You are here

Evolution of software patent law

The American Congress passed the Patent Act in 1952, and its mandate was clear and simple enough for those days. Only much later had federal courts to contend with, confront varied interpretations of the intentions of the original Act, and pass judgments thereafter.

Earlier, software by nature was considered as mathematical algorithm, and was thus not patent material. In any case, all proprietary software are automatically covered and protected by copyright not only in national contexts, but also under international agreements such as WTO, and TRIPs. Ever since companies resorted to courtroom decisions for proving the patentability of software or for upholding perceived infringements, the hitherto implicit delineations of patentable and non-patentable subject matter have been continuously eroded.

First was the ruling that software, though a mathematical algorithm, may be patented if it works in connection with a specific apparatus (“Software, Internet and Business Method Patents”). Then came the ruling that patent laws should protect all methods whether or not a useful and tangible result ensued, in one of the most famous cases relating to such patents - the State Street Bank & Trust Co Vs Signature Financial Group. (This was one of the precursors to two wholly new classes of patent classification, namely the Business Method Patent and Software Patent). Thus some feel that there has been a ‘ relaxation of standards used to determine whether an invention qualifies for patent protection'. (“ The Software Patent Experiment”).

In many more cases the judiciary has been instrumental in granting recognition to patents that otherwise would not have qualified under extant laws. Thus the courtroom indirectly created a convenient state of affairs for the stakeholders to find a way to safeguard huge investments. Indeed, the US created the Federal Circuit to deal with patent suits.

The removal of the subject matter clause (exception) gave the fillip that resulted in the boom in numbers of software patents. As a legal feature of computing, software patents can be classed with business method patents, which indeed have to be software-based.

Software patents in other countries

In Europe , software patents have been granted since 1980. The European Patent Convention's (EPC) Article 52 excludes ‘programs for computer' from being patentable and a note below includes the words ‘as such'. This is interpreted to mean that presence of ‘ further technical effect that goes beyond the normal physical interaction between the program and the computer' will make the program patent-worthy.

(“Software patents under the European Patent Convention”, )

Since April 2002, Japan has amended its laws to recognize software patents, and this has resulted in rapid rise in patent numbers, despite opposition to the policy (“Patent Inflation in Japan ”, ). As of October 2005, the Japanese Ministry of Economics, Trade and Industry METI has recognized that abuse of software patent policy exists and efforts are on to stem the ‘harmful effects' of granting such patents. (“METI Study Group Interim Report”, ). A New Year message of the JPO conveys its intentions and efforts: “We aim to attain an eleven month ‘waiting period for patent examination,' the shortest in the world, by 2013 and finally eliminate the waiting period altogether…Toward the early attainment of our vision to become an ‘IP-based Nation,'”

(“New Year Thoughts, 2006”,

India has courageously removed the software patentability clause from the Indian Patents Act in April 2005. The world's anti-software-patent lobby sees this decision as excellent third-world economy sense, and a wonderful precedent for other developing countries to emulate, to obtain fair deals on all IP-related issues from WIPO.