Justice Ayyangar’s report formed the basis of the 1970 Indian patents act, and without doubt, has been the key driving force behind the stellar success of the Indian generic industry. I would venture that the amendments and discussions that led to the 2005 version of the Act being passed, may be equally responsible for the success of the Indian software industry.
Our patent office has recalled the 2015 Guidelines and issuing a fresh set of guidelines (2016 CRI guidelines) that are in tune with the statutory provisions. We hope that proper implementation of these guidelines would enable our start-ups to innovate freely without worrying about patent litigation, and infringement notices. The guidelines may be accessed here. The Office order issued by the Controller is available here.
The 2016 CRI Guidelines reset the clock to the position prior to 2015 when the 2015 guidelines were issued, with the caveat that the 2016 CRI guidelines are much clearer and provide clear cut procedures and examples that deny patentability to Computer Related Inventions. These guidelines ensure that India’s position is now similar to the stand taken by United States and Europe while determining patentability for computer related inventions. India had the patent law in place correctly from the beginning, and ventures to apply law from other jursidictions was ill conceived.
The guidelines provide a [T]est to determine patentability of CRIs. (Clear procedures to determine patentability have been given)
(1) Properly construe the claim and identify the actual contribution;
(2) If the contribution lies only in mathematical method, business method or algorithm, deny the claim;
(3) If the contribution lies in the field of computer programme, check whether it is claimed in conjunction with a novel hardware and proceed to other steps to determine patentability with respect to the invention.. The computer programme in itself is never patentable. If the contribution lies solely in the computer programme, deny the claim. If the contribution lies in both the computer programme as well as hardware, proceed to other steps of patentability.
This time the CRI examination guidelines explicitly recognize that CRIs may fall under:
3(k) a mathematical or business method or a computer programme per se or algorithms;
(l) a literary, dramatic, musical or artistic work or any other aesthetic creation whatsoever including cinematographic works and television productions;
(m) a mere scheme or rule or method of performing mental act or method of playing game;
(n) a presentation of information;
And accordingly, other portions (other than 3(k)), may be used to determine patent eligibility for CRIs / application (Section 2.2, page 4 of the 2016 guidelines). This is an important step. Use of Section 3(m) and 3(n) may render quite a few applications unpatentable. For example, wireless communications uses several protocols, and as argued by Apple in a recent matter against Ericsson (the matter settled out of court), protocols are abstract, and hence not patent eligible. The US court did not rule on the matter. However, in India it is easy to co-relate Section 3(m) to protocols (a mere scheme).
As another example, in wireless communications, nomenclature is used to define certain channels, for example, logical channels, transport channels, data channels etc. However, this is just a naming convention, and patent applications claiming logical channels etc. may fall under Section 3(m) [scheme] or under Section 3(n) This slideshow could not be started. Try refreshing the page or viewing it in another browser.
I feel these guidelines are more confusing now. What do they mean by “Novel hardware?”. The Patent office has removed the examples of what was considered Patentable in the 2015 CRI guidelines. I am not sure how this will encourage the software Industry in India, it may just be a deterrent to Make in India where entrepreneurs rely on business models based on Innovative Software working on known hardware.
Yes, I agree with Anonymous.
What is a new Hardware?
“computer programme, check whether it is claimed in conjunction with a novel hardware” nothing much beneficial. There is very less innovation happening in hardware. We still stand on the same position. NO CHANGE
It should have been either computer program or hardware, but it states
A lot of innovation is happening in hardware and some of it is happening in India. See some of the work hapoening in non boolean computing and nueromorphic computing happening at IISc and some of the IITs. This is innovation. SW oatents have a variety of problems, the primary problem being that it is simply not feasible to have them. The sheer logistical nightmare involved in managing the combinatorial explosion of algorithm variants should be reason enough to avoid this minefield. Please examine the 8 patents Ericsson is trying to assert and see for yourself how they use the apparatus loophole to bypass 3k ! If there is any innovation in any of those patents, I would like to know ! Unix, linux, ingres, postgres, mysql, apache tomcat, jboss, apache soark, hadoop , x windows, gcc, sparc cpu, riscv cpu – the list goes on – all evolved without sw patents. So the case for innovation does not exist. Patents are not a right, they exist as a tool to encourage innovation. Here the evidence is that it hinders, hence the cri guidelines.
Everyone has a different understanding and that is the good part of debating. Read the Patent act and the definition itself makes it clear what is Patent and what rights it gives to the owner.
