From October 2016 to March 2017 the team is joined by Guest Kats Rosie Burbidge and Eibhlin Vardy, and by InternKats Verónica Rodríguez Arguijo, Tian Lu and Hayleigh Bosher.

Wednesday, 19 November 2014

Lantana's lament: no technical contribution, no patent

Technical contribution? Dream on ...!
It may not be Alice v CLS Bank International all over again, but Lantana Ltd v The Comptroller General of Patents, Design and Trade Marks [2014] EWCA Civ 1463,  a Court of Appeal for England and Wales from 13 November, shows that when it comes to disqualifying computer software from patentability [or patent eligibility, as the Americans prefer to call it] on the basis that it doesn't disclose an invention, the Brits can do the job at least as elegantly as the Yanks. The court, consisting of Lord Justice Kitchin together with Lady Justices Arden (who delivered the main judgment) and Gloster, also had a chance to revisit the Court of Appeal's earlier decision in the combined appeals in Aerotel/Macrossan -- and its three-or-is-it-four-step test of excluded subject matter, a test motivated by the alarm bells set off by the UK Intellectual Property Office when the number of hearings concerning patentability of computer programs had risen from just a couple a year to around four a week.

The following note on Lantana is brought to you via the good services of our good friend and occasional guest contributor Paul England (Taylor Wessing LLP), who writes as follows
Data transfer method makes no technical contribution

The Court of Appeal has held the invention disclosed in a patent application by Lantana Limited to be excluded from patentability, as a computer program “as such”.

Lantana’s application was for a method of extracting and transferring data between two computers, involving the computers being linked up to the internet and one making a request to the other for a file.  The local computer has a list of the documents on the other computer. The user of the local computer sends an email message to the other which automatically responds by sending a message to the local computer with the file attached. The virtue of the invention, Lantana claimed, is that it saves the need for continuous connection to the remote computer and the risks of being hacked while connected, or losing the connection. But does this make it patentable?

The appeal focused on section 1(2) of the UK's Patents Act 1977 [= European Patent Convention Art.52], which excludes patents for computer programs unless the invention goes beyond the computer program “as such”. Accordingly, following the four-step test in Aerotel Ltd v Telco Holdings Ltd; Macrossan’s Patent Application [2006] EWCA Civ 1371, if the invention makes a contribution that is technical in nature, then the invention may be patentable, subject to other objections to patentability – novelty, inventive step etc. 
While there is no comprehensive test for identifying a technical contribution, the courts have established a number of “signposts” to assist (AT&T Knowledge Ventures’ Application [2009] EWHC 343 (Pat), [2009] FSR 19 (as refined)). However, both the Hearing Officer of the UKIPO and Birss J, in the Patents Court, held that these did not assist Lantana in this case; there was no identifiable effect outside the computer program itself, such as making a better computer. As the Hearing Officer said: 
any increase in the reliability of the data retrieval process is solely due to the use of e-mail rather than any other communication method. There is no evidence of either of the two computers or the network being intrinsically more reliable themselves” 
and that, as regards the connectivity problem, 
“The contribution cannot be said to solve the problems identified. Rather it circumvents the problems of maintaining a good continuous connection by simply not using a continuous connection. 
Birss J concluded simply that “In substance the claim relates to computer software running on conventional computers connected by a conventional network”. The Court of Appeal has agreed, holding that the invention is excluded matter, having no relevant technical effect.
What is a lantana? Click here to find out
Lantana the movie here
Latest Australian jurisprudence on the same issue in Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150, noted on Patentology here.

21 comments:

Anonymous said...

Accepting its only a precis, all the way to the CoA for that???

Jeremy said...

Anonymous @09:36

Funny you should say that. When I asked Paul if he could write a note for us, I said I was surprised the CoA had agreed to take the appeal.

Paul said...

I was surprised last year when they even went to the high court for it.

Freddie Noble said...

There is a brief report of the CoA granting permission to appeal on page 214, CIPA journal, April 2014.

I am not surprised by the result in this case, but I am not especially impressed by some of the reasoning in the Judgment either.

