The USPTO Would Like to Partner with the Software Community

There are (at least) three valid perspectives on the patent situation within the software community but part of the reason so little constructive progress is made is that many people with strong opinions will often flat-out deny or reject the validity of the other perspectives because it is outside their experience.

Three perspectives of which most people ignore one or more:

- Computer technology is over-run with frivolous, vague, stupid, conflicting, and contradictory patents. Any pretense of quality control by the USPTO was lost in the 1990s. This imposes a non-trivial cost on the entire ecosystem and a complete absence of quality control is arguably worse than no patents at all.

- R&D into new computer algorithms is a non-trivial investment, frequently requiring years and millions of dollars. There is a tendency among programmers to discount the level of effort required to develop a new computer algorithm that materially extends the state-of-the-art even though most could not develop such algorithms themselves and have never been involved in such R&D. Organizations that make this investment do so looking for a return.

- Academia is already facing difficulties in computer science because much of the state-of-the-art research is being done by private companies. Much of this research is being treated as trade secrets because (ironically) patents offer flimsy practical protection. As a consequence, there are a number of areas in computer science where the leading academic papers are literally a good half decade behind the state-of-the-art that is buried in NDAs. Lack of publication means that a lot of smart people are wasting time duplicating work. Patents were originally invented precisely to avoid this outcome. One of the reasons that I stopped reading academic computer science in some areas that interest me is that I see computer science under NDA that is much more sophisticated, which is a shame.

Any practical policy will need to take into consideration all of these perspectives. It is not as convenient and simple as "all software patents are evil!" or "software patents FTW!" but it more closely reflects the real tradeoffs.

I don't think that's an accurate statement of how patents function in these industries.

Patents provide a framework for companies to transact in technology while protecting their R&D investment. It enables companies like MIPS and ARM to specialize in producing designs without having to develop finished products, and it allows companies to invest in open standards like LTE or WiMAX without having some Chinese manufacturer just take the end result and produce cheap chips implementing the standard at low cost because they didn't have to undertake any R&D investment.

And here's another one:

The idea that someone can come up with a solution:

- prevent anyone else henceforth from using it without paying them,

- enforce that by force and using taxpayer money,

- stretching over political boundaries,

- not even giving the remotest fair chance to the unborn to ever come up with that solution,

- charge a price fixed by whatever whims tickle that person's fancy and deny others of a possibly cheaper solution

- just to protect a possibly overvalued idea

.. is extremely horrifying to me.

Please refer this earlier post: http://news.ycombinator.com/item?id=4591253

The fact that it's difficult to reverse-engineer Google's technology is an accident of the kind of products it makes: web services. You can access the product while the "secret sauce" remains safely tucked away in a data center.

At the opposite end of the spectrum think of a company like Intel investing into WiMAX. Developing that standard was not cheap, but almost by definition it was something that required disclosure of the end result. Intel can try to recoup the initial investment by selling chips implementing the standard, but competitors can always undercut them on price because they didn't have to spend any money coming up with the initial design.

There is a lot of technology that's much easier to reverse engineer than it is to develop. Especially with software, where it's often very easy to dig out a copy of the firmware from a competing product and decompile it.

You might think this kind of stuff doesn't actually happen, but in the 1990's American companies hated working with the Chinese precisely for this reason. They'd take an American product and copy it right down to the silk screening on the PCB's, and sell the result for cut-rate prices.

When I was younger and far more innocent I would do things because of the sheer love of doing, learning and discovery. I'd spend hours, no, days, wiring chips together, writing code, testing, soldering, dreaming, inventing. In college I even published a paper on some work I did in robotics when pushed my my prof's to publish. It was great. I'd like to think I "invented" a few things in the process too.

Then I was "born" into the real world: One where companies who do nothing (and some who do something) have patents for stuff I had been doing for ages. You know, stuff like using pulse-width-modulation to control LED intensity (ye'old Color Kinetics), clicking buttons (multiple companies) or putting a spring in series with a microprocessor-controlled motor to control force (MIT, series-elastic actuator).