I have read the act and also happen to advise the govt. on technology. I have been filing sw patents since the mid 90s and have provided affidavits in patent cases. I have been developing sw and hw products since 1987 and did my first hw startup in 1989. I also advise IP law firms. Part of the problem is that technology in computers has outpaced the act. Hence all this confusion. The wording in the act is a trifle confusing and hence the cri guidelines.
Happens when lawyers encounter technology ! Bottom line is that there us a conclusion that innovation in sw need not be rewarded, the CRI guidelines reflect that.
As I pointed out, this is not an opinion. Examine the Ericsson payents and then we can have a debate on facts.
The whole idea is that software should not get patent protection. Innovation in SW anyway has nothing to do with patents. And no SW Indian product company is asking for it anyway.iSpirit which represents product companies does not want patents. Novel hw means simybtbat, something different from a general purpose Cpu. Also logic is SW, so hw has to be novel at circuit level.
There is a difference. Software codes are subject matter of Copyright but the integration of hardware and software enabling a functionality unique to a particular inventor should be part of Patent. Even in US, where they say abstract idea is not patentable, they never say software related inventions are not patentable. So, I dont see these guidelines in line with US as mentioned in the article. Further, Software company like TCS, WIPRO, Infosys have filed thousands of Patents over the past couple of years and I dont think these Patents have been filed for experimentation with the Patent regime. Ofcourse they wish to protect their IP.
Specifically with respect to the “Novel Hardware” term. Please tell me, if my hardware is novel, why would I bother about claiming a software, wouldnt I focus to try and claim my Hardware itself which can help me run many softwares. If the intention is to exclude software, and there is some other intention in mentioning this term “Novel Hardware” then they should give examples where novel hardware is present with Software, which is missing from the Guidelines. I don’t see any merit in this term “Novel hardware”.
A number of stakeholders gave examples of special hw. This for reasons unknown have not made it in the guidelines. But novel hw is not a variant of a processor. If you are not aware, all processors are designed as sw or rather logical constructs. So the behaviour of hw or its logic is in fact sw. So a new method of multipliers in a cpu say going from a wallace tree to a dadda, is not novel hardware. In simpler terms algorithms are non patentable irrespective of whether they are realized in sw, hw virtual machines, programmable logic or fixed ICs. So an intel x86 clone that mimics behavioue is not in violation. Bug if you use the Intel finfet technique in your process, you will be in violation. But circuit design techniques are patentable. Bottom line, novel hardware is not so ambiguous, but examples like the above would help.
Instead of special hardware please mention about what is the meaning of “novel hardware”. If the hardware itself is novel then why the applicant has to bother about the software. It’s a simple question. If the examples given in the stakeholders meetings have not made it to the guidelines, it only suggests that the Patent office itself isn’t sure what they can grant. That is why thr is more confusion now.
Infosys is publicly anti sw patent. Please see Vishal Sikka’s interview in the Times last year. TCS and Wipro are also signficantly open source based these days and will suffer if sw patents are allowed. TCS has digigov, a product for govt organizations that is based on open source components. They are just less less vocal publicly but are nevetheless not great fans of sw patents either. In fact you will not find support for sw patents in most companies, small or large and this applies even to product companies. The large companies file patents reluctantly to protect themselves against lawsuits abroad. It is a defensive measure. People who actually create products, do not really care for patents.
I beg to differ. I would appreciate if we consider the facts and not give them our own logic. I dnt knw about your concept of Patents but companies who innovate to create products value patents the most. TCS have hired more than 30 Patent attorneys inhouse just to protect their innovation. Please check the no. of applications filed by them. Dont give your own interpretation to the actions of larger companies. I would also suggest that people who understand the concept of Patents should provide inputs so tht thr is clarity as to what the patent office means by “novel hardware” for inventions involving software.
I am not giving my opinion. I happen to advise pretty much all these large companies on technology related issues, including innovation. Cannot say more. So I am stating facts not giving an opinion. I also happen to design processors and operating systems for a living among my other activities so I am one of the few people who are actually concerned about novel hw ! All these companies are in a quandry, I can assure you of that. I am saying this based on direct evidence of innumerable conversations on this issue with all the major companies.
http://articles.economictimes.indiatimes.com/2015-01-16/news/58149713_1_infosys-ceo-vishal-sikka-patents-software-industry
You have not answered the question. If the hardware is novel how section 3k cones into picture?
It is self evident. The locus of innovation has to be the hardware. In factt tbe very term sw is a redherring since by definition any innovation in the software component of a system is ignored. Legally I can see why it is worded that way but technically, just ignore anything that is sw related. So for the examiner, the process is very simple. See if there is novel hw. If there is grant a patent. The innovativeness of the the sw that accompanies the hw is irrelevant since there exists no case where the sw can make normal hw novel. Effectively all mention of sw in the patent application get redacted before examination. It is just white space. 3k is in effect meaningless.
