TO BE OR NOT BE PATENT FREE!


Even though India fared dismally in the recent rankings on E-governance, losing 26 places in the World E-Government Readiness Index 2008 and ranking seven places below China in the Web Assessment Survey, the attempt of the Indian Government to establish E-Governance throughout the country continues undeterred. This persistence has been applauded by the United Nations. In particular, India’s attempt at establishing a comprehensive national portal http://india.gov.in/, to account for the important policies and the strategy of the government in relation to issues of E-Governance has been praised.

A visit to the national portal reflects upon the firm commitment of the government to E-Governance. The end aim being to make all government services accessible to the common man in his locality and ensuring efficiency, transparency and reliability of such services at affordable costs. In its most recent attempt, in pursuance of the National E-Governance Plan (NeGP), the Indian Government released a Draft Policy on Open Standards for E-Governance. The Draft Policy lays down certain basic standards which a format adopted by the Government, for it’s E-Governance Plan, should necessarily possess.

The importance of this policy is to be understood in light of the basic standards of E-Governance identified by Sameer Sachdeva in his White Paper “E-Governance Standards in India”. To elaborate, the White Paper authored by Sameer Sachdeva for the United Nations has identified some basic characteristics which a format adopted in India needs to necessarily possess, some of these characteristics being adequate funding, privacy, authentication, interoperability, standardisation and ability to use local language. The question which then arises, against the background of the White Paper, is whether the Draft Policy is satisfactory.

It is submitted that while the policy works well against most of the parameters identified by Sachdeva, there are some significant problems. The biggest problem relates to the possible interpretation of the guiding principle for selection of standards which states that the standard should be openly implementable, i.e., every agency should have the freedom to implement the specification without royalty and without any patent related encumbrances till the life of the standard. It is submitted that on a literal reading of the principle the requirement is two-fold, i.e., a standard should be royalty free and possibly patent free. The patent free requirement creates a significant problem as both the international standards, OOXML and ODF, are protected by patents. It is a separate issue that originators of both these standards have released not to sue covenants in relation to use of their standards. This post seeks to elaborate on the possible concerns which arise if the guiding principle was to be interpreted as being patent free.


Definition and Characteristics

In order to address this debate we must first have a working understanding of open standards. The following definition is pretty comprehensive and has been developed and adopted by the International Telecommunications Union.

As per the definition:

“Open Standards” are standards which are made available to the general public and are developed (or approved) and maintained via a collaborative and consensus driven process. “Open Standards” facilitate interoperability and data exchange among different products or services and are intended for widespread adoption.

Other elements of “Open Standards” include, but are not limited to:

  • Collaborative process – voluntary and market driven development (or approval) following a transparent consensus driven process that is reasonably open to all interested parties.

  • Reasonably balanced – ensures that the process is not dominated by any one interest group.

  • Due process – includes consideration of and response to comments by interested parties.

  • Intellectual property rights (IPRs) – IPRs essential to implement the standard to be licensed to all applicants on a worldwide, non-discriminatory basis, either (1) for free and under other reasonable terms and conditions or (2) on reasonable terms and conditions (which may include monetary compensation). Negotiations are left to the parties concerned and are performed outside the SDO.

  • Quality and level of detail – sufficient to permit the development of a variety of competing implementations of interoperable products or services. Standardized interfaces are not hidden, or controlled other than by the SDO promulgating the standard.

  • Publicly available – easily available for implementation and use, at a reasonable price. Publication of the text of a standard by others is permitted only with the prior approval of the SDO.

  • On-going support – maintained and supported over a long period of time.

From the above, it would appear that the only requirements are that the file format be freely available, easily accessible, interoperable etc. However, none of the criteria are compromised upon if the format enjoys patent or other intellectual property protection. In practice, in order to ensure that “openness” (in terms of accessibility, availability, open life cycle) is met, it is presumed that the entity concerned would not enforce its intellectual property rights against the other entities using the standards. Thus, the Indian policy, if interpreted to mean ‘patent free’, would fly in the face of industrial practice.


Openness and ISO standardisation

The criterion of “openness” is important in that it is a step forward from the RAND standards presently advocated by ISO. To elaborate, as per ISO norms, a format can only become a standard if offered on reasonable and non-discriminatory (RAND) terms. Thus, it is possible that under RAND terms the originators or owners of the standards can make the standard available on payment of ‘reasonable’ royalty. Interestingly, both ODF and OOXML have exceeded the RAND requirement, and have actually made their formats freely available on the internet. Hence, the point is that in practice the international standards already satisfy the mandate of “openness”, in terms of being freely available, as stipulated in the Draft Policy. For instance, Microsoft has made the 6,000-page specification of OOXML freely available. (The Open Specification Promise is dealt with in the next section and the details are available here). Microsoft is the originator of OOXML and Sun Microsystems is the originator of ODF. (Refer to B. Winterford, Developers Warned over OOXML Patent Risk available here).

