Boy there is a lot to respond to in this one, but here goes...
Perhaps there would never be a single algorithm "discovered" by patenting. However, if you recall that patents protect functionality and that a good (and valid) patent must disclose the functionality in sufficient detail so that a person of ordinary skill in the art can make the invention by reading the patent, then by comparing the source code that an OPL would require to be disclosed with the functional disclosure from the patent, we can learn a great deal about the process of implementing software, which to me is a goal that is not only worthwhile but critical to efforts to ensure that software development becomes/remains a true engineering discipline. In this context, you should be able to see that a software patent would perform as both a requirements spoecification, a design document, and even perhaps a data flow diagram. Taking that information and implementing it is just good software engineering. I won't presume to speak for David Pressman, but I believe he would agree with me.
I hope that adresses a large chunk of your comments. As for people who are in territories that do not recognize software patents, they can get the patents in the United States. A US patent is then prior art that will prevent a patent from issuing in other countries (and vice-versa).
For an explanation about the publication problem, that does not exist in the same way in the US. See my other post on that.
The OPL I am advocating would be an OPEN license. The GPL can be used to prevent people from using code as well as to open it up. You either comply with the terms of the GPL or you can't use the code. Why? IT DOESN'T BELONG TO YOU (unless you wrote the base code and are the GPL licensor). You are using it with PERMISSION. I don't have time to look it up, but I know I have seen at least 1 story here about a lawsuit to require a company to comply with the GPL's terms. The same concept applies to patents. But -- think of the possibility of a closed-source company liking an Open Source idea so much that they put resources into implementing the patented idea. The OPL would require them to then release the source. Think it won't happen? I think it could. Apple is moving into a mixed closed/open source model with Chimera as one of the latest examples of it's cooperation with the Open Source community.
if someone else wants to use an idea from one of my programs, and they go to the effort of reimplementing it, they're welcome to it
This wouldn't change that at all except that someone reimplementing your "idea" a/k/a invention (because at that point it is no longer a bare idea) would have to release the source. Under the GPL if someone "reimplementing" did not use your code base, they are under no obligation to release their code.
Yes, "prior art" will invalidate a patent. However, there are well-defined rules as to what is considered "prior art." The United States does not have the same rule as most other countries. In most of Europe for example, there is somehting called the "absolute novelty" requirement. If you release information or the product before applying for a patent, you cannot obtain a patent.
In the US, we follow the "first to invent" rule meaning that the first person who can show inventorship gets the patent. The inventor also has one year to file a patent application following public disclosure of the invention. So up to a year of implementation can occur before you lose patent rights.
Of course the MOST important thing is that to get a patent you DO NOT need a working model of your invention. You need a finished CONCEPT, meaning that you have to know what the invention does, how it does it, and be able to describe it in sufficient detail so that someone of ordinary skill in the art can implement the invention from reading your patent. That is the reason why I believe the OPL would promote progress in Software Engineering - you could compare the implementation with the process disclosed in the patent and thereby learn a great deal about the process of software development.
I'm actually thinking of going beyond that -- my OPL would require the release of source code that performs the patented FUNCTION even though it is not derived. I am NOT trying to break the back of software patents. I believe that they are useful. But I also believe in Open Source principles. Open Source developers should be able to do their thing with the protection of the law and Closed Source developers should be able to continue their thing as well. If a Closed Source company wants to implement patented functionality from the Open Source world, well, then they will have to open their code.
The GPL is a COPYRIGHT license. Copyright only protects the expression (source code or object code). It does NOT protect the functionality. That's a patent's job.
A good patent cannot be "dodged" as easily as you think. The Doctrine of Equivalents says that you cannot make insubstantial changes and claim you do not infringe the invention. So your "write it with a different function" approach may not work. Sorry "dude" but your conclusion is based on false assumptions.
I'm sorry, but the only thing I can really say is that you appear to have no REAL idea how patents or the law work. Your "criticisms" are nothing but conclusory negative remarks unsupported by facts.
If you had bothered to READ and actually THINK about my post, you would see that what I am advocating with the OPL would provide a GREAT contribution to Computer Science and specifically Software Engineering as a discipline. Show me another potential framework for a large community to examine and critique differing approaches to provide the same functionality.
Calling something "moronic" will not do anything to help you deal with the fact that software patents are here to stay. An OPL is a way to ensure that Open Source software not only survives, but thrives, contributing to the body of knowledge in the area in the process.
If you have specific criticisms about the idea of an OPL, I'd love to hear them. If you want to rant and rave, go on Jerry Springer.
I've said it before, but it seems particularly appropriate in this thread. If Open Source code is novel, then it should be patented. Then a patent license similar to the GPL copyright license can be used to ensure that Open Source users can use the functionality in the code. If an Open Source patent License (OK, I'll coin a term -- the OPL - Open Patent License) is used in conjunction with the GPL, think what a remarkable impact it would have on code development.
