Guidelines do not clarify software patent question say submissions

Debate rages on Section 15 (3A) of Patents Bill

Most submissions on the guidelines formulated by the Intellectual Property Office of New Zealand (IPONZ) on patentability of software suggest the guidelines have not achieved the desired clarification of proposed legislation and may have served further to muddy the question of what software is or is not patentable.

Section 15 (3A) of the Patents Bill – currently awaiting its second reading in Parliament – simply states “a computer program is not a patentable invention”. This clause was inserted by the Commerce Select Committee, on the strength of a number of submissions.

Further debate has seen the Ministry of Economic Development suggest that the committee did not intend to exclude “embedded” software – that is, in broad terms, software that controls a machine, achieving a physical effect outside the realm of pure computation.

The IPONZ guidelines attempt to achieve something like that qualification, while studiously avoiding using the term “embedded”. Several submissions, such as those from Fisher & Paykel Appliances and from John Rankin of Affinity and Don Christie of Catalyst IT, recommend putting the term explicitly in the guidelines, if not in the Bill itself, and either setting ICT industry representatives to work towards a definition of the term or taking on board existing definitions.

The F&P submission suggests “there is no great difficulty” in defining what “embedded” means, and points out that there is a definition in Wikipedia.

However, this, like many Wikipedia articles, is disputed; Wikipedia editors have appended notes reading ‘clarification needed” and “dubious – discuss”.

The IPONZ guidelines themselves point out exceptions such as the recording of a discovery in print. While that constitutes a physical effect it is not one that is ‘central” to the discovery and hence does not of itself render the discovery patentable, they say.

A submission from the Airways Corporation refers to its air-traffic control software as having the undoubted physical effect of avoiding mid-air collisions; yet since it achieves that effect by communicating and displaying data, it is not embedded software within many accepted meanings of the term, particularly those that exclude any software that runs on a general-purpose computer.

“We are concerned that vital air-traffic systems would not be patentable under the current guidelines,” Airways says.

Several submissions contend that the IPONZ guidelines will have no real weight in law and will in due course be superseded by a literal interpretation of Section 15(3A) or by case law formulated by judges.

The guidelines rely heavily on a test (called the Aerotel test) drawn from UK law, but some submissions say since the Bill and guidelines are not elsewhere consistent with UK law, this creates confusion. The International Federation of Intellectual Property Attorneys contends that the Aerotel test has been outdated by subsequent UK case-law. By relying on the test, “New Zealand risks isolation in the worldwide intellectual property community,” the Federation says.

A number of submitters, for example Geological and Nuclear Sciences, suggest that NZ should not stand out on its own but harmonise its law more clearly with that of Australia, the US or Europe, where our major markets lie.

Submissions from entrenched champions of software patent such as Microsoft and NZICT, as well as some from more equivocal sources, suggest the best course might be to scrap s15(3A) altogether.

Commerce Minster Simon Power, announcing the intention to formulate the guidelines, in July last year, saw them as making it possible for the Patents Bill to proceed through its remaining stages independently of debate on the guidelines themselves. Indeed, he expected the Bill to pass before the guidelines were finalised.

Tags Patents Bill


Reed Wade


"A submission from the Airways Corporation refers to its air-traffic control software as having the undoubted physical effect of avoiding mid-air collisions; yet since it achieves that effect by communicating and displaying data, it is not embedded software within many accepted meanings of the term, particularly those that exclude any software that runs on a general-purpose computer."

I might rephrase that as "not embedded software within _any_ accepted meanings of the term".

Otherwise, for example, all banking software counts as embedded as there is a physical effect of me being able to eat lunch when it works correctly.

lovely try though



Hi Dave,

With all due respect, I don't agree that the proposed law change is clear. You will remember that I put up a blog post on this topic some time ago (see ). The questions raised there remain unanswered. Even the NZOSS/InternetNZ submission found the guidelines "hard to understand".

A further illustration of the difficulty of implementing the proposed law change comes from another NZOSS submission, which explains that "the idea that software embedded with hardware is somehow different to other forms of software does not bear scrutiny" (see ).

One point that does seem to be clear is that the Select Committee wanted to allow patents for inventions involving embedded software. "It also considered that companies investing in inventions involving 'embedded' computer programs should be able to obtain patent protection for these inventions. The committee and the Minister accept this position." (See )

In relation to ad hominem arguments that may be put forward, I would only encourage readers to play the ball, not the man.

Dave Lane



I think it's important to note that despite clamour from current monopolists like Microsoft and the NZICT (who primarily represent multinational IT firms and their NZ partners, who have a vested interest in Microsoft's continued market dominance), there is no question about the position of government on this issue - the goal is to exclude software from patentability.

Based on a broad range of submissions from both proprietary and open source-focused firms (as well as lawyers, who almost unanimously favoured keeping software patents, but who don't contribute to software innovation), the Commerce Commission Select Committee made a *unanimous* recommendation to the minister to legislate excluding software patents. It can't get any clearer than that.

Lawyers for Microsoft and others either did (or neglected to) make their arguments for retaining software patents, and these were clearly not seen by the committee as compelling.

The guidelines for making the decision on what is or isn't patenable on a practical level - though important - do not in any way challenge the underlying principle that software is not - and should not be seen as - a patentable invention. It is covered by copyright. Full stop.

Don Christie



For completeness sake it would have been relevant for the article to point to the NZOSS position:

and also the NZCS commentary on why they did not think it necessary to submit in this topic. The MED was not consulting on whether software patents should be allowed. The evidence for that has been reviewed and the conclusion reached.

John Rankin


It is not quite correct to say 'that the committee did not intend to exclude "embedded" software' and nor do I believe MED is suggesting this. The select committee worded its report very carefully, as you would expect. The report in fact states quite clearly that inventions containing embedded software will continue to be patentable subject matter. As an anonymous commenter has pointed out, it is the invention that is tested for patentability; the embedded software, just like other software, is not a patentable invention. My reading of the committee's report is that it is a "for avoidance of doubt" statement, intended to address a legitimate question about the impact of section 15 (3A) on this kind of invention.

So the intent of the guideline is twofold. First, to avoid inadvertently excluding from patentability inventions which happen to contain embedded software. Second, to stop claimants circumventing section 15 (3A) by disguising an attempt to patent a computer program as a claim for an invention containing embedded software. The extent to which it achieves these aims is likely to be a source of ongoing dialogue.

The select committee report raises an important issue and we ought to focus on how well the draft guideline addresses it. As other commenters have pointed out, the committee heard all the evidence from all sides and unanimously concluded that software and patents need to get a divorce.

Comments are now closed

Google's Nexus 9 tablet, Nexus 6 'phablet' first with Android 5.0