Hence, the question is what is the requirement to have a “patent free” approach?


Other modes of satisfying the Openness criteria?

Other steps taken by Sun Microsystems (originators of ODF) and Microsoft (originator of OOXML) to satisfy the various aspects of “openness” is to include undertaking “not to sue” covenants.

To elaborate, Microsoft released the OOXML standard under a “covenant not to sue”, which has since been updated into a document called the Open Specification Promise (OSP available here). The OSP is essentially a form of a license agreement designed to give software developers peace of mind that Microsoft won’t come after them in patent-infringement or other intellectual-property litigation upon using the specification. (Refer to B. Winterford, Developers Warned over OOXML Patent Risk available here) Also, the royalty free criterion of “openness” is met by OOXML, as the standard is freely available to be downloaded from ECMA’s website.

On the other hand we have the ODF, which as per Sam Hiser (as per his article on the O’Reilly Open Source Web Platform, available here) is the only format unencumbered by intellectual property rights (IPR) restrictions on its use in other software, as certified by the Software Freedom Law Center. Further, in order to prevent any confusion, the originator of ODF, Sun Microsystems, provides a simple covenant-not-to-sue, which covers any of potential Sun patents used in the development of ODF implementations. As per the license, Sun irrevocably covenants that, subject to certain exceptions, it will not seek to enforce any of its enforceable U.S. or foreign patents against any implementation of the Open Document Format for Office Applications (Open Document) v1.0 Specification, or any subsequent version thereof, in whose development Sun participates to the point of incurring an obligation, as defined by the rules of OASIS, to grant (or commit to grant) patent licenses or make equivalent non-assertion covenants. (Refer to Oasis-Open website and Software Freedom Law Centre website available here and here.)

Lastly, ODF has a similar policy to Microsoft’s royalty-free policy. It is pertinent to note that the royalty free approach for ODF comes from the fact that it is developed and maintained by a technical committee whose members are obligated to a royalty-free policy. Thus, the several thumb rules of “openness” seemed to have been met by ODF.

Thus, we see that both the international standards, OOXML and ODF, are “open”, “reasonably available” and “royalty free”, exceeding the RAND mandate of ISO, irrespective of any kind of intellectual property protection available to such standards.

The issue which then arises is, what does the term “patent free” mean. It is submitted that it ought to mean that the patent ought not to be enforced by any entity implementing this standard and that it ought to be available on royalty free terms. If it is taken to mean that there should be no patent at all, it would immediately exclude the ISO approved standards from the options available to the government. Thus, the government should clarify the meaning of this term immediately.

{I would like to thank Mr. Vijay Kapur, Mr. Sunil Abraham (Director – Policy, The Centre for Internet and Society) and Mr. Shamnad Basheer for their guidance and instructive comments on this post. The response to the Draft Policy on Open Standards by CIS is available here}

5 comments.

  1. Anonymous

    HI,
    Can it be argued that “patent free” means that a particular technical standard is not implicated by patents in India? First, because standards approved under the aegis of the ISO are not mandatory for implementation in India. Second, if section 3(k) of the Patent Act in India excludes certain forms on computer implemented inventions in India, should it not be interpreted as patent free. This is a hypothetical situation though.

    Reply
  2. Suchita Saigal

    Hi,
    Thanks for your comment.

    Your first argument is that it isn’t mandatory for India to adopt an ISO approved standard. It’s pertinent to note that in spite of all the recent debate regarding Microsoft’s OOXML, ISO approved standards are considered to be the best in the business. Often governments prefer standards which are ISO approved. India being no exception, Kerela uses ISO approved ODF file format.
    Thus, its not viable to look for alternatives to ISO approved standards – its a quality mark. More interestingly, the criteria laid down in the policy, apart from the “patent free” criteria, do greatly correspond to the ISO standards criteria.

    Second, Section 3(k) and its interpretation post the draft patent manual is hotly debated. However, assuming that computer implemented inventions do not get protection under the Indian Patent Act, it would be patent free and possibly in line with the policy. However, logically the “patent free” criteria is problematic because these standards do enjoy IP protection internationally, and such protection hasn’t adversely affected the other criteria such as royalty free access etc. Moreover, if one were to accept the interpretation of the draft patent manual, then such computer related of software could be granted patent protection. Consequently, if the concerned file formats could be protected. Then we see a direct clash in the possible interpretation of the Draft Policy and the patent laws.