All it would take is one killer app license under the OPL to create public demand. Then if anyone wanted to duplicate that functionality, the OPL would allow it -- BUT the corresponding GPL (or the OPL itself if properly worded) would require developers to release source code with their implementation.
I submit that this would have an even GREATER impact than the GPL. Developers would be free to try widely disparate approaches to achieving the patented functionality. The different algorithms and approaches could be compared with the best methods prevailing because the best code would be that actually used. Think of the contribution to computer science possible with widespread comparison of designs. I think the industry-wide effects would result in much higher-quality code in general.
Don't tell me that Open Source cannot get patents. If someone bothered to look, they could find a patent attorney who would be thrilled to get a patent for Open Source code as long as someone paid the filing fees (for small entities, about $350). I am a registered patent attorney and I would be thrilled to prosecute one of these applications. I'm sure I am not alone.
Re:One thing I've NEVER seen here....
on
Fair IP Laws?
·
· Score: 2
there is the silliness of patenting an idea that you yourself haven't really implemented
Define "implemented." To get a patent, someone of ordinary skill in the art must be able to pick up the patent specification and actually MAKE your invention without undue experimentation. If they cannot so that, the patent cannot issue (or is invalid if it does). So if you have great design docs, that is enough -- you don't have to write code. Someone should be able to write code from your design.
I could patent the use of computers to read minds, even though it's never been done
No, you can't. See above. You can't REALLY even believe this statement yourself. If you did, then you should be running off to a patent attorney right this second.
Re:But software IS special
on
Fair IP Laws?
·
· Score: 2
You are both right and wrong. Yes, software is special. But it IS different from other "mathematical notations" - even assuming that characterization is correct (and I don't think it is).
Yes, computer languages have strict syntaxes and grammars like mathematical languages. But we are not talking about Turing Machines here. We are talking about something that has never existed before -- a language that a machine can interpret to perform functions by following instructions (with this definition I am including even such things as punched paper rolls for player pianos as "software").
It is this core difference that makes software protectible by both copyright and patent laws.
Re:One thing I've NEVER seen here....
on
Fair IP Laws?
·
· Score: 2
I submit that there is no public benefit from software patents. (Copyright is beneficial however). I submit that software creation/research is never motivated by patent incentive, or that such cases are vanishingly rare.
Prove it. Show me some empirical data -- not just anecdotes and opinion.
I'll probably get flamed to death for this, but from where I sit it seems that all Stallman ever does is condemn and complain. I can't remember the last time I ever heard of him praising or complimenting. Wait -- never mind the last time -- I have NEVER heard of it.
I get the distinct feeling that even if every piece of software in the world was called GNU/something_or_other he still wouldn't be happy. All he is doing is marginalizing himself by looking like a chronic crybaby.
Re:One thing I've NEVER seen here....
on
Fair IP Laws?
·
· Score: 2
Your argument has a flawed premise -- the GPL says what it does about patents because it has to -- not because they are bad.
The GPL is a license -- permission to use intellectual property of another within specified parameters. It is the MOST fundamental premise of any property -- real, personal, or intellectual -- that you cannot give what you do not have.
Think of property rights as a bundle of sticks. The IP owner controls the bundle. When you license, you give someone else permission to use one or more sticks. Any IP right -- patent or copyright -- provides for the EXCLUSIVE control of the owner.
The GPL is simply acknowledging that someone else has control of the bundle and that the GPL cannot give permission to use someone else's sticks.
Problems with SPECIFIC patents - like one click - are isolated and not problems with the CONCEPT of patenting software. Yes, a patent can be invalid because it is an obvious modification of the prior art. Solution there: increase the quality of people hired to examine applications for patents. But remember -- obviousness has a legal definition in this context and does not mean exactly the same thing as it does in ordinary use.
Re:if defence or prosecution methodes were patente
on
Fair IP Laws?
·
· Score: 2
Do Carpenters need to pay for a license before they put up an A-Frame house?
Yes, they do -- or the home builder pays for the plans. These plans are protected by copyright - sometimes for over 100 years, not the 20 a patent gives.
To do math, do you have to pay extra to use the square-root key?
Again, yes you do (or did) -- advanced calculator functions were an active patent area -- it was built into the price of the calculator itself.
just because someone was the first to file a patent for a new problem should not give them the right to deny others use of it.
Ummm.. yes it does. New solutions to problems have always been patentable in other areas - software should be no different.
Ok -- I preface this with with the statement that I AM a lawyer -- and a patent lawyer at that....
The best reply in this thread is the AC's. On a very superficial level, the doctrines of prosecution history estoppel and the doctrine of equivalents seem to be at odds. However, they really aren't.
