Linked by Thom Holwerda on Tue 19th Jul 2011 17:09 UTC
Google For the first time, Google has opened its mouth against the patent trolling by Apple (and by proxy, Microsoft) against Android manufacturers. By way of Eric Schmidt, Google's chairman, the company took stand against the legal actions, and stated they aren't too worried. If need be, Google will ensure HTC doesn't lose the patent case against Apple.
Thread beginning with comment 481574
To read all comments associated with this story, please click here.
How can this be fixed?
by cranfordio on Tue 19th Jul 2011 18:11 UTC
cranfordio
Member since:
2005-11-10

How can this be fixed? I think not allowing for software patents is the wrong answer for many reasons, including just a lazy answer. But how could this be fixed to allow for proper innovation and competition, but also protect a company or individuals time, money and effort into developing new software?

I am no expert, but I think a start (just a start not the ultimate solution) would be for the patent office to have a review committee made up of software engineers that have to approve patents. Also, if there is a dispute, it has to go to arbitration between the companies with the arbitration panel made up of a representative from the patent review committee and other software engineers. Then if the arbitration panel makes a decision and one company chooses not to follow it then a lawsuit is filed and a complaint made to the ITC. I think this arbitration panel should have the right to invalidate patents if they feel thy are to vague, or covered by prior art, and the patent office has to respect that decision.

Reply Score: 1

RE: How can this be fixed?
by organgtool on Tue 19th Jul 2011 19:04 in reply to "How can this be fixed?"
organgtool Member since:
2010-02-25

How can this be fixed? I think not allowing for software patents is the wrong answer for many reasons, including just a lazy answer.

It's not a lazy answer - it's the only correct answer. Patents were created to grant a limited monopoly for an invention. Software design is an abstract concept and does not become an invention until the code has been written, compiled, and tested. But the code itself can't be patented since it falls under the umbrella of copyright. When we allow software to be patented, we allow the idea itself to be patented rather than the invention.

I am no expert, but I think a start (just a start not the ultimate solution) would be for the patent office to have a review committee made up of software engineers that have to approve patents.

That is exactly what is supposed to happen while a patent is under review. However, it is much more lucrative for the patent office to collect the submitter's $1500, rubber stamp the patent application, and let the courts sort out the mess.

Reply Parent Score: 3

RE[2]: How can this be fixed?
by Not2Sure on Tue 19th Jul 2011 19:27 in reply to "RE: How can this be fixed?"
Not2Sure Member since:
2009-12-07

When we allow software to be patented, we allow the idea itself to be patented rather than the invention.


This is not entirely accurate RE: Bilski and is a common misconception spread by people like Thom Holwerda in support of an agenda. Only software specific to a "machine" or that is a transformative process is patentable. It is murky but that is not the fault of the intellectual property theorists or jurists. There have been much reasoned debate and with industry feedback that have generated real proposals to update and reform this area while maintaining protection from "free riders".

That is exactly what is supposed to happen while a patent is under review. However, it is much more lucrative for the patent office to collect the submitter's $1500, rubber stamp the patent application, and let the courts sort out the mess.


You are impugning the motives of extremely overworked and underfunded people. The primary issue (imho) in patent reform is that the fees collected are not utilized by the regulatory agency to fulfill its mandate. The revenue is "diverted" by Acts of Congress to other pet projects leaving the AJP of the USPTO extremely overwhelmed. And yes, some unscrupulous people and corporations have abused this system in the all-encompassing quest in the United States for short-term economic gain, long-term effects be damned. It is corruption at a high level and has several times tried to be corrected by legislation that has been killed.

It is also part of the current patent reform legislation stalled in the Senate and House by the absurd "debt ceiling" crisis that has basically frozen any legislative debate on any topic.

Reply Parent Score: -1

RE[2]: How can this be fixed?
by cranfordio on Tue 19th Jul 2011 20:13 in reply to "RE: How can this be fixed?"
cranfordio Member since:
2005-11-10

I guess this is where I get confused. If software is only covered under copyright than how does it protect the investment made into developing the software? One could easily reverse engineer the software then change the code (variable names, language, interface, etc.) and then just release the software as their own. It seems that it would be hard to prove that they reverse engineered your software and then rewrote it, and since the code would then visually be different it wouldn't violate copyright.

Lets say I developed software that could take the genome of two people and determine any health issues their offspring could potentially encounter. Lets say I spent 100,000 man hours developing this software at a cost of $50/hour. So now I have spent $5 million on this software and believe I could sell 100,000 copies. So I charge $100/copy for the software. Someone else comes along and reverse engineers my software and rewrites some of the code. They spend about 500 man hours doing so at the same cost so they are out only $25,000. They sell the software for $25/copy and 80,000 potential customers buy from them because it is cheaper and I am out $3 million and they have made $1.975 million.

How does copyright protect me? I can't prove they copied the software unless they outright admit it. This is a situation where I feel patents would protect my investment. So I do feel that just saying there should be no software patents is a lazy excuse.

