Centre for Internet and Society

Arguments Against Software Patents in India

pranesh — 2012-03-13T10:43:12Z

CIS believes that software patents are harmful for the software industry and for consumers. In this post, Pranesh Prakash looks at the philosophical, legal and practical reasons for holding such a position in India. This is a slightly modified version of a presentation made by Pranesh Prakash at the iTechLaw conference in Bangalore on February 5, 2010, as part of a panel discussing software patents in India, the United States, and the European Union.

This blog post is based on a presentation made at the iTechLaw conference held on February 5, 2010. The audience consisted of lawyers from various corporations and corporate law firms. As is their wont, most lawyers when dealing with software patents get straight to an analysis of law governing the patenting of computer programmes in India and elsewhere, and seeing whether any loopholes exist and can be exploited to patent software. It was refreshing to see at least some lawyers actually going into questions of the need for patents to cover computer programs. In my presentation, I made a multi-pronged case against software patents: (1) philosophical justification against software patents based on the nature of software; (2) legal case against software patents; (3) practical reasons against software patents.

Preamble

Through these arguments, it is sought to be shown that patentability of software is not some arcane, technical question of law, but is a real issue that affect the continued production of new software and the everyday life of the coder/hacker/software programmer/engineer as well as consumers of software (which is, I may remind you, everywhere from your pacemaker to your phone). A preamble to the arguments would note that the main question to ask is: why should we allow for patenting of software? Answering this question will lead us to ask: who benefits from patenting of software. The conclusion that I come to is that patenting of software helps three categories of people: (1) those large software corporations that already have a large number of software patents; (2) those corporations that do not create software, but only trade in patents / sue on the basis of patents ("patent trolls"); (3) patent lawyers. How they don't help small and medium enterprises nor society at large (since they deter, rather than further invention) will be borne out by the rest of these arguments, especially the section on practical reasons against software patents.

What are Patents?

Patents are a twenty-year monopoly granted by the State on any invention. An invention has to have at least four characteristics: (0) patentable subject matter; (1) novelty (it has to be new); (2) inventive step / non-obviousness (even if new, it should not be obvious); (3) application to industry. A monopoly over that invention, thus means that if person X has invented something, then I may not use the core parts of that invention ("the essential claims") in my own invention. This prohibition applies even if I have come upon my invention without having known about X's invention. (Thus, independent creation is not a defence to patent infringement. This distinguishes it, for instance, from copyright law in which two people who created the same work independently of each other can both assert copyright.) Patents cover non-abstract ideas/functionality while copyright covers specific expressions of ideas. To clarify: imagine I make a drawing of a particular machine and describe the procedure of making it. Under patent law, no one else can make that particular machine, while under copyright law, no one can copy that drawing.

Philosophical Justification Against Software Patents

Even without going into the case against patents per se (lack of independent creation as a defence; lack of 'harm' as a criterion leading to internalization of all positive externalities; lack of effective disclosure and publication; etc.), which has been done much more ably by others like Bessen & Meurer (especially in their book Patent Failure) and Boldrin & Levine (in their book Against Intellectual Monopoly, the full text of which is available online).

But there is one essentially philosophical argument against software as subject matter of a patent. Software/computer programs ("instructions for a computer"), as any software engineer would tell you, are merely algorithms ("an effective method for solving a problem using a finite sequence of instructions") that are meant to be understood by a computer or a human who knows how to read that code.

Algorithms are not patentable subject matter, as they are mere expressions of abstract ideas, and not inventions in themselves. Computer programs, similarly, are abstract ideas. They only stop being abstract ideas when embodied in a machine or a process in which it is the machine/process that is the essential claim and not the software. That machine or process being patented would not grant protection to the software itself, but to the whole machine or process. Thus the abstract part of that machine/process (i.e., the computer program) could be used in any other machine/process, as it it is not the subject matter of the patent. Importantly, just because software is required to operate some machine would then not mean that the machine itself is not patentable, just that the software cannot be patented in guise of patenting a machine.

Legal Case Against Software Patents

In India, section 3(k) of the Patent Act reads:

(3) The following are not inventions within the meaning of this Act: (k) a mathematical or business method or computer programme (sic) per se or algorithms.

As one can see, computer programs are place in the same category as "mathematical methods", "algorithms", and "business methods", hence giving legal validity to the idea propounded in the previous section that computer programs are a kind of algorithms (just as algorithms are a kind of mathematical method).

Be that as it may, the best legal minds in India have had to work hard at understanding what exactly "computer programme per se" means. They have cited U.S. case law, U.K. case law, E.U. precedents, and sought to arrive at an understanding of how per se should be understood. While understanding what per se means might be a difficult job, it is much easier to see what it does not mean. For that, we can look at the 2004 Patent Ordinance that Parliament rejected in 2005. In that ordinance, sections 3(k) and (ka) read as follows:

(3) The following are not inventions within the meaning of this Act: (k) a computer programme per se other than its technical application to industry or a combination with hardware; (ka) a mathematical method or a business method or algorithms.

Thus, it is clear that the interpretation that "computer programme per se" excludes "a computer programme that has technical application to industry" and "a computer programme in combination with hardware" is wrong. By rejecting the 2004 Ordinance wording, Parliament has clearly shown that "technical application to industry" and "combination with hardware" do not make a computer programme patentable subject matter.