PHE prevents an inventor from recapturing subject matter the inventor forfeited in his exchange with the patent examiner. For example, if a claim literally "reads on," i.e., describes, covers, etc. prior art, then it is invalid and will be rejected by the examiner. The inventor has 2 choices at that stage: 1. add more elements or limitations to the claim so that the claim does not completely describe prior art; or 2. Make arguments to the examiner that the claim does not describe prior art because the prior art lacks certain features described by the claim as it is currently worded.
In choice (1), this is an amendment related to patentability (an attempt to get an allowable claim). The Fed. Cir. Festo opinion held that in this case the inventor could not claim that ANY equivalents of the elements of an amended claim would infringe. This is a VERY harsh result, albeit an easy rule to enforce, because then trivial changes could be used to defeat patent protection. Preventing trivial design changes for this purpose was exactly what the US Supreme Court wanted to address in the Warner-Jenkinson case with the doctrine of equivalents.
Option 2, arguing that the claim does not cover certain devices or subjects, creates a prosecution history estoppel. If an inventor argues that his patent claims do not cover certain subjects, he cannot later claim in an infringement suit that the same subject matter he argued to the PTO was not covered is in fact now covered by his claim (the "recapture rule").
The two concepts are related and sometimes overlap, but are distinct. The US Supreme Court reaffirmed that the desire for a bright-line rule of easy application cannot outweigh the policy of the Patent Statute - to give inventors real protections, not ones that can be circumvented by trivial changes. There is a proper way to analyze patent claims - read the claim, read the specification, then read the prosecution history. This concept has been reaffirmed in many many cases. Lawyers filing infringement suits have a duty to make sure that at least one claim in a patent completely describes a device that will be claimed infringes and they can (and should!) be sanctioned for not so doing prior to filing the suit.
This is NOT a "missed opportunity to reign in patent abuse." This IS a reaffirmation of the correct way to interpret the scope of a patent.
The Federal Circuit had held that anytime a patent claim (the part that defines the invention) had been amended for a reason related to patentability, then the inventor could never claim that a device was the equivalent of what he claimed in his patent. The Supreme Court had created the Doctrine of Equivalents to prevent people from making minor changes to devices that did not amount to a real departure from what the patent disclosed and then claiming that the altered device was not literally covered by the patent. In software terms , think of this as claiming that a while loop and a do-while loop are not essentially the same thing. Yes, there are differences (where the check is performed) but the differences are trivial.
The Federal Circuit's rule was overly harsh. It is virtually impossible to get a claim allowed at the PTO without amending it at some stage. Also, virtually the only reason you amend (aside from correcting typos) is for a reason related to patentability. So this had the effect of eliminating a very important part of patent law.
The Supreme Court simply stated that the Federal Circuit departed from the law and should correct itself. To determine what a patent claim covers, you look first at the claim itself. Then you look at how the claim is described in the specification portion of the patent. Next you read the prosecution history -- the exchange of arguments with the patent examiner. It is there that you see how the inventor distinguished his invention from the prior art and further defined the terms. Only then can you properly determine the scope of the claim.
And yes, all of this is public record readily available from the PTO.
Re:One thing I've NEVER seen here....
on
Fair IP Laws?
·
· Score: 2
Why should independent re-invention not be a defense against patent infringement?
Perhaps it should. With business method patents, it is. In fact, it is even broader covering commercial, for profit uses, so long as the person defending was using the method first. See 35 USC sec. 271(b)(1). However, I admit that I am stumped when I try to come up with a scenario where someone has a good defense under this statute that would not also result in the patent being invalidated.
One thing to keep in mind is that the US system is a first to invent system. This policy was enacted to hopefully drive innovation further and elimnate the "race to the patent office." As a consequence, there will always be second (and other) inventors. I can tell you that in 1989 I came up with a great invention -- only to discover that it had been patented in 1971 by someone else.
Re:But software IS special
on
Fair IP Laws?
·
· Score: 2
SOFTWARE PATENT LAWYER QUESTION
Do you think you can forward me to a resource or a ruling which might shed some light on this question?
1. Can I get around a software patent by only publishing the source code?
2. Non-commercially?
3. Will releasing it into the public domain remove liability?
4. Can I host public domain source code which implements a software patent?
I hesitate to answer this because it gets very close to establishing an attorney-client relationship. However, with the disclaimer that this is NOT legal advice and that ANY of these answers may change with the addition of facts not provided and an acknowledgement that you are NOT my client and do NOT become a client by my giving a response. This response is a general informational post ONLY.
Define "get around." Do you infringe? Probably not if you do not make a functional executable. You are however probably liable for being a contributory infringer if others can make an executable from the source.
Define "non-commercially." You mean free, gratis, pro bono? Does not matter except to the extent you have resporces that can satisfy a judgment if you lose an infringement suit.
