In 2011, U.S. Patent 6,625,582, entitled Method And System For Converting A Designated Portion of Future Social Security And Other Retirement Payments To Current Benefits, was assigned to Retirement Capital Access Management Company LLC. Benefit Funding Systems LLC asserted the ‘582 patent against U.S. Bancorp in June of 2012. Benefit Funding Systems LLC v. U.S. Bancorp, Case No. 1:12-cv-803-LPS (D. Del. filed June 22, 2012). In response, U.S. Bancorp filed a CBM petition requesting review of claims 1, 13, 14, 18, 30, and 31 of the ‘582 patent on March 29, 2013. U.S. Bancorp asserted that the ‘582 patent qualified for CBM review under 35 U.S.C. § 324 and Sec. 18(a) of the Leahy-Smith America Invents Act, Pub. L. 112-29, 125 Stat. 284, 329 (2011), and that these claims were invalid under 35 U.S.C. § 101.
Archive for the ‘statutory subject matter’ Category
U.S. Bancorp CBM Results in Cancellation of Retirement Capital Access Management Co.’s Patent ClaimsFriday, August 29th, 2014
Posted: September 8, 2013
On September 5, 2013, the Federal Circuit affirmed a District court holding that a computer system claim was not patent-eligible under 35 U.S.C. § 101. In Accenture Global Servs., GmbH v. Guidewire Software, Inc., a divided panel affirmed a District of Delaware decision finding system claims 1-7 of U.S. Patent 7,013,284 not patent-eligible on summary judgment. (Accenture Global Servs., GmbH v. Guidewire Software, Inc., 800 F. Supp. 2d 613, 621-22 (D.Del. 2011).) Judge Lourie authored the majority opinion of the panel (Judges Lourie, Reyna, and Rader), and Judge Rader filed a dissenting opinion.
District Court Action Results in Summary Judgment of Invalidity Based on 35 U.S.C. § 101
Accenture sued Guidewire for infringement of the ‘284 patent in late 2007. In 2011, the District court found claims 1-22 of the ‘284 patent invalid under 35 U.S.C. § 101 on summary judgment. The District court decided the summary judgment motion after the Federal Circuit and Supreme Court decisions in In re Bilski, 545 F.3d 943 (Fed. Cir. 2008)(en banc) aff’d on other grounds sub nom. Bilski v. Kappos, 560 U.S. ____, 130 S. Ct. 3218 (2010).
Federal Circuit Appeal
Accenture appealed the District court determination only as to claims 1–7, directed to a system for generating tasks to be performed in an insurance organization, but did not appeal similar method claims 8–22. The Federal Circuit affirmed the District court in a 2:1 split panel decision by leveraging the finding of patent ineligibility of method claims 8-22 that were not appealed:
We conclude that the district court’s decision on patent-ineligibility of the system claims must also be affirmed, both because the system claims offer no meaningful limitations beyond the method claims that have been held patent-ineligible and because, when considered on their own, under Mayo and our plurality opinion in CLS Bank, they fail to pass muster. Although the issue of the patent eligibility of the method claims is not before us, as it has not been appealed, it is plain to us that, as the district court held, those claims are ineligible for patent.
Because the ’284 patent’s method claims have been found to be patent ineligible, we first compare the substantive limitations of the method claim and the system claim to see if the system claim offers a “meaningful limitation” to the abstract method claim, which has already been adjudicated to be patent-ineligible. CLS Bank, 717 F.3d at 1291. Under this analysis, we compare the two claims to determine what limitations overlap, then identify the system claim’s additional limitations. Essentially, we must determine whether the system claim offers meaningful limitations “beyond generally linking ‘the use of the [method] to a particular technological environment.’” Id. (quoting Bilski, 130 S. Ct. at 3230).
According to the majority, one must consider if the method claims are similar enough to the apparatus claims to warrant a conclusion that the apparatus claims are equally not patent-eligible.
More about the ‘284 Patent
The ‘284 patent claims describe a computer system and method for handling insurance-related tasks. The Federal Circuit decision summarized the claims as follows:
Claim 1 is a claim to a system for generating tasks to be performed in an insurance organization. The system stores information on insurance transactions in a database. Upon the occurrence of an event, the system determines what tasks need to be accomplished for that transaction and assigns those tasks to various authorized individuals to complete them. In order to accomplish this, the claimed system includes an insurance transaction database, a task library database, a client component for accessing the insurance transaction database, and a server component that interacts with the software components and controls an event processor, which watches for events and sends alerts to a task engine that determines the next tasks to be completed.
