Motorola has filed its opposition to Microsoft’s motion to transfer the appeal of Judge Robart’s RAND ruling from the Federal Circuit to the Ninth Circuit (see our prior blog on Microsoft’s motion).  Recall that Microsoft argued that the appealed action was a contract action, its nature did not change when that action was consolidated with a patent action and therefore appellate jurisdiction properly lies in the Ninth Circuit.  In its opposition, Motorola argues Federal Circuit case law that consolidation of an action with a patent action vests appellate jurisdiction in the Federal Circuit.  Motorola additionally argues that Microsoft’s Complaint was constructively amended by the district court when (1) it proceeded to determine a royalty rate under patent law and (2) it made rulings about the scope of the patent claims along the way (such as construing means-plus-function patent claim limitations).

More interesting than the otherwise dry appellate jurisdiction procedural issues presented are the underlying undertones.  The battle here implies that a RAND patent licensee would prefer to have a generalist court decide the value of patents whereas the RAND patent licensor would prefer to have the specialist patent court — Federal Circuit — decide the value of patents.  This implication is supported by the Ninth Circuit’s prior affirmation in this case of Judge Robart enjoining Motorola from enforcing an injunction in Germany on standard essential patents and Judge Posner’s decision not to enter an injunction in the Motorola v. Apple case.  What will the Federal Circuit think about allowing another appellate court to determine the value of patents?

Last Thursday, December 5, the House of Representatives passed H.R. 3309 (“the Innovation Act”), a patent reform bill generally directed to perceived patent litigation abuse by certain patent assertion entities (what some call “patent trolls”).  Prior draft versions of the House bill had gone through several revisions in the past few months (see our September 24 and October 23 posts), and the bill as passed by the House contains a number of provisions that will directly affect future litigation involving Standard Essential Patents (SEPs). These provisions are discussed in more detail below.

Heightened Pleading Standards.  The Innovation Act requires specific details regarding a complainant’s patent infringement allegations.  Similar to existing Patent Local Rules adopted by district courts in E.D. Tex. and N.D. Cal., the Innovation Act requires a patentee to identify the patents, the patent claims specifically asserted, the instrumentalities accused of infringement, and an element-by-element description of how each accused instrumentality practices the asserted patent claims.  Such provisions will undoubtedly provide proponents of early stage Twombly/Iqbal motions with ample grounds to seek dismissal of bare-bones counts for patent infringement.

The Act also requires a patent plaintiff to disclose some basic patent ownership information to the court, other parties, and the U.S. Patent and Trademark Office (PTO), including the patent assignee, the assignee’s parent entity, any entity with a right to sublicense or enforce the patent, and any entity with a financial interest in the patent.

Although it does not appear that a patent-plaintiff must identify existing licensees, it must identify any SSOs to which the patent has been declared essential.  The Act specifically provides a claim for patent infringement must set forth:

[W]hether a standard-setting body has specifically declared such patent to be essential, potentially essential, or having potential to become essential to that standard-setting body, and whether the United States Government or a foreign government has imposed specific licensing requirements with respect to such patent.

The original drafts of this language created problems in requiring a plaintiff to plead “whether such patent is subject to any licensing term or pricing commitments through an agency or standard-setting body” and addressed that by focusing on more knowable facts of whether a patent was “declared” essential or potentially essential (see our September 24 post).  The provision as passed kept the more fact-specific pleading of whether a patent has been declared essential, but misses the mark by requiring disclosure of “whether a standard setting body has specifically declared such patent to be essential …”.  SSOs generally do not declare patents essential, potentially essential or the like; rather, such declarations typically are made by the patent owners themselves in letters of assurances or similar disclosures that often don’t state definitively whether the patent does cover the standard, but that it might cover the standard and what the patent owner would do if the patent actually is essential to the standard.  One wonders whether this issue will be cleaned-up when the Senate considers it.

The provision about pleading whether the U.S. or other government body has imposed licensing obligations probably is in response to some limited actions by competition agencies that require certain procedures to be followed before a patent owner may seek injunctive relief (like the U.S. FTC’s consent order with Google or what Samsung has proposed to the European Commission) or court orders that have sought to put similar limits on injunctive relief.

The revised language goes hand-in-hand with another section of the Act directed to proposed Judicial Conference rules and procedures governing discovery burdens and costs, the Act proposes that the Judicial Conference consider “documents relating to any licensing term or pricing commitment to which the patent or patents may be subject through any agency or standard-setting body” to be included among the categories of “core documentary evidence” that must be produced by a patent-plaintiff to defendants in every litigation.  Such rules would provide defendants with documentary evidence directly related to establishing RAND obligations at an early stage in litigation.