Releasing what? Your source code or the patent? Obviously if the patent is in the public domain (expired or disclaimed) then there can be no infringement. If you mean the source code, then all you are doing is enlarging the group of potential infringers and increasing your liability for contributory infringement.
Probably not without being a contributory infringer.
You are right -- this is a patent lawyer question and you need to talk to one and give him or her specifics about what you want to do. There is no single source that can answer all your questions. However a very good resource is "Software Patents' by Gregory A. Stobbs, Aspen Publishers.
Re:One thing I've NEVER seen here....
on
Fair IP Laws?
·
· Score: 2
* You aren't supposed to be able to pattent things that are "obvious," which is pretty much the same thing as saying that there are a large number of people who could come up with the same thing if asked.
Sorry, but "obvious" is a term of art that has a specific legal meaning with over 150 years of court interpretation behind it. Without giving a course in patent law, I have to simply say that it does NOT mean that.
1. Algorithms are patentable
No, they aren't.
* Most software patents are re-worded prior art. The recognised technique is to take something well known, rename all the key elements, and then patent it.
I made this challenge earlier -- find some factual support for this assertion. This is pure speculation unsupported by facts.
* Software patents have the opposite effect, encouraging litigation at the expense of new development.
Same comment as above.
Re:One thing I've NEVER seen here....
on
Fair IP Laws?
·
· Score: 2
I disagree completely. Open Source advocates are savvy enough to come up with some VERY nice licensing schemes for copyrighted work. Why not obtain patents and license the patents under the same type of regime?
Imagine this: Some innovative new GNU tool that is actually patentable and covered by a patent. Now M$ wants to copy the functionality and incorporate it into Windows. Oops! Sorry Mr. gates, but you see we have this patent.....
Could M$ afford to litigate that hard? Yes. All the more reason to make sure the patent is written well. I guarantee you can find competent patent litigators to fight the OS case on a contingent fee.
Imagine the headlines though.. "MS infringing Open Source Software Patent." Think they want to take the PR hit? I don't think so...
Re:One thing I've NEVER seen here....
on
Fair IP Laws?
·
· Score: 2
< FlameShield status=UP > First of all, I don't consider RMS to be particularly informed on the topic of patents.... </FlameShield >
Re:One thing I've NEVER seen here....
on
Fair IP Laws?
·
· Score: 2
Actually, I've advocated 5 years for software patents. I have never said they can't be improved, just that they are not inherently bad.
By the way, the term is now 20 years from date of filing. It also takes an average of 3 years to get the first response from the USPTO. 4 years to get an issuance if you can. So that leaves 16 years -- even shorter than before the change. yes, there are things that can give you extensions, but those are for very specific situations.
By the way -- if there is sufficient innovation in 5 years, isn't the patent essentially worthless if it is old technology no one is using? Most high-tech patents are what I call "front-loaded" -- the majority of the value is early in the term to allow first-mover status. After 5 years the value drops significantly.
Re:One thing I've NEVER seen here....
on
Fair IP Laws?
·
· Score: 2
I personally invented and used tabbed palettes in 1986
if this is true, then Macromedia's lawyers would LOVE to talk to you (Seriously! I am not being sarcastic here).
The points you raise are valid, real concerns. However, they are concerns that cut across all areas of development and are not unique to patent law. This is one reason why clients want their patent attorneys to conduct prior art searches for them -- to supplement what the PTO examiner will find during the examination search with the result that a stronger patent is granted in the end.
We obviously live in a huge, complex world and it is impossible to know everything that is out there. Patentees run the risk that their patents will be invalidated by prior art that no one found before. All you can do is your best.
There are VERY strict requirements for obtaining patents and clearing them from prior art. These rules are applied without mercy by the PTO. Despite popular opinion, it is HARD to get a patent.
As for the "savage consequences" I just don't think that wiping an entire portfolio is fair -- even if a single patent was procured by fraud. I am totally in favor of invalidating the patent, which the law already provides for.
the only way to get rid of one is to litigate at a cost of more than what most people make in their lives
Not true. There are procedures whereby third parties can submit prior art to the USPTO for consideration against a patent application or even after the patent has issued. It is easier now that the US has adopted the rule requiring most applications to be published 18 months after filing and before issuance as a patent.
Re:I'll bite
on
Fair IP Laws?
·
· Score: 3, Interesting
You've gotten many good arguments so far. I hope you read them, too.
I do my best to read all the comments people post in reply to mine. I post because I enjoy the discussion and want to hear what people have to say.
Especially the points that software is mathematics and software is not a machine or a process for physical transformation.