Indeed, what exactly is "technical application to industry"? "Technical" has various definitions, and a perusal through those definitions would show that barely any computer program can be said not to relate to a technique, not involve "specialized knowledge of applied arts and sciences" (it is code, after all; not everyone can write good algorithms), or not relate to "a practical subject that is organized according to scientific principles" or is "technological". Similarly, all software is, by definition, meant to be used in combination with hardware. Thus, it being used in combination with hardware must not, as argued above, give rise to patentability of otherwise unpatentable subject matter category.

In 2008, the Patent Office published a new 'Draft Manual Of Patent Practice And Procedure' in which it sought to allow patenting of certain method claims for software inventions (while earlier the Patent Office objected to method claims, allowing only device claims with hardware components). This Draft Manual was withdrawn from circulation, with Shri N.N. Prasad (then Joint Secretary of DIPP, the department administering the Patent Office) noting that the parts of the Manual on sections 3(d) and 3(k) had generated a lot of controversy, and were ultra vires the scope of the Manual (which could not override the Patent Act). He promised that those parts would be dropped and the Manual would be re-written. A revised draft of the Manual has not yet been released. Thus the interpretation provided in the Draft Manual (which was based heavily on the interpretation of the U.K. courts) cannot not be relied upon as a basis for arguments in favour of the patentability of software in India.

In October 2008, CIS helped organize a National Public Meeting on Software Patents in which Indian academics, industry, scientists, and FOSS enthusiasts all came to the conclusion that software patents are harmful for both the industry as well as consumers.

Practical Reasons Against Software Patents

This is going to be an attempt at distilling and simplifying some of the main practical arguments against patenting of software.

There are traditionally four incentives that the patent system caters to: (1) incentive to invent; (2) incentive to disclose; (3) incentive to commercialize; and (4) incentive to invent substitutes. Apart from the last, patenting of software does not really aid any of them.

Patent Landmines / Submarine Patents / Patent Gridlocks / No Exception for Independent Creation

Given that computer programs are algorithms, having monopolies over such abstract ideas is detrimental to innovation. Just the metaphors say a lot about software patents: landmines (they cannot be seen/predicted); submarines (they surface out of the blue); gridlocks (because there are so many software patents around the same area of computing, they prevent further innovation in that area, since no program can be written without violating one patent or the other).

Imagine the madness that would have ensued had patents been granted when computer programming was in its infancy. Imagine different methods of sorting (quick sort, bubble sort) that are part of Computer Science 101 had been patented. While those particular instances aren't, similar algorithms, such as data compression algorithms (including the infamous LZW compression method), have been granted patents. Most importantly, even if one codes certain functionality into software independently of the patent holder, that is still violative of the patent. Computer programs being granted patents makes it extremely difficult to create other computer programs that are based on the same abstract ideas. Thus incentives # (1) and (3) are not fulfilled, and indeed, they are harmed. There is no incentive to invent, as one would always be violating one patent or the other. Given that, there is no incentive to commercialize what one has invented, because of fear of patent infringement suits.

An apt illustration of this is the current difficulty of choosing a royalty-free video format for HTML 5, as it shows, in practical terms, how difficult it is to create a video format without violating one patent or the other. While the PNG image format was created to side-step the patent over the LZW compression method used in the GIF image format, bringing Ogg Theora or Dirac (both patent-free video format) to surpass the levels of H.264/MPEG-4 AVC or VC-1 will be very difficult without infringing dozens if not hundreds of software patents. Chris DiBona of Google, while talking about improving Ogg Theora as part of its inclusion in HTML 5 specifications said, "Here’s the challenge: Can Theora move forward without infringing on the other video compression patents?" Just the number of companies and organization that hold patents over H.264 is astounding, and includes: Columbia University, Electronics and Telecommunications Research Institute of Korea (ETRI), France Télécom, Fujitsu, LG Electronics, Matsushita, Mitsubishi, Microsoft, Motorola, Nokia, Philips, Robert Bosch GmbH, Samsung, Sharp, Sony, Toshiba, and Victor Company of Japan (JVC). As is the amount of royalties to be paid ("[t]he maximum royalty for these rights payable by an Enterprise (company and greater than 50% owned subsidiaries) is $3.5 million per year in 2005-2006, $4.25 million per year in 2007-08 and $5 million per year in 2009-10"; with royalty per unit of a decoder-encoder costing upto USD 0.20.)

Indeed, even the most diligent companies cannot guard themselves against software patents. FFII estimates that a very simple online shopping website would violate twenty different patents at the very least. Microsoft recently lost a case against i4i when i4i surfaced with a patent covering custom XML as implemented in MS Office 2003 and MS Office 2007. As a result Microsoft had to ship patches to its millions of customers, to disable the functionality and bypass that patent. The manufacturers of BlackBerry, the Canadian company Research in Motion, had to shell out USD 617 million as settlement to NTP over wireless push e-mail, as it was otherwise faced with the possibility of the court shutting down the BlackBerry service in the U.S. This happened despite there being a well-known method of doing so pre-dating the NTP patents. NTP has also filed cases against AT&T, Sprint Nextel, T-Mobile, Verizon Wireless, and Palm Inc. Microsoft was also hit by Visto Corporation over those same NTP patents, which had been licensed to Visto (a startup).
- Don't These Cases Show How Software Patents Help Small Companies?
  