Claim 8 claims a method for generating tasks to be performed in an insurance organization. The method takes an insurance transaction and applies rules to that transaction to determine tasks to be completed. These tasks are made accessible to authorized individuals who then complete the task.
The text of claim 1 and claim 8 are provided for comparison:
1. A system for generating tasks to be performed in an insurance organization, the system comprising:
an insurance transaction database for storing information related to an insurance transaction, the insurance transaction database comprising a claim folder containing the information related to the insurance transaction decomposed into a plurality of levels from the group comprising a policy level, a claim level, a participant level and a line level, wherein the plurality of levels reflects a policy, the information related to the insurance transaction, claimants and an insured person in a structured format;
a task library database for storing rules for determining tasks to be completed upon an occurrence of an event;
a client component in communication with the insurance transaction database configured for providing information relating to the insurance transaction, said client component enabling access by an assigned claim handler to a plurality of tasks that achieve an insurance related goal upon completion; and
a server component in communication with the client component, the transaction database and the task library database, the server component including an event processor, a task engine and a task assistant;
wherein the event processor is triggered by application events associated with a change in the information, and sends an event trigger to the task engine; wherein in response to the event trigger, the task engine identifies rules in the task library database associated with the event and applies the information to the identified rules to determine the tasks to be completed, and populates on a task assistant the determined tasks to be completed, wherein the task assistant transmits the determined tasks to the client component.
Claim 8 reads as follows:
8. An automated method for generating tasks to be performed in an insurance organization, the method comprising:
transmitting information related to an insurance transaction;
determining characteristics of the information related to the insurance transaction;
applying the characteristics of the information related to the insurance transaction to rules to determine a task to be completed, wherein an event processor interacts with an insurance transaction database containing information related to an insurance transaction decomposed into a plurality of levels from the group comprising a policy level, a claim level, a participant level and a line level, wherein the plurality of levels reflects a policy, the information related to the insurance transaction, claimants and an insured person in a structured format;
transmitting the determined task to a task assistant accessible by an assigned claim handler, wherein said client component displays the determined task;
allowing an authorized user to edit and perform the determined task and to update the information related to the insurance transaction in accordance with the determined task;
storing the updated information related to the insurance transaction; and
generating a historical record of the completed task.
Federal Circuit Majority Compares Apparatus Claims to Method Claims
The majority compared the method and system claims of the ‘284 patent:
It is not disputed by the parties that the ’284 patent’s system claim 1 includes virtually the same limitations and many of the same software components as the patentineligible method claims. Both claims are for “generating tasks to be performed in an insurance organization.” . . . Both the claimed system and the claimed method contain an insurance transaction database containing information relating to an insurance transaction “decomposed into a plurality of levels from the group comprising a policy level, a claim level, a participant level and a line level, wherein the plurality of levels reflects a policy, the information related to the insurance transaction, claimants and an insured person in a structured format.” . . . Additionally, claim 1 and claim 8 both contain: a client component, . . . a task assistant, . . . and an event processor, . . . . The system claims are simply the method claims implemented on a system for performing the method.
Accenture only points to system claim 1’s inclusion of an insurance claim folder, a task library database, a server component, and a task engine in attempting to show that the system claim is meaningfully different from the ’284 patent’s method claims. However, these software components are all present in the method claims, albeit without a specific reference to those components by name. (underlining added)
The majority decision further determined that the ‘284 patent does not distinguish the system and method claims:
Indeed, even the specification of the ’284 patent makes little distinction between the system and method claims. The patent describes the invention as “[a] computer program . . . for developing component based software capable of handling insurance-related tasks.” Id. col. 3 ll. 23–25. The patent then discloses detailed software descriptions of the various software components without differentiating between the system or method claims. Further, although the patent’s Figure 1 shows a schematic diagram of the invention, one that includes computer hardware, the schematic’s hardware is merely composed of generic computer components that would be present in any general purpose computer. . . . The patent thus discloses that the representative hardware for the ’284 patent is a generic computer. In fact, other than the preamble to claim 1 stating that it is a system claim, the limitations of system claim 1 recite no specific hardware that differentiates it from method claim 8. Indeed, in this case “[t]he system claims are [akin] to stating the abstract idea [of the method claim] . . . and adding the words: ‘apply it’ on a computer.” CLS Bank, 717 F.3d at 1291 (plurality opinion) (citing Mayo, 132 S. Ct. at 1294).