E.g. paragraph 19: "As the "signposts" show, there must be some identifiable inventive step outside the computer program. That means that, contrary to Lantana's case, the features of the invention must be disaggregated to determine whether they fall within or without excluded matter."

This sounds an awful lot like the "contribution approach" which has been expressly rejected both by the Court of Appeal and by the EPO.

Another thing, although I don't think it mattered a great deal, there is some discussion of whether or not it was accurate to say that in the alleged invention, one computer "controlled" another. The reason for rejecting this definition (para 49) seems to be that "That computer would presumably not operate if, for example, the local computer sent an email to it requesting a particular file at a time when it (the remote computer) was disconnected from the internet".

Of course the remote computer does not respond if it is not connected to the internet (or if it is turned off, on fire, etc), but in that case it does not fall within the scope of the claim. Presumably very many things which are patentable inventions do not work if they are turned off or disconnected.

I am also slightly surprised that the UK IPO did not push an inventive step objection. The world is full of email robots which process commands and automatically reply with results of some computation or some requested resource.

I accept that it is easy for me to criticise having spent all of 15 minutes looking at the case... And yes I think the case is unpatentable and refusal was right... but this is a complex, uncertain and controversial area of law and what that means is that it is vital that applications are refused for reasons which are correct in law, consistent, and well articulated.

Anonymous said...

Perhaps the driving force behind the decision to appeal was to try and bring UK case law into line with the approach of the EPO? There, "technical effect" is read more broadly such that the difficulty nowadays is trying to persuade the divisions that the invention is non-obvious once the "non-technical" claim features have been disregarded.

MaxDrei said...

In the judgement I do not see the patent number. I was thinking of looking at other members of the patent family. How has Lantana fared, in other jurisdictions? Is there in the Family a clue as to why they slogged on, all the way to EWCA?

Anonymous said...

http://worldwide.espacenet.com/publicationDetails/inpadocPatentFamily?CC=GB&NR=2470521A&KC=A&FT=D&ND=3&date=20101124&DB=EPODOC&locale=en_EP

Freddie Noble said...

MaxDrei,

The application number is GB1014714.8. Publication no. GB2470521 and espacenet confirms that this was a PCT with corresponding US, CA and AU cases and US priority.

Darren Smyth said...

Max, the application is GB 1014714.8, which corresponds to GB2470521. There is a PCT WO2009105274 but national equivalents according to Espacenet only in US, CA and AU. Does that help?

Mark Summerfield said...

Perhaps if the UKIPO had done its job, there would have been no need to consume so much of the courts' time and trouble?

http://enaweb.googlecode.com/svn/cgi-bin/postie_readme.txt

Search within the document for "fetch filename".

I doubt very much that this is the oldest relevant prior art. I am pretty sure I recall such things from my very earliest days online. Indeed, before the internet came to Australia (i.e. as "recently" as the early 1990s) we had a 'store and forward' network called ACSnet that necessarily employed a "fetch" protocol of this type for file retrieval, because there were no permanent connections.

Anyway, my thanks to the IPKat for the link to my post on the Australian Research Affiliates case!

MaxDrei said...

I'm not a specialist in EE but for what it's worth, I think I can see what was driving Keith Beresford to pursue this to the bitter end.

It strikes me that the claimed subject matter was new and cute. I am not minded to dispute the UKIPO finding that it was not old and not obvious.

Further, it strikes me as having "technical character" and that the claimed matter uses technical means to solve a technical problem, the very definition of matter fit for patenting, and more than a program "as such".

Why then unpatentable? Binding Precedent?

The HO found himself with no choice. Each instance of appeal found insufficient reason to reverse.

The Decision is replete with expressions of regret. Expressions like "I find myself forced to conclude..". I read in to These expressions that the HO decided with an uneasy heart.

It is a bad day when the law decides something against one's sense of fairness and commonsense. That the line has been drawn, and that's it is OK when the line is perfectly placed, but not otherwise.