  The astute reader might be tempted to ask: are not all of these examples of small companies getting their dues from larger companies? Doesn't all of this show that software patents actually help small and medium enterprises (SMEs)? The answer to that is: no. To see why, we need to note the common thread binding i4i, NTP, and Visto. None of them were, at the time of their lawsuits, actually creating new software, and NTP was an out-and-out "non-practising entity"/"patent holding company" AKA, patent troll. i4i was in the process of closing shop, and Visto had just started up. None of these were actually practising the patent. None of these were producing any other software. Thus, none of these companies had anything to lose by going after big companies. In other words, the likes of Microsoft, RIM, Verizon, AT&T, etc., could not file counter-suits of patent infringement, which is normally what happens when SMEs try to assert patent rights against larger corporations. For every patent that the large corporation violates of the smaller corporation, the smaler corporation would be violating at least ten of the larger corporation's. Software patents are more helpful for software companies as a tool for cross-licensing rather than as a way of earning royalties. Even this does not work as a strategy against patent trolls.
Thus, the assertion that was made at the beginning is borne out: software patents help only patent trolls, large corporations that already have large software patent portfolios, and the lawyers who draft these patents and later argue them out in court.
Term of Patents

Twenty years of monopoly rights is outright ludicrous in an industry where the rate of turnover of technology is much faster -- anywhere between two years and five months.
Software Industry Progressed Greatly Without Patents

In India, software patents have never been asserted in courts (even though many have been illegally granted), yet the software industry in India is growing in leaps and bounds. Similarly, most of the big (American) giants of the software industry today grew to their stature by using copyright to "protect" their software, and not patents.
Copyright Exists for Software

As noted above, the code/expression of any software is internationally protected by copyright law. There is no reason to protect the ideas/functionality of that software as well.
Insufficient Disclosure

When ordinary computer programmers cannot understand what a particular software patent covers (which is the overwhelming case), then the patent is of no use. One of the main incentives of the patent system is to encourage gifted inventors to share their genius with the world. It is not about gifted inventors paying equally gifted lawyers to obfuscate their inventions into gobbledygook so that other gifted inventors can at best hazard a guess as to precisely what is and is not covered by that patent. Thus, this incentive (#2) is not fulfilled by the current system of patents either -- not unless there is a major overhaul of the system. This ties in with the impossibility of ensuring that one is not violating a software patent. If a reasonably smart software developer (who are often working as individuals, and as part of SMEs) cannot quickly ascertain whether one is violating patents, then there is a huge disincentive against developing software in that area at all.
Software Patents Work Against Free/Libre/Open Source Software

Software patents hinder the development of software and FOSS licences, as the licensee is not allowed to restrict the rights of the sub-licensees over and above the restrictions that the licensee has to observe. Thus, all patent clearances obtained by the licensee must be passed on to the sub-licensees. Thus, patented software, though most countries around the world do not recognize them, are generally not included in the default builds of many FOSS operating systems. This inhabits the general adoption of FOSS, since many of the software patents, even though not enforceable in India, are paid heed to by the software that Indians download, and the MP3 and DivX formats are not enabled by default in standard installations of a Linux OS such as Ubuntu.

Conclusion

Currently, the U.S. patent system is being reviewed at the administrative level, the legislative level, as well as the judicial level. At the judicial level, the question of business method patents (and, by extension, software patents) is before the Supreme Court of the United States of America in the form of Bilski v. Kappos. Judge Mayer of the Court of Appeals for the Federal Circuit (CAFC, which heard In re Bilksi) noted that "the patent system has run amok". The Free Software Foundation submitted a most extensive amicus curiae brief to the U.S. Supreme Court, filled with brilliant analysis of software patents and arguments against the patentability of software that is well worth a read.

For more details visit https://cis-india.org/a2k/blogs/arguments-against-software-patents

Time Out Bengaluru - Software Patenting

pranesh — 2013-01-16T06:39:27Z

An article by Akhila Seetharaman published as a precursor to the national public meeting on software patents held on 4th in Bangalore.

Original article on Time Out Bengaluru website

In August this year, the US Patents and Trademarks Office granted Microsoft ownership of “page up” and “page down”. So in theory, no other company can scroll without permission and acknowledgement to Microsoft in monetary terms.

A number of seemingly ubiquitous software ideas have been patented: the use of tabs to shift from one hyperlink to another on a web page, the “Add to Shopping Cart” function that appears on every online store, automated online loan requests, and even reducing image size to make a webpage load faster.

“Most companies register defensive patents to protect themselves, not offensive ones,” said Sunil Abraham of Centre for Internet and Society. “Not many actively pursue patent infringement, but it is still very scary for a small-time entrepreneur.”

At a time when the Indian Patent Office is in the process of putting together a new Manual of Patent Practice and Procedure, the Centre for Internet and Society is holding a one-day consultation on the issue of software patenting in the city. Participants include the Delhi Science Forum, RedHat, IT for Change, Open Space, as well as the Alternative Law Forum.