Because the system claim and method claim contain only “minor differences in terminology [but] require performance of the same basic process,” id. at 1291, they should rise or fall together. Accenture only cited four additional limitations in system claim 1, and we have already indicated why those limitations do not meaningfully distinguish the abstract idea over the patent ineligible method claim. While it is not always true that related system claims are patent-ineligible because similar method claims are, when they exist in the same patent and are shown to contain insignificant meaningful limitations, the conclusion of ineligibility is inescapable. Thus, like the unappealed method claims, the system claims of the ’284 patent are invalid under 35 U.S.C. § 101.
Federal Circuit Majority Reviews the System Claims “On Their Own”
The majority decision included a separate section to analyze the claims on their own, stating that the system claims are ineligible for patenting because they fail to include limitations that set them apart from the abstract idea of handling insurance-related information. The majoried identified this “abstract idea” of the claim:
The abstract idea at the heart of system claim 1 of the ’284 patent is “generating tasks [based on] rules . . . to be completed upon the occurrence of an event.” ’284 patent col. 107 ll. 25, 38–39. Although not as broad as the district court’s abstract idea of organizing data, it is nonetheless an abstract concept. Having identified the abstract idea of the claim, we proceed with a preemption analysis to determine whether “additional substantive limitations . . . narrow, confine, or otherwise tie down the claim so that, in practical terms, it does not cover the full abstract idea itself.” CLS Bank, 717 F.3d at 1282 (citing Mayo, 132 S. Ct. at 1300; Bilski, 130 S. Ct. at 3231; Diamond v. Diehr, 450 U.S. at 187); see also Ultramercial, 2013 WL 3111303, at *8 (“[T]he relevant inquiry is whether a claim, as a whole, includes meaningful limitations restricting it to an application, rather than merely an abstract idea.” (citing Prometheus, 132 S. Ct. at 1297)).
The majority decided that Accenture’s position (that the claim is limited by being applied in a computer environment and within the insurance industry) was not sufficient to adequately “narrow, confine, or otherwise tie down the claim,” citing Bilski, Diamond v. Diehr, and Parker v. Flook. The decision noted that the specification was complex, but deemed the claims to recite only “generalized software components arranged to implement an abstract concept on a computer.”
Federal Circuit Majority Compares the ‘284 Patent Claims to the Ultramercial Claims
The majority differentiated claim 1 of the ‘284 patent from the claims in the Ultramercial patent (which were found to be patent-eligible by the Federal Circuit panel only recently) by finding that claim 1 is more abstract than the claims upheld in Ultramercial. Furthermore, the majority found that the instant case raise different procedural issues than the Ultramercial case, because in Ultramercial the claims had not been interpreted and were the subject of dismissal on a 12(b)(6) preanswer motion to dismiss.
Judge Rader Dissents
Judge Rader’s dissent made the following points:
- Judge Rader cited his decision in Ultramercial, Inc. v. Hulu, LLC, 2010-1544, 2013 WL 3111303 at *8 (Fed. Cir. June 21, 2013) for the proposition that (1) any claim can be simplified to an abstract idea if stripped of its concrete and tangible limitations, and (2) it is not proper for a court to go hunting for abstractions by ignoring claimed concrete and tangible limitations.
- He disagreed with the majority’s characterization and use of CLS Bank, stating: “[N]o part of CLS Bank, including the plurality opinion, carries the weight of precedent. The court’s focus should be on Supreme Court precedent and precedent from this court.”
- He disagreed with the majority’s position that Accenture’s failure to appeal the finding of invalidity of the method claims estopped it from arguing that the elements therein are directed to patent-eligible subject matter.
- He countered the logic set forth by the majority that the invalidity of the method claims urged similar invalidity of system claims by citing an excerpt from the CLS Bank decision. That excerpt stated that just because one or more method claims was invalid under § 101, that does not dictate that all associated system claims must be likewise invalid.
- Judge Rader would have held that the specific combination of computer components rendered the claim patent-eligible under § 101.