Licensing Obligations Continue Through Bankruptcy.  The Innovation Act ensures that IP licenses are not eliminated in bankruptcy, resolving an apparent discrepancy between U.S. and foreign law in favor of the American rule.  As we discussed in a post earlier this month, the Fourth Circuit recently affirmed a district court decision that licensees could rely on Section 365(n) of the U.S. bankruptcy code to preserve the existing licenses to U.S. patents.  The present legislation codifies this principle, barring a bankruptcy trustee from terminating certain licenses to patents and other intellectual property of the debtor, adding trademarks to definition of “intellectual property” in title 11 proceedings, and requiring a bankruptcy trustee to meet any existing contractual obligation to monitor and control the quality of a licensed product or service covered by a licensed trademark.

The Customer-Suit Exception.  The Innovation Act allows a manufacturer to intervene in a patent suit brought against its customer.  These provisions allow a patent suit to be stayed as to the customer while the manufacturer and patent plaintiff litigate the merits of the infringement action, so long as (1) the manufacturer and the customer consent, (2) the stay is sought within 120 days after the first complaint for infringement, and (3) the customer agrees to be bound by the court’s ruling on any issues in common between the customer and manufacturer.

Other Key Provisions.  Although we haven’t discussed them here, the Innovation Act contains a number of other key provisions, including:

  • Cost Shifting, awarding costs and attorneys fees to a prevailing party unless the position and conduct of the nonprevailing party was reasonably justified in law and fact;
  • Post Grant and Inter Partes Review, allowing a PGR petitioner to later assert invalidity defenses in a civil action that could have been, but were not raised during PGR and requiring the PTO to use district court claim constructions in PGR and IPR proceedings;
  • Expanding the scope of prior art used in transitional program for business method patents covering financial products;
  • Core Discovery and Discover Fee Shifting, limiting the types of discovery prior to claim construction and requiring Judicial Conference to develop rules and proposals limiting discovery in patent cases and to study the efficacy of the rules enacted;
  • Codifying Doctrine of Double Patenting for first-inventor-to-file patents; and
  • Demand Letters, requiring a claimant seeking to establish willful infringement may not rely on evidence of pre-suit notification that fails to set forth claimant’s infringement allegations with particularity.

Next Steps.  Now that the bill has passed the House, the Senate is expected to move quickly. Last Thursday, Senate Judiciary Committee Chairman Patrick Leahy (D-Vt.) confirmed a legislative hearing will be held on December 17 to consider the upper house’s patent reform bill (S. 1702).  The House and Senate measures will have to be merged before the Act is presented for President Obama’s signature, but it is expected the proposed legislation will be signed into law early next year.  Because the bill’s litigation provisions apply to patent cases filed after the date of enactment, and not pending cases, we may see an uptick in the filing of patent cases prior to the anticipated date of enactment (as we saw with enactment of the America Invents Act).

Today the U.S. Supreme Court granted certiorari to consider the patent eligibility of computer-implemented inventions in Alice Corp. v. CLS Bank (the docket is available from SCOTUSblog).  You may recall that a hopelessly divided en banc Federal Circuit held that the computer-implemented patent claims at issue were invalid because they were not directed to patentable subject matter, but there was no majority agreement as to why (see our May post explaining the patent claims and division within the court).

At the surface, this case concerns technical patent nuance issues about the Section 101 standard for what things can be patented.  The issue with computer-implemented inventions typically is whether the patent is simply claiming some broad mathematical concept, thought process or other basic tool of innovation.  The concern is that giving someone exclusive rights to basic tools of innovation (mathematical equations) would keep others from using those tools, thus stifling innovation:  How much progress would be made if only one person was allowed  to use addition, subtraction or multiplication.

This issue is separate from the concern whether someone is trying to patent something that is not new, would have been obvious, is too vague to denote definite bounds of the invention or other traditional patent invalidity defenses.  But often those traditional invalidity issues invade the patentable subject matter debate.  For example, a concern often is whether someone has simply claimed “do a mathematical computation that people have done before by hand or in their mind, but use a computer to do it instead.”  The mathematical computation aspect concerns patentable subject matter issues, but the more visceral reaction often is that’s not an innovation given how obvious it would have been to use a computer to do the computation once computers were available.

This concern is not limited to computer-implemented inventions.  For example, gauge your own reaction when reading U.S. Patent No. 5,443,036 granted in 1995 that is directed to a method of exercising a cat by having it chase a laser light beam.  Do you react more to whether exercising a cat should be eligible for a patent or to whether that is actually an innovation — e.g., cats had been chasing flash light beams and the such long before there were lasers and it would have been obvious to use a laser once handheld lasers became available?

