The Delhi High Court, in AB Initio v. Assistant Controller, allowed the maintainability of divisional applications, which were rejected by the Controller u/s. 3(k) for being a computer program per se. The DHC, in this judgement, continued with its restrictive interpretation of 3(k), narrowing the scope of inventions which are excluded u/s. 3(k). The Court, having wide options of ‘technical’ tests in its arsenal, applied the ‘technical effect’ test in this case. In this post, apart from discussing the background and arguments raised in the case, I will discuss how the court, relying upon the now scrapped draft 2013 guidelines, applied the technical effect test to decide the patentability of the invention. Further, I show how the present decision continues the debate on the dilution of 3(k) and digital innovation, which I have also discussed here. Lastly, I contrast the outcome of the judgement with MHC’s decision in Microsoft Technology Licensing, LLC, wherein the Court had granted the patent without looking into the question of enablement, independent of the 3(k) inquiry.
Facts & Arguments
The appeal was against the rejection of divisional applications 6500/DELNP/2011 and 6501/DELNP/2011, which were titled “A method for processing data and system thereof.” The Patent Office, among other grounds, rejected the applications on the grounds of unpatentability u/s. 3(k) of the Patent Act.
The Application highlights that earlier stored data sets would often include data with unknown characteristics, e.g. typical value of the data, ranges of value or functional dependencies among the values were not known. As a result, extensive computation resources were required to carry out a functional dependency analysis across fields of a data set.
The invention, which related to a system of data profiling, would use a profiling module to read records from data sources, compute statistics, and other descriptive information relating to the data set. The collected data was then stored in a metadata store, which could be later examined by the user. This, the applicants contended resulted in two things- (i) increased speed of functional dependency analysis; and (ii) saving computer resources.
The Patent Office rejected the patent application u/s. 3(k), contending that the invention merely provided methods to analyse data and stored the information derived from the data in a metadata store. The invention did not affect the ‘internal’ functioning of the computer. It also relied on European Patent Office guidelines to argue that a summary of data or data profiling is only ‘cognitive’ information relevant to human users, thereby it does not produce a technical effect.
The applicant, in their written submissions to the Patent Office, had argued that since the claims relate to ‘functional dependency analysis’, which is a technical task, and the increase in speed is based on internal considerations of the computer, the invention does produce a ‘technical effect.’
What is a technical effect?
In my post on the MHC decision, I had mapped out how DHC has decided 3(k) cases using a medley of undefined and vague tests- technical effect, technical solution to a technical problem, further technical effect, technical advancement etc.
In paragraph 35, the Court noted that the “test which must be applied for determining whether the claim is of a computer program per se/algorithm or otherwise is based on the determination of whether the said program/algorithm has a technical effect.”
How did it define ‘technical effect’? It held that an ‘innovative input’ in the form of process, method, or system, which enhances the computation ability of the processor, would result in a ‘technical effect.’ For instance, the court notes, that if the input system allows the processor to give more efficient and faster output, the effect is technical.
Applying the above principles to the facts, the Court, in paragraph 39, notes five aspects of the claims of the invention, which, in its opinion, indicated technical effect-
- Invention increases the speed of functional dependency.
- It saves computer resources or amounts of computation by operating on the collected information rather than raw data.
- Data profiling without creating a copy of the data, thereby creating an efficient storage system and storing it in metadata store.
- Efficient processing since profiling of the data can be performed parallelly by partition
- Invention relates to a special purpose programmable computer for processing data
Based on the above features, the Court, in paragraph 44, concluded that the above, seen together, does overcome the threshold of ‘technical effect.’ On the respondent’s argument that the invention was merely cognitive data processing, the court noted that the system provided increased functional dependency speed and workload distribution by parallel processing. Therefore, at a system level, there was a technical effect with an impact on computation processing.
Relying on the 2013 CRI guidelines
One can wonder whether the increased speed of functional dependency is sufficient to overcome the technical effect threshold. The Court, to determine what all could qualify as a technical effect, sought reliance on CRI guidelines 2013, which has since been replaced by the 2017 CRI guidelines. (Para 35) The 2013 guidelines said that higher speed, more economical use of memory, more efficient database search strategy, improved user interface etc. demonstrate technical effect. The Court, in this case, has concluded that the invention had a technical effect based on features included in the 2013 guidelines. Do the 2017 guidelines include the same?
The 2017 guidelines do not even mention ‘technical effect.’ It states that to decide the patentability of computer-related-invention, the question which needs to be asked is whether the CRI is technical in nature, involving technical advancement as compared to the existing knowledge.
In Ferid Allani, the Court had noted that ‘technical effect’ find no mention in CRI guidelines after the 2013 guidelines. It nevertheless applied the ‘technical effect’ test, noting that the meaning of technical effect is ‘clear’ due to judicial precedents and patent office practices. There are other judgements of the DHC (here and here) which have upheld applied the technical effect test.
However, the question remains, can the court rely on scrapped guidelines to decide whether the invention is patentable despite 3(k)? As Swaraj pointed out, from 2013 to 2017, there has been a ‘ping-pong’ of sorts when it comes to the CRI guidelines. For instance, the 2013 draft guidelines had a requirement for novel hardware whereas the 2017 guidelines omitted this requirement, the only requirement being ‘technical advancement.’ Clearly, 2013 guidelines were overruled and could not be relied upon for deciding 3(k) disputes (not ignoring the fact that the guidelines were merely draft and non-enforceable in the first place). Irrespective of whether a similar outcome could have been reached relying on 2017 guidelines, the exercise of judicial power by relying on an invalid legal source is fraught with danger.
Diluting 3(k): No surprises here
The Court, relying upon various judgements of the DHC (and Pratibha M. Singh’s Commentary on Patent Law), observed that “the trend of the Courts is to give a restrictive interpretation to the exclusion in Section 3(k) of the Act, and adopt a more benevolent interpretation to what would amount to technical effect.” (Para 36) This approach, the court noted, is aligned with the ‘fast progress’ of technology which is evolving incrementally to achieve greater efficiency.
In this post, I have highlighted the debate about the dilution of 3(k). On one hand, similar to the Court’s view, it is argued that if 3(k) is not read in a restrictive manner, it would stymie innovation in the digital space since almost all innovation is software-based. On the other, it is argued that allowing software-based inventions to be patentable dilutes 3(k). This dilution, in turn, allows monopolization of innovation which is non-patentable and supposed to be in the public domain.
The court, quite unsurprisingly, has stuck to the view that restricting the scope of 3(k) furthers innovation. It is anyone’s guess whether enhanced functional-dependency-analysis speed or an efficient storage system is worth being granted patent protection. As it stands, for the time being, restricting 3(k) is the way to go to further digital innovation for the DHC.
Remanding the application to the Controller
Discussing Microsoft Technology Licensing, LLC , I had criticised the MHC for granting the patent to the impugned invention without going into the question of ‘enablement’, reasoning that it was an irrelevant inquiry u/s. 3(k). In that case, the Controller had rejected the invention for lack of enablement. Therefore, the best course of action would have been to refer the application back to the Controller to decide enablement in light of the court’s findings or decide the question itself.
The Controller, here, rejected the invention for being barred u/s. 3(k), without discussing the question of inventive step. In contrast to the MHC, the Court noted that the Controller’s order was silent on the inventive step, which is a separate inquiry from 3(k). As a result, it remanded the application back to the Controller to decide the question of inventive step u/s. 2(1)(ja).
This affirms the position I had argued in the previous post, that 3(k) is not determinative of the patentability of the invention. Rather, the test of enablement and inventive step, apart from being separate inquiries, are just as important and relevant to deciding patentability.