From mobile phone technology to pacemakers in healthcare, everybody is dependent on software. “Each software patent is a 17-year monopoly on an idea,” said Anivar Aravind of the Free Software User Group Bangalore.

“If formulaic Hindi films were protected by patent laws, we would be able to make only one film,” joked Abraham. The system of software patenting wipes out smaller businesses and innovation, he said. “Software, like poetry and literary works, is already protected by copyright. After all, Bill Gates made his fortunes from copyright and not patents. But many software companies are trying to get additional protection.”

Copyright and patents are both part of intellectual property rights, but copyright restricts the expression of an idea while patents restrict the idea itself, according to Abraham. Under a patenting regime, even before a kid writes one line of code he has to read many patents.”

Kiran Patil of Turtle Linux Lab agreed. “If every little thing is patented, there’s nothing a developer can do.” He cited Richard Stallman, founder of the Free Software Movement and the GNU (a recursive acronym for GNU’s Not Unix) Project, who likened patents to explosive devices: “Software patents are the software project’s equivalent of land mines: each design-decision carries a risk of stepping on a patent, which can destroy your project.”

Worst of all, the world sees those with patents as the innovators, said Patil, which, according to him, is a big misconception.

While corporate giants like Microsoft and IBM fix exchange deals through cross-licensing, smaller companies get left out of the loop entirely. Despite not having many patents of their own, several Indian software companies support software patenting because they have huge contracts with the large software companies in the United States and Europe who do.

The Indian Patent Act of 1970 did not allow for software patents until 2002 when an amendment, which ironically excluded “computer programmes per se” from the scope of patenting, was introduced.

The amendment implied that while computer programmes themselves were not eligible for patents, programmes used in combination with hardware were. The Act was further amended through an ordinance in 2005 to narrow the scope of software excluded, but the ordinance was rejected by the Indian

Parliament and the Act effectively reverted to what it was after the 2002 amendment. “The law has left it somewhat ambiguous,” said Abraham. “Nobody is sure what can or cannot be patented. Many people are using the clause “computer programmes per se” to get pure software patents.”

This occurs either due to incompetence among patent officers or by accident, he said. “While many of the patent officers have expertise in the area of industrial inventions or medical inventions, very few know enough about software patents at the moment.”

-- Akhila Seetharaman

For more details visit https://cis-india.org/news/time-out-software-patenting

World Day Against Software Patents

pranesh — 2013-01-16T07:15:16Z

A global coalition of more than 80 software companies, associations and developers has declared the 24th of September to be the "World Day Against Software Patents". The Hindu, a national daily dedicated one page of its Bangalore edition to software patents and software freedom. Deepa Kurup contributed written two articles titled "Will patenting take the byte out of IT here?" and "How would it be if you read only one type of book?" which reflects some of the concerns of the Free/Libre/Open Source Software community.

Will patenting take the byte out of IT here? [link]

Deepa Kurup

There has been little debate on patent laws and the software industry. Today is World Day Against Software Patents.

IT software, services and outsourcing industry has been rooting for software patenting
Delhi Patent Office receives around 50 applications for software patents every month

BANGALORE: Picture this. Indian mathematicians came up with the concept of the “zero” — often touted as India’s greatest contribution to civilisation — and got a patent for it. By now they would have raked in inestimable amounts in royalty. Seems preposterous? Members of the Free Software community say that patenting every other algorithm would be somewhat in the same league.

While there has been substantial discussion on how patents will affect the pharmaceutical sector, there has been little debate about its implications on the software industry. To the layman, software patenting sounds like an abstract issue applicable to an even more abstract domain. However, with a growing software industry which is trying to spread its indigenous roots, the issue becomes an important one.

Traditionally, software comes under the Copyright Law (just like any literary work) and anyone who writes a program owns it. After Indian Parliament in 2005 scrapped an ordinance which declared “software in combination with hardware” patentable, the controversial and ambiguous clause — “software per se” — has now resurfaced in a recently formulated Patent Manual.

And how will the common man be affected by this proposed change in the patent manual? For example, when Global Patent Holdings patented usage of images on websites, a bunch of small and big companies had to cough up to $50 million each. And where does this cost reflect? “The consumer will find that products will get a lot more expensive. Take a DVD player which has about 2,000 patents (many of them software-related). Every time a local company makes a DVD player, they have to pay royalties and the costs will naturally be reflected on the sale price,” says Sunil Abraham of Centre for Internet and Society, a research and advocacy organisation.
Backdoor entry

The Free Software community feels that patents will make a backdoor entry, courtesy this manual and that ongoing public consultation (by the Patent Office) does not take their voices into account. Mr. Abraham says: “We feel that the powerful software lobbies around are pushing for this clause. If allowed, it will affect the basis of innovation, and will in turn affect the industry.” While the Bangalore consultation was “postponed indefinitely,” the Patent Office in its Delhi meeting said this issue called for an “exclusive meeting with the software industry.”