Of course I do feel the system is broken and what I would like to see is a real solution to fixing it, I am sure someone out there has put some real thought into this.

Reply Parent Score: 1

RE: How can this be fixed?
by zimbatm on Tue 19th Jul 2011 19:09 in reply to "How can this be fixed?"
zimbatm Member since:
2005-08-22

Can you cite one software patent that you feel is valid to your eyes ?

Reply Parent Score: 3

RE[2]: How can this be fixed?
by somebody on Tue 19th Jul 2011 19:36 in reply to "RE: How can this be fixed?"
somebody Member since:
2005-07-07

(;sorry, couldn't resist;)

real life trash can? it provides me with space for commercial software after:
1) company goes bankrupt
2) company stops supporting that software
3) i see software works nothing like commercials said

Reply Parent Score: 2

RE[2]: How can this be fixed?
by cranfordio on Tue 19th Jul 2011 20:19 in reply to "RE: How can this be fixed?"
cranfordio Member since:
2005-11-10

No, I can't. But this doesn't mean I don't think there are cases where they can be valid and I am not about to look through every software patent out there to find one. I just don't feel that a blanket "No More Software Patents" is the best solution.

Reply Parent Score: 1

RE: How can this be fixed?
by fran on Tue 19th Jul 2011 19:25 in reply to "How can this be fixed?"
fran Member since:
2010-08-06

It is perceived that it's not in the US's economic interest to abolish software patents.
This might change in the following decades though.

I see this happening as follows.
Recently quantity research projects/papers in other parts of the world in starting to surpass the US.
For example in 2009 the Chinese has produced more IT research papers than the US.

China Leads World on IT Research
http://bigthink.com/ideas/38603

US science chief warns: 'China will eat our lunch'
http://www.independent.co.uk/news/science/us-science-chief-warns-ch...

The tuition fees of American Universities compare to other countries.

Americans worried by soaring tuition fees
http://www.google.com/hostednews/afp/article/ALeqM5iTffhwx80QV-z6Xq...

This is a sad tale but the high cost of producing IP is going to sink patents in the US because just like their sneakers it is becoming become cheaper to manufacture overseas.
It will thus be US interest to abolish IT patents in the 20 years.

Reply Parent Score: 4

RE: How can this be fixed?
by galvanash on Wed 20th Jul 2011 06:16 in reply to "How can this be fixed?"
galvanash Member since:
2006-01-25

How can this be fixed? I think not allowing for software patents is the wrong answer for many reasons, including just a lazy answer.


Patents don't apply to everything for a reason. You cannot patent abstract ideas, natural laws (even if you discovered one), mathematical formulas, etc. You are only supposed to be able to patent "concrete" things - you can for example design a mouse trap and patent it, but you have to provide enough detail to actually build it. A patent on "a mousetrap that transports all mice in a home back to November 5th, 1955" won't fly until someone can actually describe how to build a functional time machine...

The main problems with software method patents are:

1. 99% of them are stupid and obvious. The system actually encourages this. You can find all sorts of patents that, after carefully removing the bits that have prior art, boil down to "present a link to the user to act on". If you look at such patents as what they are, it becomes immediately obvious that there are thousands of patents that are really nothing more than "click here" patents - they only differ in the details leading up to the "click" part... What is innovative about clicking links? Why does combining it with something else make something "new"?

2. Methods ARE abstract ideas. Software method patents are both abstract ideas AND mathematical formulas (software is simply a method for describing an idea to a machine - it is not a machine itself). A gratuitous error was made by the government when they allowed for the concept of method patents in general, but doubly so when they apply to software. It just doesn't make sense to patent this kind of stuff - the VAST majority of it is obvious and even the non-obvious stuff represents very little in the way of real R&D costs (which is what patents are supposed to be protecting).

3. Software is already covered by copyright. Most developers feel that for software to evolve and innovation to take place the ideal environment is one where anything and everything goes - if you want to gain financially from a software idea than do it better than your competition - don't try to lock them out of an idea just to protect your own product. Patents cause harm to the biggest advantage of the software industry - rapid iterative advancements.

4. Software ideas really have low R&D costs. There is very little material costs, you generally don't need to create a factory to productize a software idea. I'm not saying R&D for software is cheap overall, its not, but the R&D that applies to a specific patented ideas is nearly non-existent - It is the distribution and marketing that costs money - implementation costs for the idea in a typical patent is small and immaterial.

5. For software, trade secrets work just fine for the most part. I.e. if you don't want competitors to take advantage of your idea until you are ready then don't disclose it. You are not obligated to do so anyway - a competitor cannot reverse engineer your software unless you give them access to it (at which point you are already at a competitive advantage). And if they do it badly you can nail them over copyright...