    Regards,
    Suchita

    Reply
  3. Anonymous Coward

    A few things need to be clarified:
    1. Section 3(k) of the Patent Act talks of “a mathematical method or business method or a computer program per se or algorithms”. It does not talk of “computer-implemented inventions”, which is actually an area of great debate, especially with regard to the Draft Patent Manual.

    2. As regards the openness of OOXML:
    a. it is not a open stakeholders-controlled standard; its development is controlled by Microsoft and Microsoft alone.
    b. There are many elements in OOXML (included for backward compatibility with older MS Office formats) that aren’t defined and documented in the OOXML standard specification. Thus, no one is be in a position to create a viable alternative to Microsoft’s implementation (since they don’t have access to the binary legacy data formats), thus again bringing about vendor lock-in (which is the entire point of avoiding closed standards). OOXML also contains several emulation specifications, the implementation of which require an word processor to be able to emulate MS Word perfectly. That isn’t possible (for the reason stated above) and thus, again puts MS ahead of the competition in an unfair manner.
    c. The OOXML standard document is 6546 pages long. The ODF standard, which achieves the same goal, is only 867 pages.

    The following mandatory requirements of the Draft Policy on Open Standards for e-Governance are not met by OOXML:
    1. Developed in a collaborative and consensus manner and not led by a single agency or a small clique of interested
    parties (
    2. Recursively open and shall not use unpublished extensions (See above comments.)
    3. Should not duplicate already existing standards (ODF is an existing ISO standard for document editing)

    Currently, even Microsoft does not have an implementation of its ISO-approved OOXML standard.

    More later.

    Anonymous Coward

    Reply
  4. Suchita Saigal

    Dear Anonymous,

    Thank you for your feedback.

    As per my understanding, issues relating to patent protection of file formats and related standards would fall under section 3(k) and the question could be characterized as whether computer implemented inventions are more than a mathematical method, a mere application of an algorithm or a computer program per se – whether there is any technical advancement in computer implemented invention.

    Secondly, there are debates about Microsoft’s OOXML and its openness. The controversies addressed by you, have been raised against Microsoft and OOXML in several arenas. I agree with you on several points. TO clarify, in no way does the present post seek to favour Microsoft’s OOXML. The short point is that if the Draft policy were to be interpreted in a certain way (“patent free”), then both the ISO approved standards would fall out of the ambit of choice. Between the two standards, the choice is up to the government, which should decide as per the criteria in the draft policy.

    Reply
  5. Anonymous Coward

    @Suchita
    Mine is not a point about you favouring OOXML or ODF. That is barely something that a non-technical person can go into, and I couldn’t care less to see it addressed on a legal blog. Mine is a simple contention that this sentence that you’ve written: “Thus, we see that both the international standards, OOXML and ODF, are “open”, “reasonably available” and “royalty free”, exceeding the RAND mandate of ISO, irrespective of any kind of intellectual property protection available to such standards,” is false as per the Draft Policy. OOXML is not “open”. What are the criteria that you are using to say that it is “open”? You haven’t clarified that. You list out various criteria for openness (by which OOXML is not open), and then you says it more than satisfies RAND, and then you come to the conclusion, suddenly, that it is open when you discuss “other modes of satisfying the openness criteria”. Whatever happened to satisfying the criteria you listed earlier? Did I miss something there?

    You don’t note the uncertainties in Microsoft’s OSP for which it was roundly criticised (which parts are covered by IP?). One excellent thing that Microsoft did was to resolve many of the criticisms levelled against the OSP immediately. But some of the basic criticisms still remain. And the closed-binary-parts nature of OOXML is bound to make it a non-open standard. Wouldn’t you agree?

    As per the Draft Policy, while ODF is an open standard, OOXML is not. OOXML does not even satisfy the mandatory conditions for openness, leave alone the desirable conditions. Isn’t discussion of ODF and OOXML from the viewpoint of the Draft Policy on Open Standards for eGovernance the entire point of this write up? It fails to do so. Perhaps I misunderstood the purpose of the entry.

    In any case, I thought I’d point out the situation of OOXML vs. ODF vis-à-vis the Draft Policy.


    Where are you getting the “technical advancement” test from? Is it recognized anywhere in India? As per Shamnad Basheer’s most recent entry on the Kolkata stakeholders’ meeting on the Draft Patent Manual, Shri N.N. Prasad made it clear that such interpretation was not the Draft Patent Manual’s job. The same goes for your notions about CIIs.


    Is there any text that you can cite where “without any patent related encumbrances” or “patent encumbrances” has been argued to mean “patent free”? Patents grant an inventor rights beyond royalty for use. Form and method of use also, for instance, can be dictated by the patent holder in the licence. Those are patent related encumbrances.

    AC

    Reply

Leave a Reply

Your email address will not be published.