The Ping-Ponging Paradigm of Patenting Computer Programmes in India (“Software Patenting” 1999-2020)

Patents for Computer Related Inventions (CRIs) or “Software Patents” have, unfortunately, been an evergreen issue in India, with much confusion, lobbying, changes, misunderstandings, and anything else one could imagine, playing its role at some point or the other. As Shamnad had once written – its indeed confusingly confounding! Readers may recall a recent post which discussed the Ferid Allani order. As Sandeep Rathod helpfully pointed out in the comments on that post – that patent application has once again been rejected by the Patent Office. I had intended on writing a follow up post, but in the course of researching on that, I was diverted when I noticed that there doesn’t seem to be a single source that I could find, outlining the rickety road that CRIs have taken in India. So, here is my attempt at outlining the major pit-stops and potholes that CRIs have had the misfortune of bumbling along, in India. Wherever possible, I’ve tried providing a copy of the relevant documents as well. Comments and corrections, if any, are welcome. (At almost 3000 words, this is double the length of our usual posts. However splitting it into two parts didn’t seem to make sense, given that this is an attempt to put all this information in one place)

The Patent (Second Amendment) Act, 2002 – introduction of “computer programmes per se” to the Act

After India’s signing of the WTO TRIPS Agreement, there were a set of amendments required to bring the 1970 Patent Act in line with the requirement of certain TRIPS standards. “Computer Programs” were mentioned explicitly for the first time in The Patents (Second Amendment) Bill, 1999, when it listed in its exclusions: “a mathematical or business method or a computer program or algorithms”. This bill was referred to a Joint Parliamentary Committee, who then presented their Report on 21 Dec 2001. (PDF here). The JPC Report recommended adding “per se” to “Computer Programmes” with the following reasons (emphasis, my own) :-

“In the new proposed clause (k) the words ”per se” have been inserted. This change has been proposed because sometimes the computer programme may include certain other things, ancillary thereto or developed thereon. The intention here is not to reject them for grant of patent if they are inventions. However, the computer programmes as such are not intended to be granted patent. This amendment has been proposed to clarify the purpose.”

This bill was passed as the Patents (Second Amendment) Act, 2002 (PDF here) implying an agreement with the rationale behind its inclusion. The resultant provision:

“a mathematical or business method or a computer programme per se or algorithms;”

The 2004 Patents (Amendment) Ordinance and its repeal – Rejecting the dilution of S.3(k) exclusion

Dec 27, 2004 saw the promulgation of an ordinance (PDF here) to amend the Patent Act. The ordinance proposed splitting 3(k) into two sub-sections, which would’ve effectively diluted the exclusion:

“(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;”

The same day, the then Union Minister of Commerce and Industry, Mr Kamal Nath, issued what seems to be an official statement (PDF here) saying, “In IT, the trend is to have software in combination with or embedded in hardware – such as in computers or cell phones or a variety of other gadgets. Software as such has no patent protection (the protection available is by way of copyright); but the changing technological environment has made it necessary to provide for patents when software has technical applications in industry in combination with hardware. This has been a demand of NASSCOM.”  ….(and later)…  “We have introduced a provision for patenting of software that is embedded in hardware”

A plain reading of the ordinance, along with the explanatory statement by the Minister, seems to indicate that there is already a split in understanding the proposed provision. On one hand, the language of the proposed amendment essentially says, there are to be no patents for computer programes as such, but patents can be granted if the subject matter is a computer programme’s technical application to industry, or a combination of software and hardware. Whereas the Minister’s statement indicates that patents can only be provided when there is technical application in combination with hardware. The minister also indicates that NASSCOM had asked for this.

The phrases ‘in combination with hardware‘ and ‘embedded in hardware‘ seem straightforward – i.e., software alone is not patentable, but software in combination with hardware, or software embedded in hardware would be patentable (subject to the usual novelty, non-obviousness and utility standards). Certainly the drawing of specific boundaries may be a bit more difficult, but conceptually the idea seems clear. However, the meaning of the phrase ‘technical application to industry‘ seems unclear. Personally, I would imagine that any computer programme is capable of being described as having technical application. Or to phrase it in the negative: are there any computer programmes that can be posited to have absolutely no technical application to industry? I guess it would be possible if one were to narrowly construe the words ‘technical’, ‘application’, and ‘industry’.

In any case – this ordinance was repealed a few months later, on 4th April, 2005, by the Patents (Amendment) Act, 2005. Therefore the language of S.3(k) went back to “a mathematical or business method or a computer programme per se or algorithms;”. Further, Lok Sabha and Rajya Sabha debate records show that the discussions were held on the topic of disapproval of the 2004 Ordinance, in combination with the discussions on the Amendment Act 2005. This, along with a return to the previous language, would indicate a clear intention of the Parliament to prevent this type of dilution of the Section 3(k) exclusion. Interestingly, in his Press note on the Patent Amendment Bill (just before it was passed as an Act), Mr Kamal Nath stated, “It is proposed to omit the clarification relating to patenting of software related inventions introduced by the Ordinance as Section 3(k) and 3 (ka). The clarification was objected to on the ground that this may give rise to monopoly of multinationals.”