The powerful IT software, services and outsourcing industry has been rooting for software patenting. Under the guile of the seemingly innocuous clause in the Indian Patent Bill 2005, software companies and the MNC lobby is trying to carve out a slice for the specific “software embedded with hardware” industry saying that it will increase the value of indigenous home-grown software, pump up software exports and thereby rake in greater revenue.

However, the other side of the story is worth telling. Software, per se, is simply a set of instructions to carry out a certain process. Software experts put forth the argument that big corporations — with money, muscle and hired talent — will seek to impose patents along the software value chain, starting from source code to the recent demand for “embedded software.”

Sources in the Delhi Patent Office say that they receive around 50 applications for software patents every month. In the U.S. 25,000 patents are granted every year. In a software-driven world, blurring the lines between software and software “per se” could be risky. “Patenting is an expensive and tedious process. The challenge for every programmer would be to verify each time, to see if any two lines of his code would infringe upon a patent. In the U.S., a single verification can cost as much as $5,000. The fundamental issue is that if I arrive at anything independently, should I not use it only because someone had got it patented before me?” asks a senior official at Red Hat, an open source service provider.

A paper written by members of the Alternative Law Forum (ALF), the case against software patenting is presented as a very basic one. “Software evolves much faster than other industries, even with its own hardware industry. Microprocessors double in speed every two years. So, a patent that lasts up to 17 years (minimum period -15) is alarming. In this field, the idea underlying may remain the same but a product has to be replaced on an average of every two years,” it states. The paper also points out that in software “research costs are little because ideas are as abundant as air.”

Prashant Iyengar of ALF feels that patent laws will effectively curtail innovation, like it has done in the U.S. “Software, unlike other industries in India, is end-driven but is also on a “body shopping” model. Given that, a strong start-up company will be either be shut down or bought over if patent laws come in,” he explains.

How would it be if you read only one type of book? [link]

Deepa Kurup

Little or no attention is paid to what is being taught in schools and colleges

BANGALORE: A computer literacy programme in a public sector organisation teaches the following modules: MS Office, MS Power Point, MS Excelsheet and Internet Explorer. A glance through the “computer syllabus” in most schools, and the list is similar. All items on this checklist have one thing in common: proprietary software. So, if every computer user is being taught exclusively on proprietary platforms, would they ever be comfortable switching to the easier, cheaper and readily available alternatives?

Advocates of Free Software — software which can be used, studied and distributed without restriction — say that this is a ploy by proprietors to turn learners into potential customers. They allege that educational systems and the State are in cahoots with these large corporations which insist that children and learning adults be taught to only follow their system.

In a recent meeting with a State Government official about the use of Free Software on e-governance platforms, the official complained that none of his officials knew how to use it or repair it if things went wrong.

“This takes you to the root of the problem,” says Sunil Abraham of Centre for Internet and Society, Bangalore. “Students are taught to use only proprietary software. The Government is subsidising training in proprietary technology and little or no attention is paid to what is being taught in schools and colleges,” he explains.

The “back-office” tag that our IT industry has learnt to live with is also a product of this malaise, experts point out. “When students learn only proprietary software, they will qualify only as computer operators and never learn about using the nuts and bolts of the profession. This is one of the reasons why there are no innovative products that come out of this country,” says Mr. Abraham.
Simple analogy

A simple analogy would be that of a child taking up reading as a habit. If a child reads a lot of books, they say, they learn to write and express better. Academics feel that in the absence of any familiarity with Free Software, where the source is easily available, engineering students and computer graduates never get to read any code and are thus hardly familiar with the languages.

FOSS supporters have written to the Ministry of Human Resource Development and several universities to point this out. Anivar Aravind, a member of Free Software Users Group, says that the progress so far has been staggered. Recently, CDAC and Anna Univeristy (KB Chandrashekar Research Centre) came up with a Free Software syllabus and offers trained to teachers in engineering colleges.
Cost factor

A study by International Open Source Network (an UNDP initiative) study on FOSS and education states that using open source software could reduce the costs involved in ICT education significantly. In a country like ours, this fact that Open Source Software usually involves low or no cost would be perceived as an important step towards reducing the digital divide. With no licensing fee, they can be made available on CD or downloaded.

For more details visit https://cis-india.org/news/world-day-against-software-patents

Second Response to Draft National Policy on Open Standards for e-Governance

pranesh — 2009-07-07T16:49:37Z

Another draft (labelled "version 2", dated May 26, 2009) of the draft national policy on open standards for e-governance was made available to Fosscomm, while many software companies were speaking out against NASSCOM's position on the policy. CIS drafted a second response addressing both the allegations against NASSCOM as well as the few shortcomings we perceive in the draft policy.

To
Shri Shankar Aggrawal
Joint Secretary (e-Governance)
Department of Information Technology
Ministry of Communications and Information Technology

Tuesday, July 7, 2009

Dear Sir,

Sub: Comments on Draft National Policy on Open Standards for e-Governance (version 2)

I am writing on behalf of the Centre for Internet and Society, which is a Bangalore-based civil society organization involved both in research and policy advocacy. Public accountability and digital pluralism are two of our core concerns, and it is for this that we are writing to you today. As a natural corollary of our mission, we aim at representing the concerns of citizens and consumers. You would recall that we had submitted comments to the call for comments you had put out for the draft National Policy on Open Standards for e-Governance last year (archived at <http://cis-india.org/advocacy/os/iosp/the-response/>).

