Thursday, December 7, 2023

Throwing Mud at the Wall

NCS Multistage Inc v Kobold Corporation 2023 FC 1486 McVeigh J

2,693,676 / 2,820,652 / 2,738,907 / 2,766,026 / 2,820,704 (NCS) / 3,027,571 (Kobold)

This case concerns six patents relating to downhole tools used in fracking. NCS sued Kobold for infringement of five NCS patents. In response, Kobold sued NCS for infringement of Kobold’s 571 patent. It did not go well for NCS — all the claims in issue for all five patents were declared invalid, while Kobold was successful in its infringement counterclaim.

In broad brush terms, this strikes me as a case where a major new technology created new problems with obvious solutions. Horizontal fracking is relatively recent. It presents some new problems, such as clearing debris away from the tool. In a vertical well, this is accomplished largely by gravity — debris falls to the bottom of the well — but in horizontal wells, some other methods are necessary. The methods claimed by NCS were generally obvious or had already been tried: for example, if debris is getting stuck in a moving part, machine out a bit more clearance to allow free movement of the debris [1191]. Several ancillary arguments were raised along the way, of which the most important is the discussion of double patenting. I’ll deal with double patenting and overbreadth in subsequent posts. This post will provide an overview and briefly address a few of the other legal issues.

This is a very long decision. That is a direct result of an “everything but the kitchen sink” approach: “[a]fter closing submissions, the parties had narrowed the scope of claims at issue down from approximately 145 claims across seven patents to approximately 123 claims at issue across six patents, still with 128 separate issues.” McVeigh J was not impressed. She quoted Locke JA’s comments in Western Oilfield 2021 FCA 24 [9] as to “the lack of wisdom of raising so many issues on appeal,” with the result that “the appellants’ approach suggests that they themselves cannot identify any issues that are particularly strong” [5]. She reiterated that “putting all these issues before the Court was an unwise litigation strategy” and warned that

[9] Lest this become a trend for this Court, or for the parties before it, I feel compelled to state that, in my view, putting these countless issues before the Court for simultaneous determination was detrimental to the aims of the parties in presenting the most clear and effective arguments. It is also contrary to the general principle of this Court to secure the just, most expeditious and least expensive outcome of every proceeding.

While McVeigh J directed her general comments at both parties, the fault lay primarily with NCS. Kobold successfully asserted one patent and four claims, which depended on three further claims [1623]. NCS asserted five patents and put over 120 claims in issue. NCS was comprehensively unsuccessful. All claims at issue — 128 in total — were declared invalid. Costs consequences followed: “There is a clear window here for an elevated lump sum costs award to incentivize efficiencies while sanctioning conduct with so many issues making it more complex.” Accordingly, McVeigh J awarded Kobold a lump sum of 45% of actual fees, up from the baseline of 37.5%, in significant part because “the multitude of claims and issues in this case unnecessarily enhanced the complexity of this case” [1671], [1674]. I must say that the 7.5% bump in the costs award strikes me as a modest sanction given the waste of judicial resources occasioned by NCS’s strategy. But maybe I am just inclined to be unsympathetic after having waded through 200 pages of claim construction.

While NCS’s strategy was certainly annoying, that doesn’t make it unwise. And while it failed in this case, that doesn’t necessarily mean it was a bad strategy; if you make a bet with good odds, it was a still a good bet to make, even if you lose. This is one of the most extreme versions of the kitchen sink strategy aka the “throw mud at the wall” strategy that I’ve seen, but that strategy is common enough in a more restrained form. This raises an interesting question about the ‘kitchen sink’ strategy: it makes the judges mad and risks an adverse costs award, but maybe that is worth it if it increases your chance of winning. I think it is pretty clear that the kitchen sink strategy sends a message to the court that you have a weak case. But given that you do have a weak case, is the kitchen sink strategy a good bet?

Of course, we can’t do a controlled study to see whether a more focused strategy is better, but I have to admit my eyes glazed over while reading hundreds of pages of claim construction, with each claim getting only a few paragraphs of analysis. And while I can’t see inside McVeigh J’s head, I have to suspect that it would have been easy for a few meritorious points to have been swamped by the routine of dismissing the unmeritorious arguments. This is particularly so because McVeigh J often found the evidence of NCS’s expert, Mr Lehr, to be wanting. He “took several positions which appeared at odds with expertise and historical facts” [92], and his arguments were regularly inconsistent with the documentary evidence: see eg [1286]. In some instances Mr. Lehr would read in new elements that were not in the claim at issue at all [1424]. I suspect NCS’s kitchen sink strategy had the effect of undermining the credibility of their own expert by inducing him to take unsupportable positions on weak claims that should never have been raised, rather than allowing him to take a stronger position on a smaller number of claims.

