Monday, August 15, 2016

Is it Time to Revisit the Rule Against Using Extrinsic Evidence in Claim Construction?

Pollard Banknote Limited v BABN Technologies Corp 2016 FC 883 Locke J

In this declaratory action brought by Pollard, Locke J declared both claims of the ‘551 patent, owned by Scientific Games (the successor to BABN), to be invalid (Claim 1 for obviousness 242], Claim 2 for ambiguity [144]). On the infringement counter-claim, he held that the claims were not infringed in any event [2]. The decision raises a number of interesting issues, perhaps the most of important of which relates to prosecution history estoppel.

The ‘551 patent relates to a security feature for instant lottery tickets. These tickets have a scratch-off layer covering the “game data” which the customer scratches off to reveal whether the ticket is a winner and the amount of any prize. (In cheaper tickets a paper pull-tab may be used to cover the prize instead of the scratch-off layer.) To prevent fraud, the tickets normally also have validation information, in the form of numbers and/or a bar code, which is checked against the lottery’s online system on redemption to confirm that the ticket had been legitimately sold, that it was indeed a winner, and that it had not previously been redeemed [12]. As additional security, the validation information may also be hidden by a scratch off layer. This was all part of the prior art.

These aspects of lottery tickets are shown in Figure 3 of the ‘551 patent:

Figure 3. A ticket on which the game data and the bar code are hidden under separate scratch-off layers.

To understand prosecution history estoppel issue, it is helpful to start with the original claims (of 2011-09-16):

1. A printed document comprising:

(a) a substrate [ie the paper on which the ticket is printed];

(b) a play area on the substrate comprising printed indicia covered by a removable scratch-off coating, said printed indicia when present in a desired format may result in a prize being won; [ie customer scratch-to-win area]

( c) a non-play area on the substrate spaced apart from the . . . play area and including . . . a [machine-readable] bar code [with validation information] and being covered by a removable scratch-off coating. . . ; and

(d) wherein the bar code is a two dimensional bar code.

All of these elements – in particular, a play area (34, showing a $10 prize) and a separate non-play area with a bar-code (36) – are shown in Fig 3.

Fig 3 also shows two distinct scratch-off layers covering the separated play and non-play (validation) areas. This is claimed more particularly in the original Claim 3:

3. The printed document of claim 1 wherein the printed indicia and the bar code are covered by separate removable scratch-off coatings.

Claim 2 covers the same type of ticket, except where a single scratch off layer extends across both the play and non-play areas:

2. The printed document of claim 1 wherein the printed indicia and the bar code are covered by the same removable scratch-off coating.

The examiner objected to these claims on the basis of the Camarato application (No. 2,119,190), which shows this ticket:
Figures 1 and 2 illustrating a bingo-style lottery ticket. The Camarato application has a play area and a separate bar code, both of which were covered by separate pull tabs [165]. The Camarato application specified that the pull-tabs could be replaced by scratch-off layers, so this could not be used to distinguish it from the invention of the ‘551 patent. While the Camarato application only indicated a 1D bar code, rather than the 2D bar code specified in the 551 patent, the examiner was of the view that it would be obvious to substitute a 2D bar code. To overcome this objection, SG, in its response of 6 March 2013, stated that the Camarato Application “clearly does not teach coating the bar code and the play area indicia with a continuous scratch off coating” [emphasis in original] [236]. At the same time, SG amended the claims to the form in which the patent was ultimately granted. The original Claims 2 and 3 were deleted and Claim 1 was amended as follows:

1. A scratch-off lottery ticket comprising:

(a) a substrate;

(b) a play area . . .;

(c) a non-play area on the substrate spaced apart from the printed indicia of the play area and including an authentication means comprising a [machine readable] two dimensional (2D) bar code [with validation information];

(d) a removable continuous scratch-off coating covering both the printed indicia in said play area and the bar code in said non-play area. . . .

Paragraphs (c) and (d) of the original Claim 1 were combined, and a new limitation was included in the new paragraph (d), evidently in order to distinguish Camarato. In effect, the original Claim 1 became Claim 2 ,and Claim 3, to a ticket with separate scratch-off coatings, was deleted. 