We have recently received what appears to be a newer draft (version 2) of the National Policy on Open Standards for e-Governance, dated May 26, 2009. We are yet again very pleased to note the progressive nature of this document and wish to congratulate the government on its decision to promote the interests of the citizens of India over the narrow partisan interests of a few companies which wish to promote proprietary standards.

It has brought to our notice by some in the software industry that the National Association of Software and Services Companies (NASSCOM) has argued for the dilution of the definition of open standards by including standards licensed under “reasonable and non-discriminatory” terms to be considered “open”, and has also called for multiple standards in the same domain to be considered valid as a rule under the policy. We believe both these demands go against the interest of consumers of standards — which in this case is the Indian government — and are thus against the interest of citizens as well, since the Indian government handles data on behalf of its citizens.

Even “reasonable and non-discriminatory” terms of licensing of standards are in fact discriminatory as they prevent the development of free/libre/open source software based on those standards. And while having multiple implementations of a standard is beneficial as it increases consumer (i.e., governmental) choice, having multiple incompatible standards is detrimental to the government's interest as the policy itself recognizes in paragraph 4.2, and the very purpose (as enumerated in paragraphs 1, 3, and 4) of having standards is defeated. Even if the multiple standards are bi-directionally interoperable, additional costs are incurred in having concurrent multiple standards.

Thus, one hopes that the the threshold of “national interest” mentioned in paragraph 6.4.1 is set to a high level. Lastly, the views put forth by NASSCOM seem not to be truly legitimate as it has been the complaint of some that NASSCOM did not hold an open consultation with its own members before formulating its views. There are software giants, including IBM, Sun, and Red Hat, that have openly criticized the NASSCOMM position on open standards. More importantly, NASSCOM's position does not concur with what we believe is in the best interest of small and medium software enterprises, which constitute the bulk of the Indian software industry. We pray that you shall keep this in mind while considering NASSCOM's views.

We believe that apart from the technical reasons to favour open standards, there are many public interest reasons as well. We believe that the adoption of open standards is a step towards the promotion of equitable access to knowledge to all the people of our country. We further believe that public accountability will be served greatly by adoption of an open standards policy by the Central and State governments. While even developed countries (such as those of the EU) are mandating open standards in all governmental departments, processes, and interactions, it is developing countries that stand to gain most from open standards. Proprietary standards place a larger burden on developing economies than developed as developing economies have a greater need to participate in the global network by using standards, but do have lesser capabilities than developed economies in terms of paying for royalties.

On the document itself, while there are many reasons to hail it, we believe there are still a few shortcomings which we wish to bring to your notice.

Issue 1: Possibility of following letter of policy while violating its spirit

Explanation
Sometimes private companies can interfere with the standardisation process by exerting undue influence on the members of the standard setting body. That such undue influence have been sought to be applied even in India recently shows that this is not mere conjecture or idle speculation. Given this background, the document should note this as a problem and note that remedial measures could be undertaken in the event such undue influence comes to light.

Resolution
Introduce language, such as that used in the EU EIF, stating:
“Practices distorting the definition and evolution of open standards must be addressed immediately to protect the integrity of the standardisation process.”

Issue 2: Patenting and licensing of government-developed standards

Explanation
Paragraph 6.3 of the draft policy allows the government to opt for the development of a new standard by a Government of India-identified agency in case no standard is found to meet the government's functional requirements. However, it is not clear under what terms this standard will be available.

Resolution
Introduce a paragraph 6.3.1 stating:
“Any standard developed by or on behalf of the government shall be patent-free and the specifications of such a standard will be published online and will be available to all for no cost. Along with the standard, the government shall also provide, or shall cause to be provided, a free/libre/open source reference implementation of that standard.”

Issue 3: No framework provided for review or phasing out interim standards

Explanation
Paragraph 6.2 permits the government to adopt a non-open “interim” standard (one which does not fulfil all the mandatory requirements of open standards as laid out in 5.1) if no open standard exists in the specific domain for which the standard is required. This however does not have a clause necessitating the phasing out of such an interim standard.

Resolution
A review mechanism should be provided for periodic evaluation of all standards selected by the government, especially those designated as interim standards. A new paragraph 7.1.1 could be added:
“All standards selected through the processes outlined in this policy shall undergo an annual review by the Apex Body on e-Governance Standards, and all those designated as interim standards shall be reviewed biannually.”

Issue 4: Problematic definition in the glossary

Explanation
In Appendix A, the definition of “patents” (A.12) states: “The additional qualification 'utility patents' is used in countries such as the United States to distinguish them from other types of patents but should not be confused with utility models granted by other countries. Examples of particular species of patents for inventions include biological patents, business method patents, chemical patents and software patents.” Many of these references are U.S.-specific and are not valid forms of patents in India (e.g. biological patents, business method patents, and software patents).

Resolution
Delete the last two sentences in A.12

We once again wish to compliment the government on developing such a strong policy on open standards, and hope that our suggestions are incorporated into the text of the final version. We further hope that the policy will be notified at the earliest, as there has already been considerable opportunity for the public and industry to comment on the draft versions of the policy.