The Draft(ed) Manuals on Patent Practice and Procedure

There was a 2005 draft Manual on Patent Practice and Procedure (PDF here) that seems to have been prepared just prior to the repealing of the 2004 Patent Ordinance. The manual included a 14 page annexure on Computer Related Inventions, outlining examples, definitions and relevant jurisprudence. Unfortunately, given that it was only a draft manual, and based on the repealed ordinance, it is of questionable utility. It does however provide the first definition of “Computer Programme Product”:

“computer program product is claimed as “A computer program product in computer readable medium”, “A computer-readable storage medium having a program recorded thereon”, etc. In such cases the claims are treated as relating to software per se, irrespective of the medium of its storage and are not held patentable.”

Further, they also describe certain examples and then state: “These are the programs solely intellectual in its context and hence not allowable.”

They also explain ‘technical effect’ for the first time, as: “The method claim should clearly define the steps involved in carrying out the invention. It should have a technical effect. In other words, it should solve a technical problem.”

Once again – the draft 2005 Manual was prepared when the 2004 ordinance was in place, and therefore is of limited utility.

Fast forward to around March 2008, and the DIPP issued another Draft Manual of Patent Practice and Procedure. By this time, despite the explicit exclusion of Section 3(k), the Patent office had already started granting ‘software patents’. Therefore, when the draft manual once again used diluting language (i.e., narrowing the exclusion), it caused a great stir. I’m now unable to find a copy of this draft manual anywhere, however as per our old post here, in the manual “Section 4.11.6 seeks to draw a distinction between “software per se” and “software having its technical application in the industry””. Prashant had also written a post around it, here. The remainder of the year saw several rounds of consultation on the draft manual. Shamnad reported that in one of these rounds, the chair of the stakeholder’s meeting, the then Joint Secretary of the Dept of Industrial Policy and Promotion, Mr NN Prasad, is reported to have specifically instructed his drafting committee team to remove certain statements in the manual, including the word ‘technical’.

Separately, in Oct 2008, there was also a Parliamentary Standing Committee Report (88th Report on Patents and Trade Marks Systems in India – PDF here) which stated that ” the domain of “per se” in the definition needs to be clearly defined.” [On a complete tangential note – in the same Report, Justice Krishna Iyer makes a note about how the Patent Office seems to have provided contradictory interpretations in the draft Manual vis-a-vis what the Act allows, and that this manual should be done away with altogether in favour of a Patent Handbook. However, he says this in reference to how the manual discusses Section 3(d). You can view his 1 page note in the above Report, or his 8 page letter mincing no words in PDF here]

In late 2010, there seems to have been a Revised Draft Manual of Patent Practice and Procedure, as Rajiv had written here. Unfortunately, I’m not able to find a copy of this anywhere. CIS has a blog post going through some of its provisions here.

2011 Manual on Patent Practice and Procedure

In March, 2011, version 1.11 of a Manual on Patent Practice and Procedure was published – PDF here. It stated that it was the result of stakeholder feedback on the earlier draft (though, I believe it mistakenly refers to a non-existent 2009 draft, instead of the 2008 draft), and that it would be updated from time to time. It mentioned that it did not have the force of law, and was just meant to help as a guide. A reading of the short paragraph it included on Section 3(k), would lend one to understanding that the exclusion under Section 3(k) is a broad one, and that a computer programme related subject matter would be patentable only if it had a non computer-programme related claim or claims that were essential to the subject matter. The relevant text is:

If the claimed subject matter in a patent application is only a computer programme, it is considered as a computer programme  per se and hence not patentable. Claims directed at ‘computer programme products’ are computer programmes per se stored in a computer readable medium and as such are not allowable. Even if the claims, inter alia, contain a subject matter which is not a computer programme, it is examined whether such subject matter is sufficiently disclosed in the specification and forms an essential part of the invention.

These guidelines, perhaps, could have further discussed this with examples, as it did not seem to stem the confusion as to the exact scope of the Section 3(k) exclusion.

The IPAB’s decision in the Yahoo v Controller case also came soon after this (Dec 2011), and its pronouncement is worth noting: That if the claimed subject matter is excluded under Section 3(k), then there is no need to look at whether novelty, non-obviousness, and utility are satisfied, as it is non-patentable regardless.

2013 Draft Guidelines on Computer Related Inventions 

On June 28th, 2013, the Patent Office came up with Draft Guidelines on Examination  Computer Related Inventions (PDF here). These draft guidelines acknowledge the rejection of the language in the 2002 Patent Amendment Ordinance as legislative intent to maintain the original scope of Section 3(k). They also provide several definitions. Two parts in particular are worth noting: First, 3.15+3.16 mention that not all ‘technical effects’ will amount to ‘technical advancement’. Then,  5.4.5+5.4.6 attempt to explain the role of hardware in determination of patentability of a CRI, by indicating that a general purpose machine would place it within Section 3(k)’s ambit, but then also mentioning that new or novel hardware with computer programmes could be patentable.

