Patent law insights and commentary

Bruce Jones, Partner

20.10.2026

Still not over Yeda (but I’m blaming Actavis)

Remember Generics (U.K.) Limited & Ors v Yeda Research and Development Company Limited & Ors?  It was, to the best of my knowledge, the first time that it was ever explicitly acknowledged that we now had a different test for novelty and for infringement based on the same claim (I know Mobil kind of perverted novelty a little but that ‘perversion’ was at least expressed explicitly in the claim and applied symmetrically).  At which point, RIP the squeeze defence.  Why?  Well (as a quick recap)… in temporal order (as they don’t now say on telly):

• Actavis said, in the right circumstances, a claim covered not just what it was properly interpreted to say but equivalents  (the key “circumstance” here being that the equivalents were – forgive the loose terminology – obvious); and
• Yeda then went on to find that it was no longer the law that a claim lacked novelty if the prior publication disclosed subject-matter  which, if performed, would necessarily infringe the claim (say, because it was only within the scope of ‘equivalents’ to the claimed subject-matter).

It’s only right and proper to point out that Arnold J noted in his judgment that the Supreme Court had not addressed the impact of its conclusions in Actavis upon the issue of novelty.  That’s on you Lord Neuberger.

It’s been nearly ten years and it still seems like a repugnant development to me.  BUT – for as long as we are stuck with Actavis – it’s useful to consider what would be the impact of the alternative? By which I mean that of novelty being regarded as including within its scope obvious equivalents?  You could argue, say, that we shouldn’t get too precious about preserving the test of novelty and perverting it to encompass some degree of obviousness.  After all, I will always be of the view that the CJ EU in the Interflora v M&S judgment introduced a ‘confusion’ element into what should  have been a simple and pure double identity test for TM infringement.  So why not in patents too?  Well…  Imagine the burden upon the patent office of examining this stuff.  They would have to establish what obvious equivalents were and then encompass them within an examination of the test of novelty.  And all of that extra work when 99/100 applications (I’m making up the numbers here) are never the subject of a dispute.  If, as we must, accept that we are lumbered with Actavis, the final destination of this train of thought brings me back again to Gwilym Roberts’s point (I’ll stop now, Gwilym, honest): we have to let the Patent Office (of whichever stripe) grant patents on a ‘good enough’ basis.  And in this instance,  and Actavis notwithstanding, that means preserving the purity of the novelty test.

Bruce Jones, Partner

14.10.2026

Could Improver have been improved?

Following on from the seemingly endless G1/24 and G1/25 farrago, and more particularly the issue of whether a specification is ‘good enough’ to be granted (rather than this apparent, and by now almost philosophical, quest for perfection on which the EPO seems to be engaged), I started to reflect on instances in which patent law explicitly accepts imperfections that are baked into the system.

One of these is the little-considered detail in the old Improver  test which is thrown up by the second of the three questions:

Would [the fact that a variant had no material effect] have been obvious at the date of publication of the patent to a reader skilled in the art?

First of all, why did Lord Hoffman refer to the date of publication of the patent rather than the application?  I suppose one reason might be that only once the patent is granted do you have the claims upon which the issue of infringement will be determined.  But, on the other hand, once the issue of whether the ‘materiality’ of the ‘variant’ has been framed, there is no reason why the date of first publication couldn’t be used instead.  Re-reading the judgment, I think the answer is much simpler: this is the date used by Lord Diplock in Catnic and Lord Hoffman was further developing that authority.  Why did Catnic use that date?  The following phrase from Lord Diplock is a clue:

“Where it is not obvious, in the light of then-existing knowledge, the reader is entitled to assume that the patentee thought at the time of the specification that he had good reason for limiting his monopoly so strictly and had intended to do so”

So it appears to be based upon the theory that the date of publication – which under the 1949 was in effect grant – is a date linked to the framing of the claim that is granted and, because of that, this is the date upon which the ‘intention’ of the owner in framing it in that manner is then judged against what then became the second Improver question.