On the other hand, there were one or two places where the sheer volume of issues may have worked to NCS’s advantage. For example, Kobold did not carry its burden of showing that the 704 patent was obvious, because Kobold’s experts provided only conclusory statements and did not address inventiveness at the claim-by-claim level [1450]–[1452]. It is possible that these weaknesses in Kobold’s submissions were due to the sheer volume of arguments it had to deal with. (The claims were nonetheless invalid on other grounds.)

Some of the discussion was a bit difficult to follow because McVeigh J’s analysis sometimes plunged right into the weeds without making it clear why the issue mattered. Most of the hundreds of pages of claim construction were like that. While claim construction is antecedent to validity, without knowing “where the shoe pinches” it can be difficult to stay awake while reading about whether a “wedge” has to be wedge-shaped.

Priority and the use of file history

One argument that was particularly difficult to follow related to priority dates and the use of file history [1302]–[1337]. The Ravensberger patent (2,730,695), filed 11 Feb 2011 claimed an almost identical invention to the NCS 907 patent [1302], and claimed priority from US applications 12/842,099 (23 June 2010) and 12/971,932 (17 Dec 2010). The 907 patent was filed on 4 May 2011 and claimed priority to an 18 Oct 2010 US application (61/394,077). During prosecution of the 907 patent a protest was filed arguing that if the 907 patent could not claim priority from the US 077 application, then it was anticipated by the Ravensberger patent. (McVeigh J referred to the protest as being filed on 6 March 2012 [1305], but it appears she is referring to the protest filed on 26 March 2012.) The 907 patent nonetheless was granted, from which it might be inferred that the examiner had decided that it properly claimed priority from the 077 application. (Though I should say here that it is not clear to me why any of this mattered.) There was also a protest against the Ravensberger patent based on the 907 patent, apparently also raising a priority issue. NCS wanted to introduce the file history of the Ravensberger patent [1309]. (Again, I don’t understand what argument was being pursued and so I don’t know how the file history was supposed to be relevant.) McVeigh J reviewed this file history, and while she was of the view that the file history might in principle be used for purposes other than claim construction pursuant to s 53.1(1), she decided that she would not use the file history “essentially to re-argue the same protest in relation to the Ravensbergen when the Patent Office issued the patent with those priority dates” [1325]. That seems like a very reasonable conclusion, with the caveat that I don’t understand where the argument was going.

McVeigh J then stated that “[a]lthough Canadian jurisprudence does not appear to have directly addressed this question, priority can be asserted on a claim-by-claim basis” [1331]. I’m not aware of any caselaw on that point either, but nonetheless the point seems perfectly clear on the face of s 28.1(1): “The date of a claim . . . is. . . .” I expect that is why there is no jurisprudence on the point.

Then McVeigh J stated (my emphasis):

[1335] As such, it is possible for different claims to be subject to different priority dates relating to different, previously filed applications. However, I interpret this as taking place during the application process and not at the stage in an action where validity is being determined. This would amount to an ex post facto prosecution of the patent, and contrary to the role of this Court.

This is puzzling. On its face, the underlined sentence seems to to say that the Court cannot assess whether a claim can properly claim priority from a foreign filing. However, in the next paragraph McViegh J stated that “on this record with very limited argument and no expert assistance, I am not prepared to do a claim-by-claim review of priority dates” [1336]. That indicates that she would be prepared to assess priority, just not in the context of this particular patent and in particular given the (puzzling) reliance on the Ravensberger claim history. Further, in discussing Kobold’s 571 patent, McVeigh J did determine different claim dates for different claims [1586], saying “As stated earlier, I agree that it is possible to evaluate the claim date in the way NCS requests” [1580]. I take it the “earlier” is a reference to her discussion at [1302]–[1337] relating to the 907 patent. So, on the whole, it seems clear enough that McVeigh J is not suggesting that the claim date of different claims cannot be determined by the court. I would speculate that the underlined sentence was addressed to the specific argument being made by NCS — maybe to the effect that she was not prepared to review the priority date determinations made by the examiner in light of the prosecution history. That is a very reasonable position, which does not exclude the court reviewing the priority dates directly. I’ll just conclude by saying that it is perfectly clear from 28.2(1)(b) and 28.2(1), which governs anticipation generally, not just in the context of prosecution, that the claim date may be assessed by the court.