A simple truth table proves my thesis
Novel sw + novel hw – patentable
Non novel sw + novel hw – patentable
Non novel sw + non novel hw – not patentable
Novel sw + non novel,hw – not patentable
Please read this also, http://IP and business.com/Infosys-growing-Patent-portfolio Even I can give numerous examples where the companies have indicated Patent as essential part of their strategy. It’s not about giving arguments. I am very well travelled across the globe. I have interacted with lot of inventors from more than 30 countries and I understand their perspective also. Just saying thr is no innovation in software makes no good argument. For giving an argument from a different domain, consider the fact that Indian Pharma companies copy th research of foreign companies and sell the products at lower price. How will this encourage innovation. I am just trying to understand the Logic behind the terminologies given in the guidelines for CRI.
I am restricting my comments only to sw patents. These are my points
1. No company is filing patents out of a strong belief in patents. As I said,
I am making this statement based on direct evidence not because I am well travelled ! They are filing the payents due to business imperatives.
2. Very few sw ideas qualify as innovative. Take a few of the patents filed and see for yourself. But my point is different. I am stating that innovation in sw should NOT be rewarded. There is no economic case to be made. How else do you explain the CRI ? For example, if you solved the travelling salesman’s problem, it would be worth millions. But you still cannot patent it. It does qualify as great innovation. You will probably get a Turing for it. But no patent ! Why becuase it has been proven that sw innovation happens without major monetary incentive. So there is no societal need to reward it.
All this is fine but What is novel software?? If there is no invention in software then why you need to check anything with respect to software to determine Patentability.
in all the cases mentioned above, where will the sine qua non of inventive step lie ?
1. The guidelines clearly indicate that innovation has to be in HW. So novelty in SW is not really germane. The notion of sine qua non in terms of innovation is absurd to apply to SW since every work is derivative. This is specifically why all guidelines ignore innovation only in SW. It is a morass that we would be well advised not to venture into !
2. The bigger issue is that a lot of the philosophy expressed in patent law and this false dichotomy between inventive and innovation simply does not exist in SW ! As any Comp Sci theoretician will tell you, all SW logic is pure math. There is no distinction between algorithms, SW and methods. That distinction does not exist. Only in implementation can you select how specific HW is used but even there all HW (including CPUs and memory model) are modeled in the abstract anyway.
That is why I ask all my legal friends to just avoid comp sci ! Let us just stick to copyrights and forget about how SW can be innovative. Again I am speaking as a person who is both a patent law expert, who sits in industry patent law panels and somebody who teaches Comp Sci at the PhD levels at IIT. So I am well qualified to talk about both sides of the equation.
I also dabble in the area of habeas corpus rights of artificial life forms and other non-human sentient beings (like chimps, dolphins). So there is the fun problem of AI SW being sentient and hence cannot be someone else’s property !
To sum up, technology creates its own reality. Laws have to be created in cognizance of this reality and not seek inspiration in meaningless precedent. In other words when its comes to SW, patent case law is best ignored.
If that is the case then why the term “per se” is included in 3k. Doesn’t it indicate that some part of the software in conjunction with the hardware make the invention novel and inventive. If not then why keep this term “per se”
Even I would like to make two short observations specifically to the guidelines.
1. If the stakeholders meeting was so successful and everyone was convinced that thr is no innovation involving even part of software then why they have not given the examples like before, which indicates thr is no clarity on this issue. They have Infact used terminologies which have increased the confusion.
2. Reading more publications and meeting eminent people of IT industry makes it clear that they want to encourage innovation and based on tht they want to progress because innovative solutions to problems gives them monopoly in the market and it is not a simple marketing tool. It’s a right which they want to exercise.
My point exactly. References to sw and the per se construct make no sense. In fact it never made sense. It was bad phrasi g copied from legislation in other jurisdictions. But changing the act takes time. As I pointed out, you will never encounter a situation where the sw will affect the decision making process. The act effectively deals only with novel hw. I would love to be proven wrong but I am yet to come across an example that contradicts this position.
There is always a reason behind the terminologies used in the act. Doubting the understanding and intellect of the people who framed the law is not advisable. Specific exclusion of computer program per se indicates that part of computer program does have innovation. I wish Patent office would have given examples to clarify their standing either way. Whether they feel the term is incorrect or they themselves havent understood the concept or any other reason. Because these terminologies like novel hardware only increases the confusion.
I cannot say more without revealing my identity. But the act is badly worded and a confused piece of legislation. As are all sw patent laws for that matter. It is not due to a lack of intellect, technology requires a certain precision that is not a pre-requisite in law. There is a reason why no positive examples were not given, there are not any !