The courts have (rightly under the present statute) held that anything under the sun created by man is patentable. I have a hard time believing that software developers (of which I am one), who supposedly are confortable with abstractions and models, cannot understand the basics of intellectual property which is itself an abstraction. This is not "pure mathematics" as you suggest later -- it is a highly specialized process running on a machine that has the attributes and qualities of an algorithm. However, any one program does not occupy the entire field of mathematical processing the way a "pure" algorithm or mathematical theorem does.
Most extant software patents are entirely bogus, due to prior art or obviousness.
I challenge you to find some concrete factual support for this assertion instead of simply offering an opinion.
Not your industry, which you probably call a profession. None of the documents that you have ever produced as an attorney are patentable.
True. However, I am also a Software Engineer and some of the things I have developed ARE indeed patentable.
Why am I not surprised that you are not interested in constructive criticism at all and have already made up your mind?
You could not be further from the truth. I AM indeed interested in others' comments or I would not post on/. . However, my experience has convinced me that software patents are intrinsically useful and beneficial. The fact that I present arguments in favor of my point of view does NOT mean I do not listen to or consider opposing arguments. I just have not been convinced yet that the contrary point of view is correct. If and when I am, you can be sure I will post it on Slashdot.
Re:Software Patents have a repressive effect
on
Fair IP Laws?
·
· Score: 2
now I have to worry about writing some piece of code that someone else owns
Welcome to the same world other industries have lived in for years! The only difference is that because of the nature of software (i.e. - abstract, mathematical, etc.) it was never clear whether software COULD be patented. Courts have now made it clear that anything under the sun created by man is patentable. The problem was that the uninformed saw software as a mathematical algorithm only -- a pure theoretical construct, not as an application of principles or the execution of an algorithmic process.
Implement reasonable copyright protection and that's enough.
Respectfully, NO IT ISN'T!! Copyright and patent laws protect VERY different concepts. Copyright protects the EXPRESSION of a concept while patents protect the FUNCTION of the concept. Software is unique because it can be protected under both copyright and patents.
Take this example -- you write some code that does CoolThingA in Method A. You compile the code and register the object code with the Copyright Office (object code to make it difficult to copy). That means no one else can use your copyrighted code to do CoolThingA. Now someone comes along and writes some code that does CoolThingA by MethodB. What protection do you have if your code was not copied? NONE. ZIP, NADA, NOTHING, ZUPKUS. Why? Because you need a PATENT to keep others from doing CoolThingA. CoolThingA is your invention - not necessarily the way of doing CoolThingA (although a well-written patent should cover both if possible).
Slashdot Mirror
You can send email to me at compulawyer-at-netscape-dot-com and from there I can give you contact information if you would like to speak further.
Perhaps there would never be a single algorithm "discovered" by patenting. However, if you recall that patents protect functionality and that a good (and valid) patent must disclose the functionality in sufficient detail so that a person of ordinary skill in the art can make the invention by reading the patent, then by comparing the source code that an OPL would require to be disclosed with the functional disclosure from the patent, we can learn a great deal about the process of implementing software, which to me is a goal that is not only worthwhile but critical to efforts to ensure that software development becomes/remains a true engineering discipline. In this context, you should be able to see that a software patent would perform as both a requirements spoecification, a design document, and even perhaps a data flow diagram. Taking that information and implementing it is just good software engineering. I won't presume to speak for David Pressman, but I believe he would agree with me.
I hope that adresses a large chunk of your comments. As for people who are in territories that do not recognize software patents, they can get the patents in the United States. A US patent is then prior art that will prevent a patent from issuing in other countries (and vice-versa).
For an explanation about the publication problem, that does not exist in the same way in the US. See my other post on that.
The OPL I am advocating would be an OPEN license. The GPL can be used to prevent people from using code as well as to open it up. You either comply with the terms of the GPL or you can't use the code. Why? IT DOESN'T BELONG TO YOU (unless you wrote the base code and are the GPL licensor). You are using it with PERMISSION. I don't have time to look it up, but I know I have seen at least 1 story here about a lawsuit to require a company to comply with the GPL's terms. The same concept applies to patents. But -- think of the possibility of a closed-source company liking an Open Source idea so much that they put resources into implementing the patented idea. The OPL would require them to then release the source. Think it won't happen? I think it could. Apple is moving into a mixed closed/open source model with Chimera as one of the latest examples of it's cooperation with the Open Source community.
if someone else wants to use an idea from one of my programs, and they go to the effort of reimplementing it, they're welcome to it
This wouldn't change that at all except that someone reimplementing your "idea" a/k/a invention (because at that point it is no longer a bare idea) would have to release the source. Under the GPL if someone "reimplementing" did not use your code base, they are under no obligation to release their code.
In the US, we follow the "first to invent" rule meaning that the first person who can show inventorship gets the patent. The inventor also has one year to file a patent application following public disclosure of the invention. So up to a year of implementation can occur before you lose patent rights.