Grace Period

An invention must not have been obvious in light of the state of the art. Section 28.3(a) gives a one-year grace period which excludes information derived from the applicant from the state of the art. Kobold wanted to rely on a presentation referred to as “What up?” in attacking NCS’s patent. NCS objected that the presentation contained information derived from NCS and so could not be used [1258]. McVeigh J held that even if the presentation contained some information derived from NCS, 28.3(a) could not be used to exclude all information in the “What up?” presentation [1281]. That seems to be to be clearly correct. Both on the face of the Act and on a purposive analysis, it is only the information derived from the patentee that is excluded, not all information that is somehow associated with the information derived from the patentee.

Mosaicing art from different experts

NCS argued that in attacking the patent for obviousness, Kobold could only rely on references that were discussed by both of its experts, apparently on the view that where a piece of prior art is obvious, it should be obvious to both experts [1417]. McVeigh J rejected this: “I reject the proposition that references must be discussed by both (or all) experts where there is more than one expert giving their opinion for a single party. Once a piece of prior art forms part of the record, the Court is able to consider it under the obviousness analysis” [1418].

Inventive Concept

There has been a lot of controversy over the use of the inventive concept in claim construction: see here and here. My own view is that the EPO objective problem and solution approach is probably the best way: see here for my argument to that effect, and here, here and here describing the EPO approach. In this case, it seems that NCS’s expert generally wanted to characterize the inventive concept at a very abstract level, as being “active debris relief” [1097] (and see eg [1441]), but any real trouble was avoided by McVeigh J’s (clearly correct) insistence that obviousness must be addressed on a claim by claim basis: see eg [1435]–[1436]. In other instances, NCS’s expert described the inventive concept by simply paraphrasing the claims: see eg [1343]. That is a common approach, which is not necessarily a problem, but one has to ask what is gained by focusing on an ‘inventive concept’ that is just a paraphrase of the claims, as opposed to focusing on the claims themselves.


It is difficult to establish ambiguity and consequently a successful attack based on ambiguity is rare, but in this case McVeigh J held two claims to be invalid as being “incapable of being meaningfully interpreted” [1530]; and see [1527]–[1532]. There is no new law here, but successful ambiguity attacks are rare enough that this one is noteworthy. (I did not review the claims myself so I do not have any independent opinion as to whether they were incapable of being meaningfully interpreted.)

Utility related to the nature of the invention

In AstraZeneca 2017 SCC 36 [53], the SCC stated that the utility of an invention must be “related to the nature of the subject-matter,” saying “a proposed invention cannot be saved by an entirely unrelated use. It is not sufficient for an inventor seeking a patent for a machine to assert it is useful as a paperweight.” We don’t see this argument used very often, but in this case NCS sought to save the 704 patent, which McVeigh J found to lack utility because it could not function without seals, by arguing that it would nonetheless be useful for cleaning the casing, presumably by pushing it through to shove debris, in the same way that a completely inoperative tool could be used. McVeigh J rejected this argument on the basis that cleaning did not relate to the nature of the invention [1512]. There is no new law, but as with ambiguity it’s noteworthy just because we don’t see this point raised very often. (NCS’s argument was a bit less absurd than the foregoing summary suggests, because an operable tool would clean the casing as part of its perforating and fracturing capabilities — but it was a desperate argument nonetheless.)


McVeigh J allowed the successful patentee, Kobold to elect between an accounting and damages, saying “With respect to an accounting of profits, no evidence has been advanced to suggest the parties may not be entitled to this relief. With the bifurcated phase with respect to damages ahead, there is no reason to deny such relief given the Defendants came to the Court with clean hands” [1659]. The direct implication is that the successful patentee is presumptively entitled to an accounting, barring some reason for denying it, such as unclean hands. As I noted in a post on an earlier case, in a two part article forthcoming in the IPJ and summarized in a series of blog posts, I have argued that Nova v Dow 2022 SCC 43 wrought such a major change in the principles on which the quantum of an accounting is assessed that the courts should revisit the principles on which an accounting is granted.

No comments:

Post a Comment