Why do you think the guidelines went through so many iterations ? Try coming up with one example that triggers the per se clause. It is high time the act simply said sw is not patentable and save us all a lot of trouble.
Please dont take things personally. Its a platform to share and understand the situation. As a matter of fact, we don’t know what was the intention of the people while framing the law, and it has not been able to clear the concept until now. Use of the terms like novel hardware is not helping in interpretion when examples have not been provided. The liability has been put on the applicant. It appears much like the Pharma companies, people from software industry who can’t make innovative products want everything to be made available for copying. India has been put under the watch list by US due to such practice. We fear such interpretation will only worsen the scenario and Image of India will be dented.
Phrases like “If the contribution lies in both the computer programme as well as hardware, proceed to other steps of patentability” don’t provide any clarity. If it is all about novel hardware then what they want to check by saying contribution lies in both? If they would have given example of that then it would have helped else it only increases confusion.
On a more detailed comparison, it can be observed that Abeyance Version contained lesser number of negative examples (i.e. illustrative and non-exhaustive examples of non-patentable inventions) and included a few positive examples, which approach would have provided clarity to the reader/practitioner and Examiners. However in the Revised Version, the Indian Patent Office has deleted the positive examples and has provided a long list of negative examples.
The negative examples as contained in the Revised Version include the famous Yahoo Vs. Rediff case which was decided by the IPAB. It appears that a few examples are based on decided case from foreign jurisdiction (for example, the “Pension Benefit Systems Partnership” case as decided by a Technical Board of Appeal of the European Patent Office, T 931/95, seems to have been included). Having said so, no basis can be found for many of the examples as contained in the Revised Version.
It also appears that the Revised Version prescribes tests which would negatively impact the grant rate. For example, the Revised Final version contains a whole section about “Sufficiency of Disclosure” which was earlier not present. It is stated that claims for algorithm are even wider than claims for computer programme and therefore claims for algorithm are excluded from patentability in all forms. The Revised Version places a lot of emphasis on presence of novel hardware. It has been mentioned that means plus function claims shall not be allowed.
Apart from the above, the Revised Version provides a three-step test for determining patentability of CRI, which include:
Properly construe the claim and identify the actual contribution;
If the contribution lies only in mathematical method, business method or algorithm, deny the claims;
If the contribution lies in the field of computer programme, check whether it is claimed in conjunction with a novel hardware and proceed to other steps to determine patentability with respect to the invention. The computer programme in itself is never patentable. If the contribution lies solely in the computer programme, deny the claims. If the contribution lies in both the computer programme as well as hardware, proceed to other steps of patentability.
The first step in the aforesaid test has been clearly derived from steps 1 and 2 of the Aerotel / Macrossan case law from United Kingdom. In case the term “only” is to be read as “solely”, then it can be said that even the second step is based on the step 3 of the aforesaid case law. However, it is surprising to note that step 4 of the Aerotel case law has not been prescribed. Instead, a test solely pertaining to computer programme has been prescribed as step 3, which places emphasis on novel hardware. The third step furthermore states that if the contribution lies in both the computer programme as well as hardware, proceed to other steps of patentability.
By reading the aforesaid three-step test along with other sections of the revised version, it appears as follows:
If a device is claimed, the device has to be fully supported by the specification. Especially if the device is being claimed in “means plus function” form, then such means should have been clearly defined in the specification with the help of physical constructional features. (As per Sections 4.4.1 and 4.4.5 of the Revised Version)
If a device comprising a memory having instructions, in combination with a processor (which is configured to implement a method in accordance with the instructions) is being claimed, then such device will be treated as being NOT novel and hence, not patentable. (As per Sections 3.9, 4.4.1 and 4.4.5 of the Revised Version).
If a computer readable medium having instructions stored thereupon is being claimed, then the same will be treated as being directed towards a computer programme per se., and will be treated as being non-patentable. (As per Section 4.4.4 of the Revised Version).
If a device is being claimed comprising at least one novel hardware, then the same is patentable. If the device being claimed includes known hardware(s) “interconnected in a novel manner”, then the device is patentable. If a device satisfying the aforesaid conditions is being claimed in the “means plus function” form, then such means should have been clearly defined in the specification with the help of physical constructional features. (As per Sections 4.4.5 and 5 of the Revised Version)
A method implemented by a novel device (as mentioned above) is patentable. (As per 5 of the Revised Version and point 3).