The dissent recalled an earlier quote by the Court prior to grant of en banc review of CLS Bank: “[N]o one understands what makes an idea abstract.” Judge Rader finished by reiterating his view that the statute should be consulted, which offers broad categories of patent-eligible subject matter.
It seems that the Federal Circuit could not be more divided on the subject of patent eligibility than it is today. As long as the basics of patent-eligibility remain uncertain parties will have less knowledge about the strength of their patents and will expend a lot of resources posturing and briefing on these issues. That will impact the resources of the courts. Regardless of your position on patents and patent-eligibility, I think we all agree that what is needed is a patent system that is as uniform, fair, and predictable as possible. It will be interesting to see if the Supreme Court is going to take this opportunity to again weigh in on patent-eligibility.
Even though the Rehearing Request filed by Versata last week is confidential, we can glean some insight about what it contained based on the publicly available documents of record. Today the PTAB authorized SAP to file its motion to oppose Versata’s Rehearing Request, stating:
Patent owner Versata filed a motion for rehearing [ ] of the Board’s Final Written Decision [ ]. Versata’s motion raises at least two issues that SAP has not yet had an opportunity to brief. Specifically, SAP has not yet had an opportunity to brief how the following two Federal Circuit’s decisions affect this proceeding: Ultramercial, Inc. v. Hulu, LLC, 107 USPQ2d 1193 (Fed. Cir. 2013) and Versata Software Inc. v. SAP America Inc., 106 USPQ2d 1649 (Fed. Cir. 2013).
The Board authorizes SAP to file an opposition to Versata’s motion for rehearing. The opposition is limited to addressing the impact, if any, the two recent Federal Circuit decisions have on the Board’s Final Written Decision. SAP’s opposition is due no later than July 18, 2013.
The Ultramercial case resulted in a reversal of a district court holding of invalidity under 35 U.S.C. § 101. You may recall that Ultramercial’s U.S. 7,346,545 patent (the ‘545 patent) claims relate to a method for distributing copyrighted products over the Internet where the the consumer received a copyrighted product for free in exchange for viewing an advertisement (and the advertiser pays for the copyrighted content). The district court granted Wildtangent’s motion to dismiss pursuant to FRCP 12(b)(6), asserting that the method claims of the ‘545 patent did not claim statutory subject matter, but the Federal Circuit reversed the district court holding the subject matter patent-eligible under 35 U.S.C. § 101. Therefore, it is not surprising that Versata cited Ultramercial in its Rehearing Request.
Of course, it is also not surprising that Versata cited decisions from its parallel Federal Circuit appeal given the fact that the Federal Circuit upheld Versata’s district court damages decision and recently denied stay of the Federal Circuit appeal in which the Federal Circuit recently remanded the case to the district court for correction of the injunction.
SAP has until July 18, 2013 to file its opposition; however, like Versata’s Rehearing Request, SAP’s opposition brief may not be publicly accessible if it addresses protected information from Versata’s Rehearing Request.
Progressive Casualty Insurance Co. sued different insurance companies for patent infringement of 5 of its patents in 2010-2012 in the Northern District of Ohio. (Cases 1:10CV01370 and 1:11CV00082 against Safeco; Case 1:12CV01068 against State Farm; and Case 1:12CV01070 against Hartford.) One of the defendants is Safeco Insurance Company, which has Liberty Mutual as its parent. In 2012 and 2013 Liberty Mutual filed ten covered business method patent review (CBM) petitions (two CBM petitions were filed per patent). Eight of these ten petitions were instituted for trial and two petitions were denied, but each of the five patents has at least one CBM where trial was instituted by the PTAB.
Liberty Mutual and the remaining defendants moved to stay the litigation based on the CBMs instituted. Progressive opposed the motion to stay. The District Court heard oral arguments on April 11, 2013, and granted the motion stay on April 17, 2013.
The court used a four-factor test set forth in the AIA section pertaining to CBMs (AIA § 18(b)(1), P.L. 112-29, 125 Stat. 284, 331):
- (1) whether a stay, or the denial thereof, will simplify the issues in question and streamline the trial;
- (2) whether discovery is complete and whether a trial date has been set;
- (3) whether a stay, or the denial thereof, would unduly prejudice the nonmoving party or present a clear tactical advantage for the moving party; and
- (4) whether a stay, or the denial thereof, will reduce the burden of litigation on the parties and on the court.