Yours sincerely,

Pranesh Prakash
Programme Manager
Centre for Internet and Society

For more details visit https://cis-india.org/openness/publications/standards/second-response

Lecture by Eben Moglen and Mishi Choudhary

pranesh — 2011-08-23T02:55:59Z

The Software Freedom Law Center, National Law School, and the Centre for Internet and Society organised a lecture by Mishi Choudhary and Eben Moglen for students of NLS on Saturday, December 13, 2008.

Saturday, December 13, 2008 had Mishi Choudhary and Eben Moglen of the New York-based Software Freedom Law Center speaking to the students of the National Law School of India University in Nagarbhavi, Bangalore, in a talk organized by CIS.

Mishi Choudhary, who will head the Software Freedom Law Center in New Delhi, spoke on "Globalising Public Interest Law: The SFLC Model". She told the students about the importance of non-profit legal work as well as its viability as a career choice. She also laid out the background to the work that SFLC does, and traced a brief history of software patent cases

Eben Moglen chose to speak on "Who Killed Intellectual Property and Why We Did It?". He started off by talking of the interconnections between law and societal change: how law can't keep pace with the changes we see around us, and how law actually sometimes changes in the reverse direction, while trying to maintain the status quo.

This is not a new phenomenon, he noted, and that when law is responsive to anybody, it listens to the 'people of the past' more carefully than the 'people of the future'. This, he says, is compounded by the fact that the primary mode of change in the law is not legislation (since there is nothing legislators hate more than legislating), and that the better lawyers usually represent only those who can afford to pay them, hence resulting in systemic injustice. He emphasised that the clients of the SFLC, on the other hand, are people who create software worth billions of dollars, but who do not own it.

On that point of creation for the purpose of sharing and not owning, a student raised the question of why proprietary rights shouldn't exist in creations of the intellect. In response Mr. Moglen pointed out that while his personal opinions might be different, the Software Freedom Law Center does not seek to bring into dispute the concept of property rights in software, nor the fundamentals of patent law: it is merely concerned with the scope of patent law, and seeks a literal enforcement of patent law as it exists in most jurisdictions.

Another question that cropped up was on the economics of software creation and the anti-competitive nature of free software. To this, Mr. Moglen provided a brief summary of the tragedy of the anticommons by using land to be acquired for public works in the centre of a city as an example. In software, this problem is only exacerbated, he pointed out. Most physical creations over which patents are granted have something like 8 or 10 steps. Software code is different because it contains thousands of instructions. Even big companies face the anticommons problem; but they manage to evade it by cross-licensing agreements which results in efficient transactions for them since it involves no exchange of money whatsoever. Small companies are in a worse situation, since they don't have those kinds of patent portfolios to be able to enter into cross-licensing agreements, no matter how innovative they are. Thus, in effect, the system is rigged against them. This provides a partial answer to the antitrust question, he noted. Competition law is actual in favour of free software. The right to practise a trade or profession, and the right to speech get implicated in any case where a FLOSS-based company is hauled up before a court being accused of conspiring with other to take cost to zero.

Mr. Moglen further explained that when it comes to software, the problem of patenting is very different. A 20-year monopoly is more reasonable from the viewpoint of physical creations. Patent law, however doesn't tailor the rights that are granted by a patent. The problem starts right from the process of granting a patent. The job of a patent office being to apply the tests of non-obviousness, novelty and utility, most patent offices can do a reasonable job in most fields of technological endeavour, since there is a large body of innovation with which the proposed patent can be compared. Software, however, is a recent field with a large number of applications coming in all at once. While the patents that are sought might include claims on ideas and applications that existed in software in 1956, those aren't easy for the patent offices to dig up, since the field of software patents and software itself have not existed for the same length of time.

For more details visit https://cis-india.org/a2k/blogs/lecture-by-eben-moglen-mishi-choudhary

The National Public Meeting on Software Patents

pranesh — 2011-08-23T03:02:56Z

On Saturday, October 4, 2008, the Centre for Internet and Society, with the support of eighteen other organization, held a meeting on the National Public Meeting on Software Patents in the United Theological College campus. The aim of the event was to explore various issues surrounding software patents, especially from the perspective of the draft Patent Manual.

After introductions by Sunil Abraham of CIS, the discussions were kicked off by Sudhir Krishnaswamy (an Assistant Professor at National Law School), who spoke about typology of laws; principle-based arguments for excluding software from patenting; policy-based arguments for the same; and lastly, strategies for combating the patent manual. About the rationale behind excepting software ("computer programmes per se") from patentability, he theorised that given the location of "computer programmes per se" in section 3(k) of the Act, surrounded as it is by "mathematical or business method" and "algorithms", the exception seems to be a principle-based one and not a policy-based one. He also talked about what he saw as the practical realities of the Patent Office, and questioned the role the Draft Manual would actually play in the decisions of Patent Examiners.

He listed out economic arguments as:

Inapplicability of the incentive arguments. The software industry does not need patents since copyright covers software, and even if incentives are required, that is incentive enough;
Return on investment. Short shelf-life, and hence 17-year patent terms are irrelevant when the shelf-life is so small;
New intermediaries are created, who are neither producers nor consumers of software. These intermediaries who help in price-discovery. They discover value in patents which were previously thought neglected by the process known as patent trolling.