A method implemented by a known device / hardware is patentable subject to the contribution lying in terms of BOTH the method and the known hardware. (As per 5 of the Revised Version and point 3). While, it is ambiguous as to the nature of contribution expected in terms of the hardware, we believe that if any one of the following (as taken from the previous manual, which is based on EPO Guidelines) is satisfied, we have reasonable chance for obtaining protection for the method claim:
– Higher speed
– Reduced hard-disk access time
– More economical use of memory
– More efficient data base search strategy
– More effective data compression techniques
– Improved user interface
– Better control of robotic arm
– Improved reception/transmission of a radio signal
On a comparison between the version dated August 21, 2015, and the Revised Version, it can be observed that as Abeyance Version was a little bit favorably structured.
On a more detailed comparison, it can be observed that earlier versions contained lesser number of negative examples (i.e. illustrative and non-exhaustive examples of non-patentable inventions) and included a few positive examples, which approach would have provided clarity to the reader/practitioner and Examiners. However in the Revised Version, the Indian Patent Office has deleted the positive examples and has provided a long list of negative examples.
The negative examples as contained in the Revised Version include the famous Yahoo Vs. Rediff case which was decided by the IPAB. It appears that a few examples are based on decided case from foreign jurisdiction (for example, the “Pension Benefit Systems Partnership” case as decided by a Technical Board of Appeal of the European Patent Office, T 931/95, seems to have been included). Having said so, no basis can be found for many of the examples as contained in the Revised Version.
It also appears that the Revised Version prescribes tests which would negatively impact the grant rate. For example, the Revised Final version contains a whole section about “Sufficiency of Disclosure” which was earlier not present. It is stated that claims for algorithm are even wider than claims for computer programme and therefore claims for algorithm are excluded from patentability in all forms. The Revised Version places a lot of emphasis on presence of novel hardware. It has been mentioned that means plus function claims shall not be allowed.
Apart from the above, the Revised Version provides a three-step test for determining patentability of CRI, which include:
Properly construe the claim and identify the actual contribution;
If the contribution lies only in mathematical method, business method or algorithm, deny the claims;
If the contribution lies in the field of computer programme, check whether it is claimed in conjunction with a novel hardware and proceed to other steps to determine patentability with respect to the invention. The computer programme in itself is never patentable. If the contribution lies solely in the computer programme, deny the claims. If the contribution lies in both the computer programme as well as hardware, proceed to other steps of patentability.
The first step in the aforesaid test has been clearly derived from steps 1 and 2 of the Aerotel / Macrossan case law from United Kingdom. In case the term “only” is to be read as “solely”, then it can be said that even the second step is based on the step 3 of the aforesaid case law. However, it is surprising to note that step 4 of the Aerotel case law has not been prescribed. Instead, a test solely pertaining to computer programme has been prescribed as step 3, which places emphasis on novel hardware. The third step furthermore states that if the contribution lies in both the computer programme as well as hardware, proceed to other steps of patentability.
By reading the aforesaid three-step test along with other sections of the revised version, it appears as follows:
If a device is claimed, the device has to be fully supported by the specification. Especially if the device is being claimed in “means plus function” form, then such means should have been clearly defined in the specification with the help of physical constructional features. (As per Sections 4.4.1 and 4.4.5 of the Revised Version)
If a device comprising a memory having instructions, in combination with a processor (which is configured to implement a method in accordance with the instructions) is being claimed, then such device will be treated as being NOT novel and hence, not patentable. (As per Sections 3.9, 4.4.1 and 4.4.5 of the Revised Version).
If a computer readable medium having instructions stored thereupon is being claimed, then the same will be treated as being directed towards a computer programme per se., and will be treated as being non-patentable. (As per Section 4.4.4 of the Revised Version).
If a device is being claimed comprising at least one novel hardware, then the same is patentable. If the device being claimed includes known hardware(s) “interconnected in a novel manner”, then the device is patentable. If a device satisfying the aforesaid conditions is being claimed in the “means plus function” form, then such means should have been clearly defined in the specification with the help of physical constructional features. (As per Sections 4.4.5 and 5 of the Revised Version).
A method implemented by a novel device (as mentioned above) is patentable. (As per 5 of the Revised Version and point 3).