The impact of all of this arises when we consider what result changes to that date have.  Simply put – and as a general matter – the later the date, the more favourable the position is likely to be for the patentee.  Imagine that you have written a claim that, because of the strictures of A123, you cannot then change to have the scope you wish it to have.  If the second Improver question was one that was posed at the date of filing, you would be unable to do anything.  But because it’s a test to be applied at grant, as a patentee you have several years in which you can import into the state of the art all manner of material that means the answer to the second Improver question can then be answered more favourably for you.  Seems to me the only genuinely fair way to frame that question would be to change the date to the date of filing.

Bruce Jones, Partner 

17.06.2026

Claim Interpretation at the EPO: Why G1/24 and G1/26 Matter

Fractal-isation of claim interpretation at the EPO

This is a very nerdy topic even for a very nerdy corner of the patent-verse.  But….it’s just too captivating to not watch.  The EPO now appears to be on a path of eating itself and its resources through a process of fractal-esque referrals of questions to the Enlarged Board of Appeal.  The issue?   The EPO has referred back to itself a question of interpretation arising from….a decision on interpretation: G1/26 concerning G1/24.

Three reasons the EPO has reached this point

First, because the EPO doesn’t have the benefit of s125 (so it ends up wondering if it had to ‘choose’ between a69 and a84);

secondly, because it doesn’t have to deal with infringement; and

thirdly and possibly most critically, because, unlike under common law, it feels it has to provide a comprehensive quasi codification of the law rather than allowing it to evolve as disputes are resolved.

The risk of endless referrals on patent claim interpretation

Thus, we end up with a decision of the Enlarged Board  – surely not in any way revelatory – that the specification should “always be consulted to interpret the claims” (whoever drafted s125 take a bow). But does this clarify things?  Apparently not. In T 0873/24 The Technical Board of Appeal has decided to refer questions to the Enlarged Board of Appeal on the application of G1/24 to the assessment of added matter.  So the answer in G1/24 – supposedly an answer to a question of universal applicability – is now the subject of the same forensic focus as was the question before the Enlarged Board in G1/24.  Or to put it differently, we now have a different set of words to pore over, pick at and agonise as to their real meaning.  How long before the key question becomes whether G1/26 (if accepted by the EBA) simply produces a decision that itself spawns more referrals as to its meaning…..until the EPO’s resources are endlessly taken up interpreting its own words, spoken in the course of interpreting other words of its own?  An alternative would be to allow the law to evolve organically rather than trying to frame broad conceptual propositions.  The words of Birss LJ in his DABUS judgment when confronted with a somewhat abstract argument are hard to improve on:  “That does seem odd, but it is something I prefer to decide in a case in which it actually matters”

Bruce Jones, Partner 

09.06.2026

Why careful patent drafting matters more than ever…

My embody lies…

With the EPO tying itself into what, to this writer, seems like entirely unnecessary knots over claim interpretation and practice in relation to amendment of the specification, now is a good time to pause and remember that many of the potential pitfalls arising from this can be avoided with careful patent drafting from the outset. Within the UK profession at least, it is often seen as a very ‘American’ thing to place so much focus on describing examples of the invention.

Why embodiments matter in patent drafting

More attention often goes on carefully creating what are, in essence, hierarchical introductory clauses that recite language imagined might, at some future point, assist in providing flexibility for claim amendments. This difference in focus is probably also responsible for the different way that part of a patent specification is described. In the UK, we refer to it as the ‘specific description’ – a term that almost seems to relegate its significance within the overall description of the specification. By contrast, our American colleagues refer to embodiments. This difference has always carried weight for me. A carefully and, critically, thoughtfully crafted set of embodiments provides an indispensable reservoir of genuine technical material upon which to draw. Actual, explicitly described alternatives are often the best way to demonstrate that claimed width is both supported and within the content of the application as filed. What does this mean for clients? Put simply, you may need to spend a little more time developing alternatives that can be included, and a little more money having the specification drafted to include them. But those costs will be entirely dwarfed by the costs later down the line of having to prepare an extensive argument for permissible claim amendments instead.