It promised so much. At long last, a South African court ruling on the patentability of computer programs. And a ruling on the patentability of business methods. But in the end the case of Standard Bank and MTN v 3MFuture proved to be something of a damp squib.
The facts were quite simple. An inventor devised and patented a 'Transaction Authorisation System' that was designed to solve the problem of bank card fraud in transactions, particularly but not necessarily where the card holder and the supplier are in different locations. He met with Standard Bank on a number of occasions, seemingly with a view to selling or licensing his technology, but no deal was done. Standard Bank subsequently partnered with MTN to form a company called MTN Mobile, which used technology that the inventor felt infringed his patent. So the inventor, who had transferred his rights to 3MFuture, sued for patent infringement. Standard Bank and MTN defended the action by claiming that the patent was invalid for various reasons, including lack of novelty, obviousness (no inventive step), and the fact that it comprised a computer program and a business method.
For the benefit of those who are new to this, I should point out the South African Patents Act provides that it's not possible to patent a computer program 'as such'. Yet, as we all know, there are many software patents in South Africa. How can this be? Well there's still never been a South African case in which the validity of a software patent has been examined, but the general thinking is that South African courts will follow the approach that's been adopted in USA and Europe. Which is basically this: as long as the software achieves a technical result or solves a technical problem the patent is not for a computer program 'as such' and it is therefore valid. The Patents Act also says that it's not possible to patent a method of doing business 'as such'.
When the case came to court, the inventor abandoned a number of the claims in the patent and in the end, relied on just three. This had the result that the trial court did not have to consider the issues of computer program patents. It did, however, hold that the three remaining claims in the patent were valid because they hadn't been anticipated by any prior art, and they had not been obvious. This meant that the inventor was entitled to an interdict but this did not have to be decided because the functionality had been discontinued. Standard Bank and MTN took this decision on appeal to the Supreme Court of Appeal ('SCA').
The SCA handed down its decision on 22 November 2013. It made it clear right from the outset that this wasn't going to be the big computer software or business method judgment many had been hoping for. Judge Nugent said that, although Standard Bank and MTN were claiming revocation on multiple grounds, the matter could be disposed of by simply dealing with one, namely novelty. He said this: 'I assume for present purposes that the method is indeed an invention, and need only deal with the objection against novelty, because I consider it to be decisive of the appeal, which means lack of inventiveness does not arise.'
The judge went on to explain what novelty is: 'An invention is deemed to be new if it does not form part of the state of the art immediately before the priority date of any claim to the invention. The state of the art includes all matter that has been made available to the public, whether in the Republic or elsewhere, by written description'. He said that, although Standard Bank and MTN had argued that that patent had been anticipated by three earlier US patents, only one of those US patents needed to be considered.
The judge reminded us that the test for anticipation was set out in the famous old case of Gentiruco AG v Firestone SA (Pty) Ltd 1972 (1) SA 589 (A). In that case the court said that the test involves comparing the claim in the patent with the prior art document to determine whether the prior publication describes the same process. The court in the Gentiruco case said this: 'Hence for it to "describe" the invented process etc., it must set forth or recite at least its essential integers in such a way that the same or substantially the same process is identifiable or perceptible and hence made known or the same or substantially the same thing can be made from that description.' The judge also reminded us that the court in the Gentiruco case said that when interpreting the patent specification 'words must be read grammatically and in their ordinary sense... (but)... if it appears that word or expression is used, not in its ordinary sense, but with some special connotation, it must be given that meaning.' And he reminded us that the SCA in the case of Aktiebolaget Hassle v Triomed (Pty) Ltd 2003 (1) SA 155 (SCA) said that the principles of construction are to be applied purposively, 'so as to extract from the language the essence, or the essential elements, or the "pith and marrow" of the invention, thereby achieving the purpose of a patent specification.'
The judge said that 3M's invention sought to overcome fraud in electronic transactions involving bank cards by providing account holders with a way of 'enabling' or 'disabling' the account through a 'remote device' (a cell phone or computer). Which is just what the earlier US patent sought to do, the only difference being that it spoke of 'activation' through a phone rather than 'enabling'.
3M Future had argued that its claim was different from that contained in the US patent because it was confined to the account being 'switched on' by the account holder for a single transaction only, something that did not apply to the US patent. But the judge rejected this, because he wasn't convinced that the use of the word 'selectively' in 3M Future's claim so limited the patent – he suggested that it might have been different if it had used words like 'singly' or 'discretely' or 'one by one'. And he didn't feel that the reference in the claim to 'a transaction' was significant, saying that this was the kind of 'meticulous verbal analysis' that modern construction of patent specifications eschews.
There was therefore no reason to suppose that the inventor had intended the system to be used for simply one transaction at a time: 'In my view the method that is claimed is not confined to authorising only single transactions but extends to authorising multiple transactions at the selection of the account holder.' This had been anticipated by the earlier US patent. Which meant the patent had to be revoked and the appeal upheld.
So another fairly standard decision in which a patent is revoked on the basis that it wasn't new. And still nothing on the issue of software or business method patents. The long wait goes on!