A method implemented by a known device / hardware is patentable subject to the contribution lying in terms of BOTH the method and the known hardware. (As per 5 of the Revised Version and point 3). While, it is ambiguous as to the nature of contribution expected in terms of the hardware, we believe that if any one of the following (as taken from the previous manual, which is based on EPO Guidelines) is satisfied, we have reasonable chance for obtaining protection for the method claim:
– Higher speed
– Reduced hard-disk access time
– More economical use of memory
– More efficient data base search strategy
– More effective data compression techniques
– Improved user interface
– Better control of robotic arm
– Improved reception/transmission of a radio signal
Actually all your examples are negative cases as per the new guidelines.Tbey are just sw or algorithm. They will not qualify as methods. An MPEG 4 compression algorithm has been specifucally shown as something that cannot be patented. So compression and codec techniques have been shown specifically as negative cases. The key is novel hw. Improving efficiency of sw, which your examples 1,2,3,4,5,6 and 8 demonstrate do not qualify. Example 7 will qualify if you improved the arm’s mechanicals. If you just improved the 3d motion sw it will not qualify. Also see 4.4.1 andd 4.4.4. Sw discused as method claims will also be disallowed ! Examples like yours were discussed extensively I believe.
2 may also qualify if the hd head mechanism is new.
I am surprised the novel hw criteria is causing some much concern. It just means a piece of hw that is sufficiently improved over current hw. I agree tnat this is a tough criteria, so the bar is set very high. So typically new types of digital logic will NOT qualify. You will need to look at circuit level techniques. New sensors will qualify.
Yes, it is a touch confusing because there are no examples from the patent office. May be because there are no examples from existing Patents like you have mentioned but at least some illustrative example like in terms of XYZ would have helped. If the hardware is novel then why we should look for the contribution of the software because the software is not making the hardware novel, it can only relate to the functionality of the hardware based on the software. If someone has to invent a new hardware every time for filing a Patent then there is no sense of keeping the second part of the check as mentioned in the guidelines with respect to checking the contribution of software because according to the stakeholders (as it appears from your explanation), there is no contribution of the software towards an invention. Alternately, does it mean that the hardware is novel but the inventive step is determined based on the contribution of the software?? But in that case also, the software is essential for Patentability and the point of no invention in software may not stand which leads us to the term “per se”. The definition of what the Patent office considers to be Novel hardware would have helped. Because in terms of method claims, novel hardware could also mean novelty in functionality of the hardware i.e a method which makes the operation of the hardware novel in terms of its functionality.
That is precisely what I said. As it stands, as far as I can see sw plays no role in determining the patentability. Create novel hw and you pass the first hurdle.Novel hw is not easy to create. In that respect, these guidelines ensure that only true innovations get patented. In my opinion, novel hw cannot be just variznts on gp cpus. Compared to all other jurisdictions, I think this are the best worded guidelines. It could have had more positive examples, but cannot complain too much about the wording. The intention is very clear. Technology policy in India finally seems to be coming of age. The net neutrality guidelines are another example of where govt guidelines are the best around.
You may be right. But unfortunately due to lack of examples, no specific definition for novel hardware and not giving reason for what they think is th contribution of software for patentability, I agree with others that these guidelines are still confusing. When the prosecution begins, I just hope we get clarity from the Examiners.
Due to lack of examples, the definition of novel hardware and the absence of a reasoning as to why would th Examiner check for contribution of the software when my hardware itself is novel, it appears to be a confused guideline. With clarity, the Patent office could have settled this issue. But we hope that once the prosecution begins, we may get clarity on the same from the Examiners. But unfortunately looking at the bigger picture, it is evident that the earlier guidelines were kept in abeyance because of lobbying from small start ups who want to copy softwares from MNC’s to build thr business. It is reflected from this article: http://thetechportal.in/2016/02/24/no-software-patent-without-novel-hardware-component-indian-patents-office/ Which reasserts the fact that, inventions are being discouraged for profits and politics in our Country, much like in case of the Pharma Industries. But thr is a strong apprehension that all these things will have a long term negative effect on India’s image at a world stage and eventually the objective of business will also not be achieved.
Actually none of the large sucessfull companies n the US belive in patents. Google, facebook, twitter, the list goes on. I am yet so see any great innovative sw patent and I have been examining sw patents since 1994. I refuse to file sw patents for any of my inventions and due to contractual obligations had to lend my name to two. So I can assure you as an inventor in hw and sw, patents have little value except in analog, sensors and silicon techniques. The issue is not copying ideas but the sheer impossibility of figuring out if you are infringing. Please examine a few patents and let us discuss them specifically. Apple has released its programming language swift to opdn source and does not assert patents in this language. The irony is that Apple OS is derived from patent free open source, bsd and machos. So no one has made a sucessful case for the economic rationale for sw patents. The reality is that the world is going open source. And patents have no place in open source. And American companies other than the usual BSA gang actually think highly of us. This is the bigger picture.