I think the EPO does it better. First, when doing Art 52 eligibility, it does not have to investigate whether, on the date of the claim, computers were fetching files from each other or instructing each other by email. OK, that's the UK way, but isn't it putting the cart before the horse?

Second, the EPO is not constrained by Binding Precedent, from reaching a commonsense decision on each set of facts before it.

I guess that Lantana had bigger fish to fry, and that is why it pushed this all the way to a Final Decision.

Freddie Noble said...

Max, you are "not minded to dispute the UKIPO finding that it was not old and not obvious". Is that because you think that the prior art cited by Mark Summerfield is not relevant for some reason, or because you just do not want to get mixed up in inventive step when the case is about eligibility? Or for some other reason?

I think it's a jump to conclude that the hearing officer decided the case with an "uneasy heart". The language in the decision may have been chosen for all sorts of reasons. I don't have any more insight into the Hearing Officer's mind than you do, but I would note that Applicants sometimes think that Examiners and hearing officers have a personal vendetta against their case. For that reason hearing officers tend to bend over backwards to justify their decisions and to point out that they are bound by precedent. The message is "it's not personal", not "I personally disagree".

I do not agree that binding precedent in this case prevented a "commonsense decision". You may disagree with this refusal and that is fine, but this case is a long way from being "obviously patentable to any right-thinking person". Binding precedent does not prevent cases from turning on their own facts - and indeed trying to draw technical parallels with prior cases is generally not successful on either side. s. 1 Patentability is highly fact-dependent (and yes, prior art matters too in the UK). The Aerotel / Macrossan appeal is important case law because of the test set out in it, but a technical comparison of the thing you want to now patent with Aerotel's "special exchange" is nearly pointless because their patent was filed in the 1980's, and there are many things which were not obvious then which are obvious now.

On the matter of whether prior art matters, as you say, in the UK it does. Whether it *should* is certainly contentious. Many people tend to prefer the EPO approach because it is easier to apply. However, that doesn't make it right or an intellectually honest interpretation of the statute / treaty. The judgment in Aerotel / Macrossan makes a compelling argument that the EPO are rewriting history in their interpretation of this law.

My comment on the lack of an inventive step objection (and Mark Summerfield's link does indeed seem to be relevant prior art) came from a feeling that the UK IPO must know that a patentability objection has the potential to open a very large can of worms, and if I were at the IPO thinking about refusing an application under section 1 I would try especially hard to justify refusal on grounds of lack of novelty or inventive step first, to avoid all those worms. Of course, the UK IPO may not care about worms.

Anonymous said...

Is it worth bothering to look for an electronic copy of a UK patent?

MaxDrei said...

Thanks Frddie. It does indeed seem that the Summerfield art blows away the claim. Be that as it may, the UKIPO decision was given in ignorance of it and it suits my argument to run with the facts as they were on the UKIPO table. With that prior art landscape ie with a novel and not obvious claim, I find it hard to swallow the UK methodology on excluded matter. That the EPO way is "intellectually dishonest" but the UK way not is something I don't accept.

Incidentally, I like the US distinction between "patentability" over the prior art and "eligibility" relative to Art 52 EPC issues. With this terminological distinction, we know what we are debating. Many in the EPO use the shorthand "patentability" as shorthand for "Art 54 + Art 56, but not Art 52 EPC" That's fine with me.

Mark Summerfield said...

The corresponding Australian application was abandoned (applicant did not even request examination, when directed to do so), as was the US application, after a final Office Action in which the examiner cited a combination of three prior art documents in order to find a broader claim than the one at issue in the UK to be obvious!

My point is not really about whether or not the claims cover eligible subject matter. My point is more that the UKIPO appears to have relied on the PCT search for prior art, and that no examiner seems to have conducted a search beyond the patent literature that would have shown that the entire concept of an email-based file retrieval process was old by at least the year 2000 (and, as I have suggested, was probably old by 1990, although some more diligent searching would be required to establish this). It took me only about two minutes to find a relevant document via Google.