As expected, there was some stiff opposition to this requirement of novel hardware. However, as Aparajita noted in her post examining the stakeholder feedback on these Guidelines – though unpopular, this interpretation by the Guidelines seemed to be in line with the law as it existed, and to remove the novel hardware requirement would be to ignore the legislative intent in drafting the text of “computer programmes per se” (i.e., the 2004-2005 events regarding the Patent Amendments).

2015-16-17 (!) : Guidelines for Examination of Computer Related Inventions 

Till here, it had seemed like there was a fair attempt at explaining and interpreting Section 3(k). However, the 2015 Guidelines seem to have gone well beyond what the explicit language of the section allowed for (see my earlier post here) – as it suddenly opened the door to not just making computer programmes patentable, but also business methods and mathematical models – which had a blanket exclusion under Section 3(k). As I had mentioned in that post, these Guidelines were guiding patent examiners towards making decisions that the law explicitly disallowed. As could be expected, this met with fierce criticism and protest from several stakeholders from civil society, open source movements, etc. And also as expected, it received a warm welcome from industry bodies and law firms / lawyers.

The Guidelines, which had been notified on 21st August 2015, were then ordered to be kept in abeyance, via a 14th December 2015 order (PDF here) by the then Controller General, Mr Om Prakash Gupta, until contentious issues were resolved. In the mean time, Chapter 08.03.05.10 of the 2011 Patent Manual was to be used.

As a side note that may be relevant to interested readers: the Delhi High Court decided Ericsson vs Intex in early 2015, wherein it stated, ““thus, it […] appears to me prima facie that any invention which has a technical contribution or has a technical effect and is not merely a computer program per se […] and […] is patentable”. See Kartik’s 2 part post here and here for more.

19th February, 2016 saw a new set of Guidelines on Examination of Computer Related Inventions (PDF here). These Guidelines took a u-turn, and once again broadened the exclusion under Section 3(k). It introduced a 3-step test to checking for patentability of CRIs – which included (i) properly identifying the actual contribution, (ii) denying outright if the contribution was a mathematical or business model or algorithm, (iii) requirement of novel hardware, or contribution to be in both the computer programme as well as hardware before proceeding to other steps of patentability. Rajiv did a two part post examining these guidelines here and here. (Don’t miss the 47-comment long comment thread on part 1 of those posts!). I would argue that in light of the lack of clarifying interpretations from a Court or Parliamentary proceedings, these guidelines provided the best policy interpretation of Section 3(k), as well as the most clarity in how to go about examining such applications.

The 2017 Revised Guidelines for Examination of Computer Related Inventions

The novel hardware requirement in the 2016 Guidelines immediately threw up objections from the pro-‘software patent’ camp. And soon enough, on 30th June 2017, the Revised Guidelines for Examination of Computer Related Inventions, 2017 (PDF here) were issued via an Order from the Controller General. Balaji did a wonderful post on the basis of RTI findings on the question-filled circumstances that led to the issuance of the 2017 Guidelines. Amongst other things, he’s noted that the Controller General was omitted from some of the exchanges, that Deity conducted its own stakeholder meeting separately, and that for some reason, the DoT was also brought in to review the Expert Committee’s recommendations to the Controller General. The post is definitely worth reading if one is interested in trying to understand this ping-ponging of policy positions.

In any case, if readers haven’t guessed already, there was yet another reversal of position, with the three step test, along with the ‘novel hardware’ requirement was removed. All the examples, describing patent-ineligible applications, from the earlier guidelines were also removed. You can read more on our earlier post here. Significantly, it now includes a further dilution as can be seen in clause 4.5 of the Guidelines:

“…it is important to ascertain from the nature of the claimed Computer-related invention whether it is of a technical nature involving technical advancement as compared to the existing knowledge or having economic significance or both, and is not subject to exclusion under Section 3 of the Patents Act.”

Given that this seems to once again raise more questions than it answers, especially with the lack of any examples – it wouldn’t be a surprise if there is yet another Guideline framing exercise coming up soon.

The 2019 Manual of Patent Office Practice and Procedure

On March 1st, 2019, a Draft Manual of Patent Office Practice and Procedure (PDF here) was released along with a call for comments on it. Compared to the 2011 Manual, the 2019 Draft Manual doesn’t seem to substantively change much, but does remove language regarding what types of CRIs would not be patentable – thus in effect not really clarifying the scope of Section 3(k) here either. Later in the year, the Controller General’s office released the 2019 Manual of Patent Office Practice and Procedure (PDF here), as being in effect from Nov 26th, 2019, superseding the 2011 Manual. From a quick glance, the only difference between the Draft and the Final version (with regard to Section 3(k)) is the introduction of a line stating “For the purpose of this clause, refer the Revised Guidelines for Examination of Computer-related Inventions (CRIs), 2017.”

After all this back and forth – the question remains though – where does all of this leave us? Perhaps simply more confused than when we started off on this, and without really having an idea of when this will be clarified!

PS 1 – For additional reading, do check SFLC’s page on software patents here. Having regularly contributed to the debate in India, they have a lot of first hand information as well.

PS 2 – For some posts on how Controllers have dealt with Section 3(k) issues, you can scroll through Rajiv’s posts from this link