I am not so sure on this aspect. Software companies like TCS have been quite vocal on CRI as evident frm thr submissions in CRI guidelines which is available online. If they dnt wnt softwares inventions, they wouldnt have advocated the fact tht inventions in software should not be dependent on hardware. Also, Companies do believe in protecting their IP. Otherwise they won’t be spending so much in litigation. There are numerous examples where companies have benefitted from exercising thr Patent rights thru litigation. I dnt think it is justified to allow others to copy what has been invented by someone else. Innovation and Invention are two different things. I have seen many eminent ppl, even scientists actually use the terms interchangeably without realizing tht there is a underlining difference between the two. Thr r a few articles on tht online. While I agree that not all software related inventions are patentable but I do believe that certain Software inventions (which are process or method driven) can be novel and inventive inventions and hence patentable. Instead of checking for subject matters the emphasis should always be on novelty n inventive step for determining patentability. Many software related invention are discarded because they are not inventive and very obvious, which is perfectly justified. Even certain software inventions tht appear to be quite frivolous should not be granted patents. But for the inventive inventions it is never justified to allow others to copy things and create unfair competition in the market. Let’s see what impact we get because of this from around the world.
Wrt large organizations, you should look at how they act and what they say ! Most of them have relationships with companies favoring patents and cannot afford to take a public anti-patent position. So please ignore the CRI representation. If TCS for example is completely pro patent, why is that for their own products and for cost effective e-gov cost effective solutions they push open source ? The Digi GOv framework from the TCS govt. group is based on open source jboss. You think TCS would like to pay patent royalties to Oracle because of use of Java patents ? Or some Linux related patents asserted by MS ? Licenses like GPL3 are very specifically anti-patent. Infosys has publicly talked about this dichotomy in Mr. Sikka’s interview. Please read it. Till Open source came along, large companies were building services on top of proprietary stacks and hence took a pro-patent position. Now that Open source is prevalent, such positions are not tenable and hence the dichotomy within organizations. So please do not get taken by the superfluous posturing. If you want to see an IT executive squirm, ask him his position on this issue ! Yiu will not get a straight answer. The reply usually goes something like this “You see, it is a very complicated issue. We have to take a nuanced approach considering our various partners and customers” ! I cannot blame them. They do have a duty to their shareholders. And people like you get taken in by this. I advise most of the big companies. In their hearts everyone wants patents to go away. I am one of the leading computer science researchers in my field, fairly well know around the world in my field. Among other things I do research on next generation computers that can mimic and even surpass human brains. One can see that inventions coming out of this work will be fairly innovative since no systems like these exist now. So I am not speaking from a position of someone copying ideas but as someone who is one of the leading innovators around. You still do not seem to understand the basic issue. Which is the fact that sw patents hinder innovation. Society benefits when sw patents are not granted, technology improves further and more investments get made. Innovation actually reduces when sw patents are given. Why do you think Android succeeded ? Why do you think google, facebook and ebay are succeeding ? They file patents but do not assert them. You can violate their patents since they want you to ! They are defensive patents. If you actually write sw for a living, you will see that patents are useless.
The companies choose what they want to Patent and what they want to put it in open source. Like other software companies, TCS could hv easily refrained frm making any submissions. They wer nt forced to make thr observations in CRI, so even if u are correct it’s difficult to believe what u r advocating on their behalf.
A lot of companies were compelled to make submissions even though they did not want to. You need to know how the service industry works to understand this. Have you looked at the patents filed by Indian companies ? But you are missing the bigger picture. The entire big data industry runs on open source. Name one major piece of SW that is proprietary. Even MS uses open source components in its big data experiments. Look at Deity mandate of last year. It is mandatory to use open source unless you can prove open source cannot do the job. Every single govt tender you see will mandate open source barring few exceptions. In this scenario how exactly do you propose to introduce SW patents ?
I understand the service industry. That is because of copyright issues and not patents. How many IT companies submitted their response?? Why the othr big comanies not compelled to give reaponse?? All these are jus assumptions and I dnt think it is advisable to mention things on behalf of other IT companies. Copyright infringement has a criminal liability but Patent has civil. They want to know the source codes so that they don’t copy as it is and become liable under copyright. Which company would want to put themselves liable for criminal proceedings? Innovation as a term is for allowing usage of the software to make modifications in source codes. Thts y softwares are put in open source for allowing innovation. But invention is different. Please understand, technology n law are two different things, ppl who have dealt with both will understand the difference. Thr r softwares tht don’t use the open source. I dnt think MS word and Adobe are open source softwares. I am not sure what u r referring to by mentioning about experiment part of things. If it is, then it may be for source code and then they dnt form part of the proprietry software. Thr is a huge difference between Patents and copyrights. These guidelines just appear to be an attempt to make everything open source and allow easy copying for the companies who cant file for inventions.