It is interesting that the Liberty Mutual litigation was previously stayed pending the outcome of ex parte reexaminations, yet the Court found the benefits of inter partes covered business method review compelling enough to order another stay pending the outcome of the PTAB trials. Some of these benefits observed by the Court include:
- CBM proceedings are inter partes rather than ex parte, which allows Liberty mutual “a better platform to advocate its interests.”
- CBM proceedings are “presided over by a panel of three administrative judges whom are required to have ‘competent legal knowledge and scientific ability,’ 35 U.S.C. § 6(a), as opposed to a single patent examiner.”
- To institute CBM review, the petitioner must show the claims are likely invalid, 35 U.S.C. § 324(a), which is more onerous than meeting the “substantial new question of patentability” standard required to initiate ex parte reexaminations.
- The Court also found the short timeline of the CBM proceedings (to be completed within 18 months of institution of trial), to be attractive and likely to decide issues before the Court.
For further information the order for stay provides the details of the Court’s findings and has a detailed table attached at the last page showing the different CBMs and their status.
The first ever covered business method patent review stems from a patent litigation between Versata and SAP over two Versata patents relating to pricing products in mulitlevel product and organizational groups. The district court action began in 2007 when Versata sued SAP for alleged infringement of its U.S. Patent Nos. 6,553,350 and 5,878,400 (Versata Software, Inc v. SAP America, Inc., No. 2:07-cv-00153 (E.D. Tex)).
A first jury trial was held in 2009, and a second jury trial was held in 2011. In the first trial, the jury found both patents directly infringed and inducement of infringement of one claim of the ‘350 patent. Shortly after the first trial, the Court granted JMOL of noninfringement of the ‘400 patent. A second trial was held on the ‘350 patent and a jury found the ‘350 patent infringed, resulting in a $392 million judgment and an injunction dated September 9, 2011 (stayed pending appeal).
SAP filed its petition for covered business method patent review on the first day it was available via the AIA: September 16, 2012. The petition included grounds for challenging the ‘350 patent under 35 USC §§ 101, 112, and 102.
About a month later, SAP filed a notice of appeal of the district court decision in the Federal Circuit (filed on October 11, 2011). Briefing was completed some months later, and oral arguments were held on February 4, 2013. About a month before the oral arguments (January 9, 2013), the PTAB instituted trial on the challenges under 35 USC §§ 101 and 102, but declined review on the challenges under 35 USC § 112. So as of January, 2013, the dispute was being pursued both in the Federal Circuit and in the PTAB.
Based on a series of communications between the parties and the PTAB, SAP agreed to drop its challenge under 35 USC § 102 in exchange for an expedited trial on its challenges under 35 USC § 101. That trial is scheduled for Wednesday April 17, 2013, where both parties will get an hour to present their arguments concerning 35 USC § 101.
Versata took one further action to challenge the covered business method review by suing the PTO in the Eastern District of Virginia on March 13, 2013. Versata Development Group, Inc. v. Rea, 1:13-cv-00328-GBL-IDD (E.D. VA). Versata alleges that the ‘350 patent is not a covered business method patent, and therefore the PTO violated its statutory authority by institution of the review proceeding. Versata also alleges that the PTO has no authority to entertain challenges to the claims of the ‘350 patent under 35 USC § 101, and that the PTO exceeded its statutory authority by institution of the review proceeding on this ground.
So there is a race to final decision between the district court action and the PTAB proceeding, and Versata’s suit of the PTO may drive the final outcome. Johann von Goethe was quoted as saying “Every second is of infinite value.” Perhaps that is more true than ever for SAP and Versata. It will be interesting to see how this story unfolds this week and in the months to come.
A company called CoreLogic Solutions, LLC owns U.S. Patent No. 5,361,201, relating to a process for appraising real estate property. The ‘201 patent was filed on Oct. 19, 1992 and issued on Nov. 1, 1994. Absent some kind of patent term extension, the ‘201 patent will expire soon. The ‘201 patent is the subject of both a litigation and a covered business-method (CBM) patent review with an accused infringer/petitioner Interthinx, Inc. The litigation is CoreLogic Information Solutions, Inc. v. Fiserv, Inc., et al., Civil Action No. 2:10-CV-132-RSP (E.D. Texas). The CBM patent review is CBM2012-00007. Claim 1 of the ‘201 patent is as follows:
1. A computer-implemented process for appraising a real estate property, comprising the steps of:
collecting training data;
developing a predictive model from the training data;
storing the predictive model;
obtaining individual property data for the real estate property;
generating a signal indicative of an appraised value for the real estate property responsive to application of the obtained individual property data to the stored predictive model;
developing an error model from the training data;
storing the error model; and
generating a signal indicative of an error range for the appraised value responsive to application of the individual property data to the stored error model.