Of course the MOST important thing is that to get a patent you DO NOT need a working model of your invention. You need a finished CONCEPT, meaning that you have to know what the invention does, how it does it, and be able to describe it in sufficient detail so that someone of ordinary skill in the art can implement the invention from reading your patent. That is the reason why I believe the OPL would promote progress in Software Engineering - you could compare the implementation with the process disclosed in the patent and thereby learn a great deal about the process of software development.
I'm actually thinking of going beyond that -- my OPL would require the release of source code that performs the patented FUNCTION even though it is not derived. I am NOT trying to break the back of software patents. I believe that they are useful. But I also believe in Open Source principles. Open Source developers should be able to do their thing with the protection of the law and Closed Source developers should be able to continue their thing as well. If a Closed Source company wants to implement patented functionality from the Open Source world, well, then they will have to open their code.
A good patent cannot be "dodged" as easily as you think. The Doctrine of Equivalents says that you cannot make insubstantial changes and claim you do not infringe the invention. So your "write it with a different function" approach may not work. Sorry "dude" but your conclusion is based on false assumptions.
If you had bothered to READ and actually THINK about my post, you would see that what I am advocating with the OPL would provide a GREAT contribution to Computer Science and specifically Software Engineering as a discipline. Show me another potential framework for a large community to examine and critique differing approaches to provide the same functionality.
Calling something "moronic" will not do anything to help you deal with the fact that software patents are here to stay. An OPL is a way to ensure that Open Source software not only survives, but thrives, contributing to the body of knowledge in the area in the process.
If you have specific criticisms about the idea of an OPL, I'd love to hear them. If you want to rant and rave, go on Jerry Springer.
All it would take is one killer app license under the OPL to create public demand. Then if anyone wanted to duplicate that functionality, the OPL would allow it -- BUT the corresponding GPL (or the OPL itself if properly worded) would require developers to release source code with their implementation.
I submit that this would have an even GREATER impact than the GPL. Developers would be free to try widely disparate approaches to achieving the patented functionality. The different algorithms and approaches could be compared with the best methods prevailing because the best code would be that actually used. Think of the contribution to computer science possible with widespread comparison of designs. I think the industry-wide effects would result in much higher-quality code in general.
Don't tell me that Open Source cannot get patents. If someone bothered to look, they could find a patent attorney who would be thrilled to get a patent for Open Source code as long as someone paid the filing fees (for small entities, about $350). I am a registered patent attorney and I would be thrilled to prosecute one of these applications. I'm sure I am not alone.
Define "implemented." To get a patent, someone of ordinary skill in the art must be able to pick up the patent specification and actually MAKE your invention without undue experimentation. If they cannot so that, the patent cannot issue (or is invalid if it does). So if you have great design docs, that is enough -- you don't have to write code. Someone should be able to write code from your design.
I could patent the use of computers to read minds, even though it's never been done
No, you can't. See above. You can't REALLY even believe this statement yourself. If you did, then you should be running off to a patent attorney right this second.
Yes, computer languages have strict syntaxes and grammars like mathematical languages. But we are not talking about Turing Machines here. We are talking about something that has never existed before -- a language that a machine can interpret to perform functions by following instructions (with this definition I am including even such things as punched paper rolls for player pianos as "software").
It is this core difference that makes software protectible by both copyright and patent laws.
Prove it. Show me some empirical data -- not just anecdotes and opinion.
I get the distinct feeling that even if every piece of software in the world was called GNU/something_or_other he still wouldn't be happy. All he is doing is marginalizing himself by looking like a chronic crybaby.
The GPL is a license -- permission to use intellectual property of another within specified parameters. It is the MOST fundamental premise of any property -- real, personal, or intellectual -- that you cannot give what you do not have.
Think of property rights as a bundle of sticks. The IP owner controls the bundle. When you license, you give someone else permission to use one or more sticks. Any IP right -- patent or copyright -- provides for the EXCLUSIVE control of the owner.
The GPL is simply acknowledging that someone else has control of the bundle and that the GPL cannot give permission to use someone else's sticks.
Problems with SPECIFIC patents - like one click - are isolated and not problems with the CONCEPT of patenting software. Yes, a patent can be invalid because it is an obvious modification of the prior art. Solution there: increase the quality of people hired to examine applications for patents. But remember -- obviousness has a legal definition in this context and does not mean exactly the same thing as it does in ordinary use.
Yes, they do -- or the home builder pays for the plans. These plans are protected by copyright - sometimes for over 100 years, not the 20 a patent gives.
To do math, do you have to pay extra to use the square-root key?
Again, yes you do (or did) -- advanced calculator functions were an active patent area -- it was built into the price of the calculator itself.
just because someone was the first to file a patent for a new problem should not give them the right to deny others use of it.
Ummm.. yes it does. New solutions to problems have always been patentable in other areas - software should be no different.
