22 March 2015

When Does ‘Novel’ Not Mean ‘New’?

News News NewsOne of the fundamental requirements for an invention to be patentable is that it must be novel.  The word derives ultimately from the Latin novus or novellus, meaning ‘new’.  It has other counterparts in modern Romance languages, including Spanish (novillo), French (nouveau), Italian (novello) and Romanian (nuia).  In English, the word tends to connote something a little more nuanced than merely ‘new’ – perhaps a note of surprise, unexpectedness or particular originality.  It also has the alternative meaning of a long-form work of fiction, contracted from the Italian novella storia.

And then there is the meaning of ‘novel’ within patent law, where it means ‘new’ in the relatively narrow sense of not having been previously publicly known or discovered in essentially the same form subsequently claimed as an invention.

The key word here, however, is ‘publicly’.  Patents can be – and have been – validly granted for inventions that had been known and used by one or more other people prior to the later inventor filing a patent application.  Basically, if an invention exists, but has not been made available to the public, then it remains available for someone else to independently invent and, if they so wish, to patent.  This could happen because the invention has been locked away in the original inventor’s basement and never disclosed or, in a commercial setting, because the company responsible for its original development made a deliberate decision to retain the invention as a trade secret.

So far this may seem quite sensible.  However, consider the following scenario.  A first person imports a product from overseas, and offers it for sale in Australia.  Samples of the product are placed on display in a showroom, where they are seen by many potential buyers, however none are actually sold.  The product has an internal mechanism that cannot be seen during normal use (or display), but which can be readily accessed for inspection or servicing, should anyone wish, or need, to do so.  The product is subsequently withdrawn from sale.

Later on, a second person develops, and files a patent application for, an equivalent product having an internal mechanism with essentially the same features as the earlier unsuccessful imported product.  A patent is granted.

The question, then: is this granted patent valid, or does it lack novelty in light of the earlier imports?  More specifically, is it enough that the imported products were on public display, such that anyone who wished to could have inspected and disassembled them (either before or after purchase), or is it necessary to show that some member of the public actually did inspect and disassemble a sample in order to show that the later patented invention is not novel?

The answer, according to a recent ruling of a Full Bench of the Federal Court of Australia, is that it is necessary to establish, by evidence, that some person was ‘in fact interested in the internal componentry’ or ‘was in fact free to examine the product as he or she pleased’, or ‘actually had the opportunity to ascertain the internal workings’ of the product: Damorgold Pty Ltd v JAI Products Pty Ltd [2015] FCAFC 31.


The recent decision is the result of an appeal from the decision of a single judge of the Federal Court in Damorgold Pty Ltd v JAI Products Pty Ltd [2014] FCA 150.  At issue in the case was Australian Patent No. 760547, owned by Damorgold, and covering ‘a blind control mechanism’.  The patent was originally filed on 25 August 2000, and claimed priority from a provisional application filed on 25 August 1999.  Significantly, at that time a ‘prior use’ of an invention (as opposed to a prior publication in a document) was only relevant to the validity of a patent if it occurred in Australia.

Damorgold claimed that JAI infringed the patent by selling a range of products and components in its roller-blinds range.  Since July 2007, the components that JAI offered for sale, sold and supplied in Australia included various roller blind spring assist mechanisms, or ‘spring assists’.  JAI’s spring assists were used in combination with ‘slide controls’ and ‘blind cylinders’ to form a mechanism for extending and retracting a blind.  JAI admitted that this was the only reasonable use of its spring assists, and that it was aware that customers who purchased the spring assists were likely to use them in combination with slide controls and blind cylinders to make roller blind mechanisms.

Damorgold claimed, and the judge at first instance found, that all but two of the claims of the patent would be infringed by the use of the spring assists to make roller blind mechanisms, and that JAI was liable for that infringement under section 117 of the Patents Act 1990.

However, the judge also found that the majority of Damorgold’s patent claims were invalid as a result of the earlier importation by JAI, in around 1996, of a product called RolaShades.  Much of JAI’s evidence was based on the recollections of persons who were employees at the time, and it therefore had the kind of difficulties you might expect in establishing the precise facts based on memories of events now almost 20 years in the past.  It failed, for example, to establish that there had been any sales of the RolaShades products, despite one witness claiming to have seen invoices (which could not be found and produced).

Factual Findings

The primary judge made a number of factual findings based upon the available evidence:
  1. JAI was in possession of a number of the RolaShades products before the priority date of the Damorgold patent (at [147]);
  2. the RolaShades product was mounted in blinds on display at JAI’s showroom, and was also shown at customers’ places of business (at [151]);
  3. there was, however, no evidence that any of the RolaShades products were ever sold, and the samples were all retained by JAI and not left with customers (at [151]);
  4. is was important to appreciate that potential customers were entities who made blinds (i.e. trade, not retail, consumers), and had an interest in the working of the product (at [153]); and
  5. the RolaShades product could be easily disassembled and reassembled, should someone wish to do so – this was demonstrated in court by one of the witnesses (at [152]).