Apart from these, he also briefly talked of the legal arguments around software patents, and argued that the question is not only about copyright vs. patent, but also about property vs. contract. He asked questions such as: "What role does copyright play in the software industry, or is contract more important?", and pointed out that while this might have been addressed around a decade ago, those questions need to be revisited given the current scenario. Further, he proposed that the strategies should not revolve solely around the Patent Act and Draft Manual, but around pre- and post-grant oppositions as well.

Prabir Purkayastha of the Delhi Science Forum and Knowledge Commons spoke next, giving a quick run-through of the history, both legal and philosophical, surrounding software patents in India and in the U.S. and Europe (pointing out that most of the wordings of Draft Manual on this point are borrowed from a similar document in the U.K.). He asked the question of why people are opposing software patents. Is it because it is damaging to 'public interest', because it bad for Indian domestic software industry, or because it is an abstract idea which is sought to be patented in the guise of something else? He concluded that ultimately it is not the manual that groups are opposing, but the notion of software patents themselves. Thus, he focussed on how the phrase "per se" used in the Act ought to be interpreted by the Patent Office so as to give credence to the Indian Parliament's rejection (in 2005) of the 2004 patent ordinance (in which section 3(k) read: "a computer programme per se other than its technical application to industry or a combination with hardware"). Lastly, he talked about the various strategies to be employed in the fight against software patents, including pre- and post-grant oppositions.

Dr. Nagarjuna G. of the Free Software Foundation of India focussed on what he termed "the absurdity of software patents". He emphasised how software requires an interpreter or hardware, and hence talk of "software per se" often becomes meaningless. Further, he underlined how embedding software in hardware was not innovation in itself, and stressed ont he changing notions of software and hardware as we evolve technologically. His equation of software with abstract ideas gives us a glimpse into the foundation of his objection to software patents.

First up in the second session (which was more focussed on the manual, and the law in India) was Prashant Iyengar of the Bangalore-based Alternative Law Forum. He first listed out the different kinds of objections to software patents, including the point that there are only limited ways of thinking about programming, as Donald Knuth's The Art of Computer Programming shows. Then he went on to go through the history of software patents in India, from the first software patent, granted in 1996, through the 2002 Amendment, the 2004 Ordinance, the 2005 Amendment, and the 2005 and 2008 Draft Manuals. He looked at the vocabulary surrounding software patents, including the words "per se" and "as such", and the cases and legislations from which the language used in the Draft Manual might have been borrowed. He also started a fruitful debate on the different ways to attack the implicit inclusion of that which is not "computer programmes per se" within the scope of patentable subject matter.

After Prashant was Venkatesh Hariharan of Red Hat. He spoke on the practical benefits and harms of software patents, and spoke at length about the difference between legal protection of software in the form of patents and via copyright. He pointed to data showing that lawyers are the ones who benefit most from software patents, and that software developers were the ones who suffered most. Pointing to such practical issues such as how does one go about coding a simple e-commerce transaction when more than 4000 patents have already been granted in that area, he brought down the level of discussion from abstract notions of laws and legalities to practical experiences of software programmers.

Next, Pranesh Prakash of the Centre for Internet and Society made a presentation on a small sample of software patents that have been applied for in India, and pointed out the infirmities in both the patents that have been applied for, as well as the problems in uncovering these patents because of various errors on the Indian Patent Office website. Going through a few of the patent applications, he showed how a great number applications have very badly worded abstracts, filled with weasel words, whose sole purpose is obfuscating the fact that what is being applied for is a software patent. This, he pointed out, made it difficult to both determine the scope of the applications (subject matter) as well as the innovations contained in the invention (novelty and non-obviousness), and thus difficult to examine from the perspective of pre-grant oppositions.

After these presentations, the meeting continued with the Open House session which had many people making presentations, including Abhas Abhinav of DeepRoot Linux, Arun M. of FSF India, and Joseph C. Matthew, who is the IT Adviser to the Chief Minister, Kerala. With the wrapping up of this session, the proceedings for the day came to a close.

Coverage in the press

Audio Recordings and Slides

The Principles of Patent Law and Introduction to Software Patents

Sudhir Krishnaswamy (National Law School) | mp3 | ogg
Prabir Purkayastha(Delhi Sience Forum) (Knowledge Commons) |mp3 | ogg
Nagarjuna G.(Free Software Foundation of India) | mp3 | ogg

Software Patents in India: The Indian Patent Act and the Draft Patent Manual

Prashant Iyengar(Alternative Law Forum) | mp3 | ogg
Venkatesh Hariharan(Red Hat) | mp3 | ogg

Software Patent Applications in India

Pranesh Prakash (Centre for Internet and Society) | mp3 | ogg | ppt

Open House

Abhas Abhinav (DeepRoot Linux) | mp3 | ogg

Arun M.(Free Software Foundation of India)| mp3 | ogg
Joseph Mathew (IT Adviser to the Chief Minister, Kerala)| mp3 | ogg

For more details visit https://cis-india.org/openness/blog-old/the-national-public-meeting-on-software-patents