6. It puts the emphasis on the wrong part of the equation. Good software is not good ideas - it is good execution of ideas. Patents on software methods are mostly garbage - they don't provide anything meaningful to society (which is at least part of the reasoning behind patents - sharing ideas so that once the patents expire they are available to everyone). Go back in 20 years and look at 99.9% of the software patents filed this year - they will mostly be useless drivel. A machine patent tells you HOW to build a machine - a software patent most definitely does NOT tell you HOW to build said software - it contains little or no valuable knowledge. If patents required showing your code I might feel differently, but they don't...

Reply Parent Score: 6

RE[2]: How can this be fixed?
by Not2Sure on Wed 20th Jul 2011 17:49 in reply to "RE: How can this be fixed?"
Not2Sure Member since:
2009-12-07

1. 99% of them are stupid and obvious. The system actually encourages this.


You are making an exaggerated empirical claim that is verifiably untrue and a false statement.

It is true there are low and non-quality patents and that their rate of grant have dramatically increased in the last decade, unfortunately the rate of review has remained steady or declined over the same time period because of funding issues. However, 99% is not accurate and is hyperbole. As far as the system encouraging low-quality or non-quality patents, that is false. They are not valid patents under the law. That is fact.

2. Methods ARE abstract ideas. Software method patents are both abstract ideas AND mathematical formulas (software is simply a method for describing an idea to a machine - it is not a machine itself). A gratuitous error was made by the government when they allowed for the concept of method patents in general, but doubly so when they apply to software. It just doesn't make sense to patent this kind of stuff - the VAST majority of it is obvious and even the non-obvious stuff represents very little in the way of real R&D costs (which is what patents are supposed to be protecting).


HA! Not just an error but a gratuitous error. That's great. It makes sense to a great number of people, just not you. Software method patents as you describe are not what is described in Bilski which is the most recent guidance provided on the matter. I do agree and I think most people regardless of their views on intellectual property that absurdly broad software patents are harmful.

Again you are arguing that the failure to properly implement the software patent process as codified is somehow reflective of patents themselves which is a fallacious argument.

To make your case you would have to show that the software patent process even when implemented correctly fails in principle to serve its purpose. And this is what propagandists do not do. Instead they start waving around some admittedly grotesque patent abuses and say "SEE SEE! LOOK AT ME! THE SKY IS FALLING" rather than look at the suggested reforms.

The issue is the reform of the patent office not software patents.

3. Software is already covered by copyright. Most developers feel that for software to evolve and innovation to take place the ideal environment is one where anything and everything goes - if you want to gain financially from a software idea than do it better than your competition - don't try to lock them out of an idea just to protect your own product. Patents cause harm to the biggest advantage of the software industry - rapid iterative advancements.


This is why most developers are not entrepeneurs. Alos rapid iterative advancement is the latest fad in sofware methodology. Ten years ago the biggest advantage of the software industry was carefully staged, waterfall planning and requirements engineering. Seriously. And anything that is a subject of "rapid" iterative advancment is most likely not a proper subject of a software patent grant.

In any case, patents per se do not hinder rapid progress in software development except for the the speed of the grant/review process. And in fact it could be argued that a tiered or rapid schedule for software (which has already been proposed) promotes it because as opposed to trade secrets in which the innovation must be kept secret to be protected, the patents require open communication, encouraging innovation.

Again, you're arguing for the need to reform the patent office. Many are in agreement and reforms have been proposed.

5. For software, trade secrets work just fine for the most part. I.e. if you don't want competitors to take advantage of your idea until you are ready then don't disclose it. You are not obligated to do so anyway - a competitor cannot reverse engineer your software unless you give them access to it (at which point you are already at a competitive advantage). And if they do it badly you can nail them over copyright...


So, you are going to create a fabulously inventive method for doing something in the browser in javascript. How exactly are you going to protect that as a trade secret? Trade secrets, patents and copywright all have a role to play in legal strategies for software companies that is true. Trade secrets however foster secrecy (that is their requirement). Patents foster transparency and innovation (that is their requirement).

Different social aims.

Go back in 20 years and look at 99.9% of the software patents filed this year - they will mostly be useless drivel. A machine patent tells you HOW to build a machine - a software patent most definitely does NOT tell you HOW to build said software - it contains little or no valuable knowledge. If patents required showing your code I might feel differently, but they don't...


You should look at the FTC suggestions for patent reform. Suggestions include requiring clearer language in patents to make claims more definite, while also helping others understand the boundaries relating to those patents so that infringement can be avoided. It also recommends that patent examinations be enhanced so that the scope of a patent claim can be fully understood (yet doesnt include how to fund this, which is aggravating). FTC also wants more consideration of how third parties will foresee how a patent might evolve.

In addition FTC is seeking to reform the adjudication process. They wants royalty damages to be capped to a reasonable amount to allow them to be paid. They are also advocating for a method to exclude unreliable expert testimony in setting damage estimates, which could lower costs considerably for alleged infringers. You should see some of the absurd descriptions of market value of infringements that have been accepted on their face by courts during the penalty phase.

Reply Parent Score: 1