The Outcome Before The Primary Judge

In summary, the primary judge found (at [154]) that:

In my view, the potential customers … would have been free to disassemble [the product], and would have been interested in looking at its mechanical workings. The acts of … displaying the RolaShades product to various customers, with a view to sale, was making the RolaShades product publicly available. There was a means for any customer to inspect the RolaShades product. The customers had an incentive to inspect and [JAI’s representative] had an incentive to allow the customers to do so. Each customer had the opportunity to handle, view, and test the RolaShades product. An examination of the component features would have been easy, and would have revealed all the relevant integers of the claimed invention. Such an examination would have readily enabled a person skilled in the art to put the claimed invention into practice.

Furthermore, he found (at [157]) that:

It does not matter that only a few, or perhaps only one, potential customer was approached and had the opportunity to examine the RolaShades product. It does not matter that the RolaShades product was not actually dis-assembled. Each potential customer, in my view, had the opportunity of a complete examination of the internal componentry of the RolaShades product. There was no restriction placed upon any potential customer as to what he or she would do in making use of the information so obtained from an examination of the internal componentry.

There is good authority for this position.  The courts have repeatedly emphasised that the requirement for information to be ‘publicly available’ is satisfied in circumstances where at least one member of the public has access to the information, and is free (e.g. not restrained by an express or implied obligation of confidentiality) in law and equity to make use of it (see, e.g., Insta Image Pty Ltd v KD Kanopy Australasia Pty Ltd [2008] FCAFC 139, and cases cited therein, going back to 1887 in Humpherson v Syer 4 RPC 407).

It must be said, however, that the question of what information is actually made available to the public is one of fact, and will be dependent upon the evidence in each particular case.  The kinds of issues considered by the primary judge – how easily accessible was the information, how much interest did the various parties have in ascertaining and sharing that information, and so forth – will generally be critical to the outcome.

Damorgold appealed the finding that its claims were not novel in view of the RolaShades product to the Full Court.  Notably, JAI did not appear to defend the appeal, and so the court had the benefit of submissions from only one side.  The appeal court reversed the decision of the primary judge, by a 2-1 majority (Bennett and Yates JJ, with Jessup J dissenting).

The Majority Decision

The majority appears to have placed great weight on what the evidence established had actually happened, as opposed to the extent of access that had been made potentially available to JAI’s customers.  As Bennett J stated, at [58]:

Accepting that the customer was a person of skill in the art, which is not in dispute, it is necessary, to establish lack of novelty, that the evidence established that the disclosure of the RolaShades product enabled the person to understand the invention by disclosing the essential integers of it. Mere observation of the way that it worked did not provide such an enabling disclosure. There was no evidence that any customer was in fact interested in the internal componentry of the RolaShades product, or that any customer was in fact free to examine the product as he or she pleased, or that any customer actually had the opportunity to ascertain the internal workings of the RolaShades product. (Emphasis added.)

Similarly, Yates J stated, in concurrence at [95]:

The primary judge placed determinative significance on the fact that potential customers had the opportunity of a complete examination of the internal componentry of the RolaShades product and could have disassembled the product had they wanted to do so in order to carry out such an examination. Those are, undoubtedly, important findings. But the evidence went further and enabled the primary judge to conclude that, in fact, no potential customer disassembled the product and no such complete examination took place. In other words, there can be no doubt that the features of the RolaShades product that could only be discerned upon disassembly of the product were not disclosed to customers. Thus, in my view, the evidence positively established that all the essential features of the invention as claimed were not disclosed (i.e. were not made publicly available) by the acts relied upon. (Emphasis added.)

The Dissent

In my view, the majority here showed a lack of deference to the primary judge, who was the only one of the four to have had first-hand access to the original evidence and cross examination of the witnesses (including the demonstration of disassembly and reassembly of the RolaShades product).  The comments of Jessup J (at [68]) in dissent are, to my mind, apposite:

The primary Judge’s ultimate finding of fact on this aspect of the case was that “the potential customers … would have been free to dis-assemble [the product], and would have been interested in looking at its mechanical workings”. … it has not been demonstrated that this finding was in error. It is true that the finding was not of a primary fact, in the sense of relating only to something that either happened or did not happen. But his Honour’s use of the term “been free” involved, in context, more than a statement of what would have been legally permissible: it involved a conclusion about the practicalities of situations in which the RolaShades product was shown by [JAI’s representative] to his potential customers. It was entirely a matter of choice for the potential customer himself or herself whether to have the product disassembled.

Was The Case Wrongly Decided?

With the greatest of respect to the estimable Justices Bennett and Yates, I think this case has been wrongly decided, for (at least) three reasons.