The contentious phrase was "continuous" in paragraph (d). As Locke J noted, the prosecution history "can only mean that this phrase was intended to be limited to a single scratch-off coating. This argument was obviously successful, since it was followed by a notice of allowance” [236]. However, in this litigation, the position SG took on the construction of “continuous” was “entirely opposite to the position it took before the Examiner in order to obtain allowance of the 551 Patent” [115]. Before the court, SG argued that “the word ‘continuous’ does not suggest that there is a single scratch-off coating, but rather that the coating, whether there is only one or more than one, completely hides (is continuous over) each of the printed indicia and the bar code” [111]. Locke J called this a “remarkable” shift in position, saying it was “breathtaking to see SG now attempt not just to take a different position on the construction of claim 1, but also to argue that, by doing so, it does not reintroduce the problem of obviousness in light of the Camarato Application that it had previously argued was avoided applying its first position” [237].

It may be “breathtaking” to see a patentee attempt to game the system in this manner, but as Locke J noted, under current Canadian law as set out by the SCC in Free World Trust 2000 SCC 66 [66], prosecution history is inadmissible [79]. This means it is possible for a patentee to give up ground in prosecution in order to obtain its patent, and then reclaim exactly the same ground during subsequent litigation. This is not a merely theoretical possibility; it is exactly what happened in this case. Locke J, ignoring the prosecution history as he was required to do by Free World Trust, adopted the broader construction of the term advanced by SG in litigation, rather than the narrower construction SG had put forward in prosecution [119].

While Locke J followed Free World Trust, he made some trenchant observations about the SCC’s reasoning. The SCC stated that “[i]f significant representations are made to the Patent Office touching the scope of the claims, the Patent Office should insist where necessary on an amendment to the claims to reflect the representation” 2000 SCC 66 [66]. That is the counsel of perfection, but as Locke J noted “[t]he SCC did not address the possibility that the Patent Office may fail to insist on amendments to claims to reflect representations made by the applicant” [80]. The SCC also said that “[t]o allow such extrinsic evidence for the purpose of defining the monopoly would undermine the public notice function of the claims.” But “[t]he SCC also did not explain how the patent gives public notice of the claims, but the prosecution history, which is likewise available to the public, does not” [80]. The SCC said that allowing the use of prosecution history would “fuel[] the already overheated engines of patent litigation” [66]. But in this case at least, the opposite is true; Locke J pointed out that “SG’s argument would never have made it to a trial in the US where the principle of file wrapper estoppel applies” [238]. The SCC also asserted that allowing the use of prosecution history would “increase uncertainty” [66]. But in this case it would have reduced uncertainty, as the prosecution history clearly established the meaning of a hotly disputed term. Finally, the SCC also said that allowing the use of prosecution history would “undermine the public notice function of the claims” [66]. That would be a powerful argument against the use of prosecution history as a general interpretive tool, but that is not what is at issue. Even though the SCC referred explicitly to “file wrapper estoppel,” it seems to have neglected the fact that the rule at issue operates solely as an estoppel. A third party can always rely on the claims, read in light of the specification, to determine that its activities are non-infringing, because prosecution history cannot be used to expand the scope of the claims, but only to prevent the patentee from reclaiming subject matter that was given up in prosecution.

In the end, SG’s ability to ignore the concession it made during prosecution did not help it, even though Locke J accepted the broader construction advanced by SG in litigation. SG had taken the narrower position in prosecution to avoid an obviousness rejection, and the broader position meant the main difference between the claimed invention and the Camarato application was the use of a 2D bar code. Locke J held that this modification was obvious, and therefore Claim 1 was invalid [242]. This does not mean that we need not worry about the possibility of the patentee resiling from the position taken in prosecution; in this case, for example, there was a real question as to whether the Camarato application was part of the state of the art [198]. Locke J concluded it was, but if he had come to the contrary conclusion, then SG would have been able to have its cake and eat it too.

Locke J’s decision is an important development relating to the use of extrinsic evidence. On its facts, “[t]his case highlights a potential risk in taking a simpler approach to claim construction by ignoring extrinsic evidence, such as the prosecution history of the patent in suit,” as it shows that the risk of the patentee trying to game the system by reversing its position on claim construction is real, not merely theoretical [239]. That, in combination with Locke J’s observations about the practical reality of the SCC’s reasoning in Free World, “raises the question whether it is time to revisit the rule against using extrinsic evidence in claim construction” [80].

1 comment:

  1. The prospect of prosecution history estoppel might be made more fair if it cut both ways, precluding a court from taking a different position than a patent examiner in evaluating a claim. The idea of allowing examiners and judges to game the system against patentees by changing their approach, while not affording patentees the same flexibility, would appear to make for a game that has different rules for different sides.