The best reply in this thread is the AC's. On a very superficial level, the doctrines of prosecution history estoppel and the doctrine of equivalents seem to be at odds. However, they really aren't.
PHE prevents an inventor from recapturing subject matter the inventor forfeited in his exchange with the patent examiner. For example, if a claim literally "reads on," i.e., describes, covers, etc. prior art, then it is invalid and will be rejected by the examiner. The inventor has 2 choices at that stage: 1. add more elements or limitations to the claim so that the claim does not completely describe prior art; or 2. Make arguments to the examiner that the claim does not describe prior art because the prior art lacks certain features described by the claim as it is currently worded.
In choice (1), this is an amendment related to patentability (an attempt to get an allowable claim). The Fed. Cir. Festo opinion held that in this case the inventor could not claim that ANY equivalents of the elements of an amended claim would infringe. This is a VERY harsh result, albeit an easy rule to enforce, because then trivial changes could be used to defeat patent protection. Preventing trivial design changes for this purpose was exactly what the US Supreme Court wanted to address in the Warner-Jenkinson case with the doctrine of equivalents.
Option 2, arguing that the claim does not cover certain devices or subjects, creates a prosecution history estoppel. If an inventor argues that his patent claims do not cover certain subjects, he cannot later claim in an infringement suit that the same subject matter he argued to the PTO was not covered is in fact now covered by his claim (the "recapture rule").
The two concepts are related and sometimes overlap, but are distinct. The US Supreme Court reaffirmed that the desire for a bright-line rule of easy application cannot outweigh the policy of the Patent Statute - to give inventors real protections, not ones that can be circumvented by trivial changes. There is a proper way to analyze patent claims - read the claim, read the specification, then read the prosecution history. This concept has been reaffirmed in many many cases. Lawyers filing infringement suits have a duty to make sure that at least one claim in a patent completely describes a device that will be claimed infringes and they can (and should!) be sanctioned for not so doing prior to filing the suit.
The Federal Circuit had held that anytime a patent claim (the part that defines the invention) had been amended for a reason related to patentability, then the inventor could never claim that a device was the equivalent of what he claimed in his patent. The Supreme Court had created the Doctrine of Equivalents to prevent people from making minor changes to devices that did not amount to a real departure from what the patent disclosed and then claiming that the altered device was not literally covered by the patent. In software terms , think of this as claiming that a while loop and a do-while loop are not essentially the same thing. Yes, there are differences (where the check is performed) but the differences are trivial.
The Federal Circuit's rule was overly harsh. It is virtually impossible to get a claim allowed at the PTO without amending it at some stage. Also, virtually the only reason you amend (aside from correcting typos) is for a reason related to patentability. So this had the effect of eliminating a very important part of patent law.
The Supreme Court simply stated that the Federal Circuit departed from the law and should correct itself. To determine what a patent claim covers, you look first at the claim itself. Then you look at how the claim is described in the specification portion of the patent. Next you read the prosecution history -- the exchange of arguments with the patent examiner. It is there that you see how the inventor distinguished his invention from the prior art and further defined the terms. Only then can you properly determine the scope of the claim.
And yes, all of this is public record readily available from the PTO.
Perhaps it should. With business method patents, it is. In fact, it is even broader covering commercial, for profit uses, so long as the person defending was using the method first. See 35 USC sec. 271(b)(1). However, I admit that I am stumped when I try to come up with a scenario where someone has a good defense under this statute that would not also result in the patent being invalidated.
One thing to keep in mind is that the US system is a first to invent system. This policy was enacted to hopefully drive innovation further and elimnate the "race to the patent office." As a consequence, there will always be second (and other) inventors. I can tell you that in 1989 I came up with a great invention -- only to discover that it had been patented in 1971 by someone else.
I hesitate to answer this because it gets very close to establishing an attorney-client relationship. However, with the disclaimer that this is NOT legal advice and that ANY of these answers may change with the addition of facts not provided and an acknowledgement that you are NOT my client and do NOT become a client by my giving a response. This response is a general informational post ONLY.
You are right -- this is a patent lawyer question and you need to talk to one and give him or her specifics about what you want to do. There is no single source that can answer all your questions. However a very good resource is "Software Patents' by Gregory A. Stobbs, Aspen Publishers.
Sorry, but "obvious" is a term of art that has a specific legal meaning with over 150 years of court interpretation behind it. Without giving a course in patent law, I have to simply say that it does NOT mean that.
1. Algorithms are patentable
No, they aren't.
* Most software patents are re-worded prior art. The recognised technique is to take something well known, rename all the key elements, and then patent it.
I made this challenge earlier -- find some factual support for this assertion. This is pure speculation unsupported by facts.
* Software patents have the opposite effect, encouraging litigation at the expense of new development.
Same comment as above.