First, the primary judge made specific factual findings upon which his conclusion was based.  There was no suggestion that he had identified or applied the wrong legal principles.  The majority of the Full Court simply decided that they required greater certainty as to whether or not the internal workings of the RolaShades product had been made available to the public.  It seems to me to be absurd that someone could make a product available for sale, and place it on display without any specific physical security or legal restraint to prevent close inspection, and yet this would not make the workings of the product publicly available in the absence of evidence that someone did, in fact, either purchase or closely inspect the product.

Second, it seems to me that JAI employees who had access to full details of the assembly and operation of the RolaShades product should, in this case, be regarded as members of the public, to whom the product was made available.  Many of the earlier cases – including the Full Court’s own relatively recent decision in Insta Image – involved use of an invention by the inventor prior to filing a patent application.  By definition, an inventor is not a member of the public for the purposes of determining whether an invention is made publicly available.  The inventor has a choice as to whether, and when, to make a public disclosure.  The inventor equally has the right to maintain the details of an invention in confidence for as long as she or he wishes to do so.

That was not the situation in this case.  The RolaShades product was already available elsewhere in the world.  It was somebody else’s invention/design.  Samples were purchased by JAI on the open market, and imported into Australia.  There was no suggestion that the supplier was making any effort to maintain confidentiality, or to restrict distribution of the product.  While JAI and its employees had the option of limiting access to the samples of the product in their possession, they were under no obligation to do so.  Other members of the Australian public were equally able to acquire and import the RolaShades from the same, or different, overseas sources.  To put it another way, JAI and its employees had no proprietary right to protect in the design of the RolaShades product.  They were, in all relevant respects, members of the Australian public to whom full details of the internal workings of the product had been made available.

Third, the outcome simply seems absurd.  A product that was commercially available overseas was imported into Australia and offered for sale.  Although there is no evidence that any were sold at that time, the fact is that the product was here, was promoted, and was available for purchase by the public at large.  A few years later, somebody filed an application for a patent on the same arrangement.  Despite the fact that the product was, by then, not new to Australia – indeed, not even newly-available to the Australian public – the court has found that the resulting patent claims are novel.

Conclusion – A Few Closing Thoughts

Perhaps the outcome in this case was a result, at least in part, of the fact that JAI did not appear to defend the appeal.  There was, therefore, no representative in court to put an alternative argument to that presented by Damorgold.  However, this did not prevent one of the three judges from dismissing the appeal.

Unfortunately, since JAI appears to have withdrawn from the proceedings, it is unlikely that there will be an application for special leave to appeal the decision to the High Court.  This is a shame, because the decision certainly appears to raise the bar on the nature of evidence required to establish anticipation by prior use, potentially requiring proof of detailed events involving specific individuals, which may have occurred many years or even decades prior!

On the other hand, the law has since changed in Australia such that, for applications filed on or after 1 April 2002, prior use includes public disclosure by acts occurring anywhere in the world, and not just within Australia.  This means that sales or close inspections of products occurring overseas would now normally constitute a disclosure to the member of the public.  Even so, Damorgold may raise the standard of proof required to establish that overseas use resulted in a sufficient disclosure to the public.


Mark Summerfield said...

It is interesting to see that the courts are moving away from strict application of the reverse infringement test. I know here the court was asked to determine if the display of the blind was part of the prior art base at all, but as you've presented it here even the importing (within the definition of exploit in Schedule 1) of that product into Australia by JAI would have been a reverse infringement of the claims. However, as you say, for post 2002 patents, this is no longer an issue...

Mark Summerfield said...

I do not believe that the reverse infringement test has ever strictly applied to prior use. In General Tire & Rubber Company v Firestone Tyre & Rubber Company Ltd [1972] RPC 457 (from which the test is generally regarded as arising) the Court of Appeal was concerned with a prior publication. The court was looking to address the problem that the patentee (in the patent specification) and a prior inventor (in an earlier publication) might have 'so described their devices that it cannot be immediately discerned from a reading of the language which they have respectively used that they have discovered in truth the same device' (at 485). This led the court to the notion of working out whether following the instructions in the prior publication would inevitably lead to the production of something falling within the scope of the later claims.

The Australian Patents Act 1990 sets out the prior art base in explicit terms, with the definition in the Dictionary expressly referring to 'information made publicly available through doing an act'. So the mere 'doing' of the act is not sufficient to deprive a claim of novelty, whether or not it might constitute a form of 'pre-infringement'. It is necessary to determine what information was disclosed as a result.

This decision certainly seems to carry this requirement to its logical and most literal conclusion! It is notable, however, that in Insta Image the Full Court accepted that, at least in the circumstances of that case, the inquiry concerning novelty was the same under the 1990 Act as it would have been under the 1952 Act, and there are plenty of older cases in which evidence of actual inspection/disclosure was not required to establish anticipation.

I am not sure that the law can truly be said to be settled at this stage, which is why it would have been good for the High Court to have a look at the issue.

Post a Comment

Copyright © 2014
Creative Commons License
The Patentology Blog by Dr Mark A Summerfield is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Australia License.