That's when I realized that what I had been "sold" about engineering, entrepreneurship and research and invention was a huge pile of bullshit. The realization was that government incompetence in perhaps one of the most important human fields of endeavor --invention-- has all but ruined what I loved to do before this epiphany.

Now you had to study Sun Tzu. Now, without your own bullshit patent to act as a shield, you had live in fear of being sued for daring to make a screen area clickable or drag-able, attach a spring to a motor, dim an LED with PWM or the myriad of "obvious to someone skilled in the art" things that are patented. Or you had to hire lawyers to ride shotgun on everything you might want embark on.

The alternative is what most of us actually have to do: We use the tools we know and what we learned to create things and explore ideas. Unless you operate at Google/Apple/Samsung scale you have no choice but to stick your head in the sand and hope that someone doesn't come after you because you app sends an email (or whatever).

I don't know what the solution might be. I have done my share of heavy-duty hardware-intensive R&D spanning years. I get it. Sometimes the difference between something flying and not is found in a seemingly minor detail. The wheel was obvious to the second guy who so it, right?

At the same time, I think it is beyond-obvious to most in tech that the US Patent Office has done a dismal job of filtering that which is "obvious to someone skilled in the art" from true invention. I am not going to draw the demarcation line myself, the subject is too complex and each discipline has it's own boundaries.

That said, if you have ever invested any time in the process of patent search I would be surprised if you did not share my sentiment: the vast majority of patents I have seen should never have been issued.

I fully expect lawyer types to say something akin to: "You have to read the claims carefully. The differences can be subtle but important". I've heard this before. My answer has always been the same: As an engineer I don't have to dive into intricate lawyer-ese to know bullshit when I see it. A spring attached to a motor to control force, you know, "F=kx", is way beyond obvious to anyone who took first-semester Physics. Yet, you lawyers manage to craft convoluted language that takes "F=kx" in the form of a motor and a spring and turns it into a patent that I now have to worry about if I do robotics (which I have done in the past).

Perhaps one of the problems is that the USPTO (I can't speak for other countries) is like a self-feeding fire: The more patents they approve the more people and companies have to file protection patents, which brings in more and more money to support more and more bureaucrats. Every patent you approve is a patent that has "job security" written all over it. If, starting on Monday, they approved patents at a rate of 10% of what they did last year they would probably have to fire 75% of their staff (just guessing).

Maybe the way to fight nonsense patents isn't to make an intellectual claim at all. Maybe the only way is to act politically in order to funnel, say, 50% of USPTO revenue to the Department of Health, or Education or completely outside of government entities. The USPTO would cease to be a self-feeding fire and they might just be forced to only pass real patents.

What if there was a rule that everyone involved in a patent that is invalidated is fired and they loose their pension? Violent, yes, but it would sure raise the bar very, very quickly to a super-high level. Which is exactly where the bar should be.

The other aspect of this is that perhaps patents should cost a lot more money and be subject to a significantly more public process where prior art is defined far more liberally than it is today. If a patent application costs started at the lesser of a million dollars or some percentage of last year's revenue the bullshit patents would go away. In the case of organizations like Google/Apple/Samsung it might cost them ten million dollars to apply for one patent and dozens of millions of dollars to actually get one patent. If a patent costs fifty million dollars I suspect we are not going to see many "slide the button this way" patents filed. Yes, this is off-the-hip and not well thought out. I get it. Take it as more of a random though out of frustration than a coherent idea.

Sometimes you just want to cry when you learn about some of the consequences of the government-sponsored monopolies created through the USPTO. Here's a particularly touching example:

http://www.electronista.com/articles/12/06/14/speech.applica...

The teachers at my kid's school asked me if would collaborate with them to help create (free) iPad apps for their developmentally-challenged students. Of course, I will, but the above bullshit patent and the many more that must exist in the dark files at the USPTO sometimes jar your reality. It is a sad note that rather than sit down and start writing code one of the first things I have to do is a patent search.

Sometimes I wish things were like when I was younger and more innocent: I could choose to help someone with my knowledge of technology and we'd all benefit from what might result (commercial or not). That, sometimes, is a tough choice to make today.