While I appreciate the fact that eminent technologists like you are bringing valuable inputs on this platform which ideally the patent office should have done with clarification in thr guidelines, with my exeprience of prosecuting hundreds of applications for more than a decade with all major Patent offices frm around the world, I would like to mention that Open source software and software Patents are two different things. Those companies who build thr software without using the open source models will now be forced to give away thr invention because of this wierd concept of novel hardware. Understanding of open source software and the licensing (legal understanding as to what has been licensed) of these would appreciate that both the open source and the software patents have been co-existing for many years. The licensing tool is to majorly guard against copyright protection of the source codes. There are lot of articles written on it from round the world on this. Readers may see an article from Laura Majeus on this for clarity on what is the purpose of open source and software patents. These guidelines are just a way for smaller companies who cant do research and build their own invention to openly copy the inventive software of other companies. I believe the litigation happening around the world is an example of the strength of patents, the value of software patents. There is a huge difference, what you want to protect as Patent and what you want to put it in open source. As i had mentioned, thr is a difference between innovation and invention, which as I sd not may people understand and get confused very easily. Not everyone would agree with me and similarly not all IT guys agree with the concept of open source. I have met plenty of IT guys from IITs as well who believe thr invention shud get recognition thru patents. Every professional has a different point of view and it may not be appropriate to generalize it as the voice of every IT professional or in my case the opinion of every legal professional. But I really appreciate that you have been trying to help out thru this platform in understanding certain aspects, as I sd, I hope the Patent guidelines wud hv dne that with examples.
I would like you to look at the patent related courses in gpl3 and yoiu would understand the implications. Opoen source and patents are inextricably intertwined, since the newer licenses disalloiw assertion of patents. And since open source is the de facto standard, patents in SW are going the way of the dodo. Check what the world uses for its web servers, blogs, servers and cloud servers. Also most innovation comes from smaller companies not larger companies. And actually mostly from universities, ucb and MIT in particular. Netscape, Mozilla, jboss, Cassandra, spark, open office – all came from smaller companies. IBM has a 3500 member Apache spark team but the SW comes from a small startup. Looks at IBMs own portfolio these days. It’s open source offerings outsells its prop. offerings. I do not know which IT guys you talk to, but the ones I deal with do not seem bothered ! And these are lead architects at the world’s top 10 companies. And yes there is not much internal support for patents even at MS. It is the non techies and legal folks who believe in them. Check with the last 10 Turing award winners, the comp sci equivalent of the Nobel prize. Happen to know a couple of them. Not one will support patents. If you really want to understand this issue, audit commonly used software and see where the innovation is happening. P!ease not do not talk in generalities. Looks at operating systems, databases, data analytics, machine learning, processors (yes processors are SW), cloud servers, storage systems, iot stacks. See where most of the world,’s programmers are spending their time. Your own phone is probably Android !
Likewise, I wish you could understand the difference between innovation and invention. I guess that’s why you need experts in Patents to understand both technology and law. Even I dnt knw wht IT guys you deal with but the ones I have met want to have Patent protection. Someone may be a great genius in research and technology but need not understand what you get in copyright and what you get in patents for softwares. If thr was no distinction, software wouldnt hv got seperate mention in copyright for expression of the idea. Thr is a difference between expression of idea and the idea itself. The source codes relate to the first, while the novel and inventive solution to a problem relates to the other (leading to invention). Need to understand whats the difference in allowing the modification of source codes for innovation and preventing the copying of inventions. You may be presenting one aspect not the other.
I am feeling very nervous with the word “Patent Prosecutor”. Surely, I have a Mr. Hyde in me (whether DR. Jekyll is present or not !) Let us be friendly to all and use the term “Patents Advocate”. “Amo Ergo Sum”
It is educative and instructive to read Professor Madhu’s rich Contributions. His profound academic knowledge is equally matched with a vast professional experience. I am grateful for Comments evoked by all the others as well and for the rich learning opportunity for me with the well informed technical comments
I think people who have been advocating to be from the industry and speaking on their behalf should read the article in TOI today from the most eminent IPR professional of this Country, who happens to be a senior Advocate at supreme court . http://www.pressreader.com/india/the-times-of-india-new-delhi-edition/20160317/282527247547616
The NGO software freedom law center appears to have got it wrong or it appears to be a move to influence the Patent office in taking such steps for allowing easy copying of ideas. The situation is really bad and we hope a representation is made to the Govt for making them understand that you can cut off from the rest of the world and do trade with them. The eminent technologists and IP professionals from around the world are not fools sitting there, they are better equipped and qualified to give guidance on law relating to IPR than anyone else.