PACER reveals a jury verdict form just entered in the Texas litigation on September 28, 2012, where Interthinx was found to not have infringed claims 1 and 10 of the ‘201 patent, and that Interthinx failed to show claims 1 or 10 invalid. This jury decision comes only 9 days after the CBM petition was filed in the USPTO/PTAB.
What will happen next is anyone’s guess. Interthinx may elect to wait and see if a CBM trial is instituted. It may settle with CoreLogic to avoid appeal. Or it may do a little of both. It may also pursue the CBM in an effort to challenge the patent and extinguish any possible reversal and remand by the Federal Circuit on appeal of the trial court action by CoreLogic.
If the CBM PTAB trial is pursued, it will be interesting to see how the PTAB regards a prior litigation decision, such as the one in this case, given the differences between administrative review and litigation. The differences in interpretive standards and burdens of proof are summarized in the slide below:
We already learned from cases such as In re Baxter International (Fed. Cir. 2012), that the PTO can find a patent unpatentable even after the Federal Circuit affirmed a lower court decision that the patent was not invalid. As stated in Baxter:
“[T]he PTO in reexamination proceedings and the court system in patent infringement actions “take different approaches in determining validity and on the same evidence could quite correctly come to different conclusions.’[citing Swanson, quoting Ethicon]”
But that does not mean that the PTAB trial judges will ignore district court decisions and that courts will ignore PTAB decisions. It remains to be seen how much one process defers to the other.
Today I had the pleasure of co-presenting at the Midwest IP Institute on various post-issuance proceedings with Kevin Rhodes, Chief Intellectual Property Counsel and President of 3M Innovative Properties Company. A PDF of our joint presentation is found here.
The presentation provides a comparison between IPR (inter partes review), PGR (post grant review), and CBM (covered business method) patent review. It contrasts these proceedings to ex parte reexamination (EPX). The goal was to present the available options for review of patents now that inter partes reexamination is no longer available.
The presentation further covered administrative trials in the PTAB. A hypothetical was used to demonstrate the use of litigation, IPR, PGR, CMB, and EPX depending on strength of 35 USC § 101 and § 112 arguments as opposed to 35 USC § 102 and § 103 prior art invalidity arguments. Different scenarios were used to demonstrate the complexity of the analysis.
My thanks to Kevin Rhodes and 3M for allowing me to post these slides.
If you are monitoring adoption of post-issuance filings you may have noticed that five more IPR filings were filed on the second day of operation of the PTAB trials portal pursuant to the new IPR and CBM patent review options from the America Invents Act. The list of today’s filings is reproduced below:
4 IPRs were filed by Intellectual Ventures Management, LLC and one by Ariosa Diagnostics. This brings the total count to 13 pending petition matters (10 IPRs and 3 CBMs) on the petitions docket.
It is interesting to note different styles presented by the petitioners in the various petitions. Some rely more on declaration evidence than others. Some include substantial legal background for positions taken and some are more direct and pointed in arguing their positions.
It is also interesting to see the way that claim construction is being postured in these early petition filings. For example, for matters that are in more advanced stages of litigation petitioners have resorted to admissions by the patent owner and Markman decisions for proposed claim constructions. In a presentation to our local bar I predicted that PGR, IPR, and CBM would provide a kind of intermediate state of claim interpretation somewhere between BRI and a Phillips-type construction. I believe that is starting to take shape as evidence from both prosecution and district court is relied upon for claim construction proposals.
It is encouraging to see highly technical arguments being pursued in these early filings. For example, CBM (and PGR in general) may provide an attractive vehicle to challenge statutory subject matter and indefiniteness before the PTAB, as opposed to making these challenges in district court litigation. The patent community may well embrace these post-grant proceedings if it sees consistently high quality decisions by the PTAB.