Presumably the applicant, and his patent attorney, were unaware that he had essentially reinvented the wheel, and not one examiner who looked at the case did any better job of finding the most relevant prior art. Had the applicant at any stage been made aware of this, he may well have elected to abandon the application at an earlier stage. Certainly, it is hard to imagine why anyone would proceed all the way to the Court of Appeal on the subject matter issue if there were no realistic prospect of also overcoming the prior art.

All this matters because a significant source of opposition to 'software patents' is a perception that 'bad' patents are routinely granted for 'old' ideas. Sometimes this criticism is unjustified, because the critics rarely bother to read the claim and find out what the patent actually covers. But sometimes the criticism is entirely valid, and the fact that the quality of search and examination is often inadequate results in greater reliance on eligibility grounds to eliminate those claims that are perceived to be overly broad, or otherwise suspect.

This is bad in principle. Eligibility grounds should not be doing work that is more properly the domain of prior art grounds. This only leads to the kind of sloppy and inconsistent reasoning we see in the US Alice and Ultramercial cases, and the Australian Research Affiliates case, in which judges are deciding for themselves what is 'old' and 'new', 'abstract' and 'concrete', without the benefit (or burden) of actual evidence. The result is that the law is left in a sorry and uncertain state, and confidence in the patent system generally suffers.

Anonymous said...

Some would say that the objective IS to leave the patent system in a state of suffering - that NO patents (on anything) is the true aim.

MaxDrei said...

Thanks Mark. So is the EPO doing it better, with its weak, and prior art independent eligibility filter?

Or is its eligibility approach "intellectually dishonest", as England asserts?

Anonymous said...

Anon at 23:25

If you are correct, I think the firm of patent attorneys we use may be part of the conspiracy.

Freddie Noble said...

Paragraph 4 of Mark's post above is my point exactly - but made more clearly and elegantly than I did.

Max, perhaps when there is a case about something which really is novel and non-obvious (but said to be excluded matter) we can have a proper debate about "eligibility" as you put it. I think it's going to be hard to do that based on this case. Someone at AIPPI congress summed this up as "bad patents make bad case law" when talking about the "Alice" case. I agree with that and think that Lantana may be similar.

For what it's worth (probably not very much), if the prior art had been limited to computers, internet, email, and some remote file retrieval protocols relying on continuous connections, would the invention be patentable? I think maybe. In that case I can see a technical contribution for something like the reasons articulated by the Applicant. The "new bit" may be just software, but the *contribution* is a better file retrieval system which solves some technical problems.

I haven't looked into exactly what prior art was "on the table" in great detail. There is a clue at paragraph 39 of the Judgment that there was relevant prior art on the table which was a bit more than computers, internet and email but a bit less than the document cited by Mark. This put the IPO, the High Court and the Court of Appeal in murky waters when they didn't need to be.

MaxDrei said...

Freddie thanks. You say:

"...when there is a case about something which really is novel and non-obvious (but said to be excluded matter) we can have a proper debate about "eligibility""

and then you contemplate swimming in "murky waters"

but my point is that when the EPO chooses to confine what is "novel" and "obvious" to novelty and non-obviousness within technology, it keeps out of those murky waters. It has been having a "proper debate", under the rubric of Art 56 EPC, for decades without any great difficulty or distress.

You might argue that it is mere political sleight of hand to put off till Art 56 whether the claimed subject matter, borderline eligible, is patentable because, when you get to Art 56, you have to have the same debate anyway. But I don't see it like that. At least by then you have "done" novelty, once and once only. I like the way that the EPO thus makes clear to the user community that when an inventor makes a novel contribution to a field of technology, the only non-obviousness that "counts" is technological non-obviousness and that accountancy, business or marketing ingenuity, regardless how non-obvious it is, is not enough to get you a patent from the EPO.

I know what the EPO means by "non-obvious" but not what you have in mind. Perhaps we both need to look at the English PTFE corkscrew case, that distinguished between technical and commercial obviousness?

Anonymous said...

MaxDrei @ 14:13,

Keeps out?

Is that "per se" or "as such," - being that "technological" is (and remains) undefined (no matter the reason why), one can hardly say "keeps out" with a straight face.

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':