Imagine this: Some innovative new GNU tool that is actually patentable and covered by a patent. Now M$ wants to copy the functionality and incorporate it into Windows. Oops! Sorry Mr. gates, but you see we have this patent.....
Could M$ afford to litigate that hard? Yes. All the more reason to make sure the patent is written well. I guarantee you can find competent patent litigators to fight the OS case on a contingent fee.
Imagine the headlines though .. "MS infringing Open Source Software Patent." Think they want to take the PR hit? I don't think so...
< FlameShield status=UP > First of all, I don't consider RMS to be particularly informed on the topic of patents.... </FlameShield >
By the way, the term is now 20 years from date of filing. It also takes an average of 3 years to get the first response from the USPTO. 4 years to get an issuance if you can. So that leaves 16 years -- even shorter than before the change. yes, there are things that can give you extensions, but those are for very specific situations.
By the way -- if there is sufficient innovation in 5 years, isn't the patent essentially worthless if it is old technology no one is using? Most high-tech patents are what I call "front-loaded" -- the majority of the value is early in the term to allow first-mover status. After 5 years the value drops significantly.
if this is true, then Macromedia's lawyers would LOVE to talk to you (Seriously! I am not being sarcastic here).
The points you raise are valid, real concerns. However, they are concerns that cut across all areas of development and are not unique to patent law. This is one reason why clients want their patent attorneys to conduct prior art searches for them -- to supplement what the PTO examiner will find during the examination search with the result that a stronger patent is granted in the end.
We obviously live in a huge, complex world and it is impossible to know everything that is out there. Patentees run the risk that their patents will be invalidated by prior art that no one found before. All you can do is your best.
There are VERY strict requirements for obtaining patents and clearing them from prior art. These rules are applied without mercy by the PTO. Despite popular opinion, it is HARD to get a patent.
As for the "savage consequences" I just don't think that wiping an entire portfolio is fair -- even if a single patent was procured by fraud. I am totally in favor of invalidating the patent, which the law already provides for.
Not true. There are procedures whereby third parties can submit prior art to the USPTO for consideration against a patent application or even after the patent has issued. It is easier now that the US has adopted the rule requiring most applications to be published 18 months after filing and before issuance as a patent.
I do my best to read all the comments people post in reply to mine. I post because I enjoy the discussion and want to hear what people have to say.
Especially the points that software is mathematics and software is not a machine or a process for physical transformation.
The courts have (rightly under the present statute) held that anything under the sun created by man is patentable. I have a hard time believing that software developers (of which I am one), who supposedly are confortable with abstractions and models, cannot understand the basics of intellectual property which is itself an abstraction. This is not "pure mathematics" as you suggest later -- it is a highly specialized process running on a machine that has the attributes and qualities of an algorithm. However, any one program does not occupy the entire field of mathematical processing the way a "pure" algorithm or mathematical theorem does.
Most extant software patents are entirely bogus, due to prior art or obviousness.
I challenge you to find some concrete factual support for this assertion instead of simply offering an opinion.
Not your industry, which you probably call a profession. None of the documents that you have ever produced as an attorney are patentable.
True. However, I am also a Software Engineer and some of the things I have developed ARE indeed patentable.
Why am I not surprised that you are not interested in constructive criticism at all and have already made up your mind?
You could not be further from the truth. I AM indeed interested in others' comments or I would not post on /. . However, my experience has convinced me that software patents are intrinsically useful and beneficial. The fact that I present arguments in favor of my point of view does NOT mean I do not listen to or consider opposing arguments. I just have not been convinced yet that the contrary point of view is correct. If and when I am, you can be sure I will post it on Slashdot.
Welcome to the same world other industries have lived in for years! The only difference is that because of the nature of software (i.e. - abstract, mathematical, etc.) it was never clear whether software COULD be patented. Courts have now made it clear that anything under the sun created by man is patentable. The problem was that the uninformed saw software as a mathematical algorithm only -- a pure theoretical construct, not as an application of principles or the execution of an algorithmic process.
Implement reasonable copyright protection and that's enough.
Respectfully, NO IT ISN'T!! Copyright and patent laws protect VERY different concepts. Copyright protects the EXPRESSION of a concept while patents protect the FUNCTION of the concept. Software is unique because it can be protected under both copyright and patents.
Take this example -- you write some code that does CoolThingA in Method A. You compile the code and register the object code with the Copyright Office (object code to make it difficult to copy). That means no one else can use your copyrighted code to do CoolThingA. Now someone comes along and writes some code that does CoolThingA by MethodB. What protection do you have if your code was not copied? NONE. ZIP, NADA, NOTHING, ZUPKUS. Why? Because you need a PATENT to keep others from doing CoolThingA. CoolThingA is your invention - not necessarily the way of doing CoolThingA (although a well-written patent should cover both if possible).