate all of the programs that can go on a particular computer. It's the set of integers from zero to the largest number that would fit in memory (most of these are not useful, however).
I don't see the point, but I can trivially debunk this:
1) The order in which you "interpret" or execute those bytes gives you a completely different program. So if you can fit in 1GB in a high-end smartphone, the number of all possible programs approaches factorial(1e9). That's a number about 8 billion digits long, if Google serves correctly.
2) Now consider that you fill that memory up with programs that generate other programs, or variations of themselves. Something like, say, Conway's game of life. I cannot begin to calculate how many different programs that could generate.
These are absolutely humongous numbers, no different from the number of ways physical objects can be arranged. Now consider that only an infinitesimally small part of them are actually useful. I find that an argument to the patentability of software than against it.
> ... whereas the computer's properties are not altered by the information on it--the computer is still performing an instruction loop over the data it contains.
The computer's properties are not altered?!? It goes from being a heap of semiconductors to actually doing something! How is that different from a machine going from a heap of components to an useful implement?
And each program makes it do something different so each program changes the purposes of the computer, and so becomes a different tool! A simple thought exercise: you're on Youtube. Someone asks you what you're doing. Is your answer:
a) "I'm using the computer"
b) "I'm watching videos"
Interesting isn't it? Now repeat the exercise with games, programming, document editing. And then explain to me how running different software on the same machine does not change its function.
However TFA is being silly in claiming it threatens any significnt portion of MSFT's licensing revenue. Companies that size don't typically license individual patents, they license portfolios of patents.
Now the following is all speculation, since these licensing deals are very closely guarded, but from the few I've heard of: I'm guessing Microsoft's "smartphone" (or maybe the "linux") portfolio has dozens of patents, each of varying value, of which this was just one. So the portfolio's value will decrease a bit, but I'm guessing not by much.
I'll let alone the fact that your definition of "software patents" is flawed. Let me address your other points.
> There are so many conclusive arguments against software patents that the result is overdetermined.
Or, also known as "being in an echo chamber".
> Black boxes are never patentable.
The concepts of a Person of Ordinary Skill and Doctrine of Equivalents is not "black boxes". Are you saying all the so-called "bad" patents you've seen were things you couldn't implement by reading them?
>1. Computer programmers are overwhelmingly against software patents.
Echo chamber. The vast majority of developers barely know what a patent is. Many others think they're a point of pride, and that's where all these software patents are coming from. The rest are just really vocal.
> For the first time in my life I see educated people advocating the abolition of the entire patent system, even at the cost of life saving drugs, just to get patents out of computer software
"Educated" does not mean "well-informed", and it does not mean they know anything about patents. Economists are still having trouble quantifying the benefits and harms of patents, let alone software patents, so I'm guessing your "educated people" are basing their opinions on personal biases rather than evidence.
> Programmers hate patents in their industry by something like ten or twenty to one.
Wow, this echo chamber seems to have reached resonance. Here's a statistic closer to the truth: 99% of people complaining about patents don't know jack about how patents work and simply regurgitate what they tech media tells them, and tech media is not only clueless, it's deliberately misleading.
> 2. Software patents are harmful to innovation. They created almost all the modern patent trolling crisis. They shutdown startups and innovative projects and block open source.
Studies [1, 2], albeit with only publicly available data, have found no empirical evidence of a "modern patent trolling crisis", and I have heard of more small companies being ripped off by big guys (see i4i) than being shut down by patent trolls. On the other hand, I can list some studies that find "software" patents are no worse than other types of patents [3, 4, 5], and can actually be pretty useful for startups. [6, 7].
> 3. Software is math and both math and mental processes are SCOTUS identified ineligible subject matter.
"Machines are made of metals which occur in nature, which are SCOTUS identified ineligible subject matter."
> 4. The quality of software patents we see is uniformly bad...
I'm genuinely curious: how do you judge the quality of a patent?
> 5. When we bought our computers, ever since the first general purpose computers in the 1940s, we have done so expecting to run programs on them. Using a machine for its expected and customary use is not subject to any patents beyond the patents on the machine itself.
"When we first mined metals, we have done so expecting to make things out of them. Using a metal for its expected and customary use is not subject to patents beyond the patents on the metal itself."
> The current state of the patent system is such that having been granted a patent legally supersedes in power the act of enabling technological advancement. That is to say that the power of the patent system has surpassed its social justification...
I'm not sure what your first sentence actually means, but there is not enough evidence to support the following sentence. I know it's a popular opinion around here, but really, it's because this place is an echo chamber when it comes to certain topics.
There are studies that show how NPEs have beneficial effects (see work by Michael Risch, Jay Kesan, Anne Layne-Farrar etc.), and others showing their harmful effects are not as bad as media makes it look. The GAO report on NPEs, for instance, found that they are no big problem, really.
Unfortunately, the data available is limited in a big way: there is no data at all on demand letters that trolls like Lodsys send. There's an act that would help track these things too (http://www.patentlyo.com/patent/2013/11/patent-reform-2013-d...), and that's the one I'm really looking forward to. That data would give a much clearer picture of the NPE situation.
No, I work as a software engineer, but don't let that get you down from your moral high horse.
But I have also worked for a small company that was ripped off by the big guys, almost went under, and could only fight back with patents. So yes, I think the work patent lawyers do is often useful and important.
Now let me guess: You are annoyed because I pointed out flaws in an article that was confirming your biases.
Hmm, I understand, to varying degrees, pretty much everything that happens there, at least down to the logic gate level. Once it hits the physics or quantum level I'm lost. It might be because I have an EE background, but it's not impossible to be aware of a sizable chunk of the underlying technologies that you use everyday.
What is truly mind-boggling is all the manufacturing processes and logistics that make such technologies a commodity. Maybe this is the point he wanted to make?
> Finally, last but not least, that is why our patent system is broken: technology has done such an amazing job at hiding its complexity that the people regulating and running the patent system are barely even aware of the complexity of what they're regulating and running.
Uhh... the people running the patent system all have technical backgrounds. Patent lawyers and examiners must have a technical degree (edit: at least in the US). That does not mean they are technical experts, and in practice it's far from being so, but the theory is sound. Sure, politicians step down from up high once in a while to shake things up as they're lobbied to, but they are not involved in the actual technical nuances.
>... the patent discussions about modern computing systems end up being about screen sizes and icon ordering,
No, those are only the patent discussions people in this particular bubble hear about. As an experiment, go to patentlyo.com, and browse the archives to see the wide and varied range of patent lawsuits that happen all the time: you don't need to know anything about patent law, the introductory paragraph usually gives a good, brief overview of the technologies involved.
Ahh, Bessen. Interesting that he cites his own "trolls cost $29B" study [1] without mentioning that it's been "called into question" [2, 3, 4] to put it diplomatically. It's not like he's unaware of criticism of his work either, since he has responded to criticism in the past [5].
And par for the course, this article of his has several flaws as well:
1. Giving examiners the power of irreversible rejection is pretty bad, if you think about it. Valid patents are regularly rejected for the very same reasons that people argue invalid patents are allowed. Contrary to popular belief, if examiners are overworked and lack enough time, their default reaction is usually to reject the application. I have seen many a BS rejection issued because the examiner just wanted to meet his quota. And I'm not even a patent agent or lawyer!
2. (Nitpicking) Continuations are not the only way to continue examination after a rejection. You can also file for a Request for Continued Examination (RCE) and continue prosecuting the same application.
3. Continuations are a very valuable tool and not just fodder for abuse. Like TFA says, it allows applicants to claim multiple inventions off a single patent application. But that is often what happens! A single invention could have multiple facets that are inventions in their own right. As a random example, a lightbulb that uses tungsten filament and bulb with an inert gas could actually be three inventions: a) the material being used for the filament, b) the use of an inert gas in the bulb, and c) the combination of both that creates an even longer lasting bulb! What may happen is you file a patent for the combination but may realize each aspect is valuable by itself, and you may later want to claim them separately.
And sometimes, this is not in your control. A patent is allowed to only claim a single invention. An examiner may look at your claims and decide that they cover two (or more) separate inventions, and issue a "restriction", which essentially forces you to choose one invention to continue. If you still want to protect the other claims, you have to file a continuation.
And even further: Companies regularly dump hundreds of pages of technical specs (or a professor dumps a dozen papers) on to a patent lawyer and ask them to file on all the inventions covered in there. Sometimes this happens under a pretty tight deadlines (e.g. the product has already been on sale, or the papers published, for almost a year, which puts a bar on when a patent can be filed). In that case, lawyers write up one gigantic spec, slap on claims for whatever they think is the invention, and file it. Over time, they can then sit the inventors down and hash out the real invention and cover them in continuations.
4. I looked at the file wrapper for the Apple patent. "continuous" was not the only word added. Almost a third of the claim was amended to make it more precise of what they were claiming. I personally don't think this is a stellar patent, but such misinformation must be pointed out.
5. Strangely (or maybe not so strangely) enough, academics who write so much about patents have no idea how examination actually happens, or even how patents even work. To show invalidity, an examiner must show a one or more references that completely cover each and every element of the claim. If they cannot, no matter how obvious it seems to us, they have to allow it. The reason for this is that the decision must be an objective one, and all our opinions are inherently subjective. Supporting a rejection with previously published information, which is a recorded fact, is the only currently known way of doing this objectively.
6. Bessen implies other countries don't have patent litigation problems because their examiners can issue an absolute rejection. First of all, I highly doubt that is the case - there always avenues of appeal. Secondly, he ignores the legal environment in other countries, such as "loser pays" in EU, which tend to discourage lawsuits in general, not just frivolous ones. Thirdly, this makes no difference in the quality of patents - I have seen a ton of US patents and their international counterparts, and they pretty much all have the same claims. And they can all be just as each other. Nokia actually prevailed in some lawsuit in the EU over a patent that claims, without (much) exaggeration, pausing downloads when something more important is to be downloaded.
7. Rambus, as underhanded as their FRAND patent shenanigans were, was not a patent troll. By this standard, all fabless semiconductor companies would be patent trolls.
> For a group of people with highly specialized skills (who get underpaid when compared to lawyers and doctors)...
Too broad a brush stroke there. The vast majority of software development does not require highly specialized skills and is more comparable to nurses and paralegals than doctors or lawyers. (Not trying to be disparaging here, either to nurses/paralegals or developers, just running with your analogy.)
> Where is the risky investment in research that patents are supposed to be making possible?
Focusing on the "risky investment" part, the investment is not just monetary: I spent a full year of my grad research barking up the wrong tree. All I got out of it was a few pages to pad my thesis with.
Just like publications, as useful as it is to future researchers, you can't make a thesis out of "we tried this, it doesn't work", and a huge chunk of research work is exactly that.
Little do they know that even in marriage the dynamics don't change much, only the number of people involved (married people stereotype joke :-))