So the deeper dive beyond the surface technical Section 101 issue is what balance will the Supreme Court take in addressing this issue.  Will the Supreme Court put its thumb on the scale to keep the door open to what innovations can be patented and rely on traditional defenses of whether its actually a concrete “innovation”, or will it put its thumb on the other side of the scale.  We’ll eagerly watch and hope that, as they address whatever actual problems may exist with the patent eligibility standard and provide the guidance that the Federal Circuit failed to give, they do no harm to our patent system for innovation in the process.

Qualcomm and Ericsson have filed amicus briefs in the appeal of Judge Crabb’s dismissal of Apple’s declaratory judgment action that sought a court-determined FRAND royalty rate under Motorola patents, where Judge Crabb dismissed the case after Apple would not agree to be bound by that FRAND determination (see our July post). Qualcomm and Ericsson generally agree that the dismissal was appropriate because the prospective licensee would not agree to be bound by the court-determined FRAND.

Qualcomm’s Amicus Brief

Qualcomm’s amicus brief asserts that the dismissal in this case was reasonable legally and as sound policy: “Where a potential licensee seeks an adjudication of a SEP owner’s compliance with its FRAND commitment through specific performance, a district court may properly require the potential licensee to agree to be bound by the terms affirmatively adjudicated as complying with that commitment.” A contrary ruling “could frustrate the objectives of SSO IPR policies to ensure balance of interests and could waste judicial resources, as it would create disincentives to negotiate and lead to continued litigation.”

Qualcomm urges the Court not to upset the balance of incentives and obligations embodied in FRAND commitments by allowing non-binding adjudication of FRAND terms. If parties have a good-faith dispute over FRAND terms, a potential licensee seeking to compel an offer on adjudicated FRAND terms should be bound by that determination and enter into a license agreement on those terms. Otherwise, judicial resources are wasted and the balance between the SEP holder and prospective licensee is unduly disrupted in favor of the licensee. Litigation will then become a means for prospective licensees to inject cost and delay to the clear disadvantage of an SEP owner.

Qualcomm asserts that standardization affords consumers with benefits of rapidly evolving technology, increased competition among implementers of standardized products and services, and enhanced performance/features at lower cost. However, these benefits are not realized if licensees do not negotiate FRAND in good faith, such as by seeking specific performance of FRAND commitments in court without being bound by the outcome if the licensee deems it undesirable. SEP owners would be denied compensation and unlicensed implementers would gain a monetary competitive advantage over licensees who have previously entered into FRAND licenses with the patent holder.

Qualcomm further asserts that dismissal was consistent with maintaining the balance of competing interests and applicable law governing FRAND commitments. First, the district court correctly concluded that a non-binding opinion would only improve one party’s negotiating position. Second, the general contract principles of contract law support the district court’s decision. Although the SSO’s IPR policies did not expressly require potential licensees to negotiate in good faith, Apple is claiming to be a third party beneficiary of the SSO-patentee contractual agreement and must abide by the contractual obligation to negotiate in good faith. And “[a]sking a court to compel specific performance of the FRAND commitment while simultaneously refusing to be bound by the outcome of that adjudication cannot be reconciled with the potential licensee’s good faith obligation.”

Ericsson’s Amicus Brief

Ericsson’s amicus brief asserts that allowing a prospective licensee to proceed with a declaratory judgment action without being bound to the ultimate determination renders the requested judgment no more than an advisory opinion. “It is one thing, however, for a district court to resolve a FRAND-related licensing dispute between two parties that have consented to be bound by the court’s determination of what is fair, reasonable, and nondiscriminatory. It is another thing entirely for a prospective licensee to ask a district court to exercise its discretionary and equitable powers to determine FRAND licensing terms—all the while refusing to be bound by the court’s determination.”

Because Apple refused to be bound by the court’s determination, principles of equity and contract justified dismissing Apple’s claims for equitable relief. Specifically, Apple took the position that it would not “pre-commit” to terms found by the court to be FRAND, instead preserving the option to bargain down from the royalty terms set by the court. “Apple thus asked the district court to exercise its discretionary and equitable powers to determine fair, reasonable, and nondiscriminatory terms—while reserving the right to use those FRAND terms as ‘a ceiling’ in subsequent efforts to secure terms that, in light of the court’s determination, could be unfair, unreasonable, or discriminatory.”

Ericsson further asserts that policy considerations favor Motorola, noting that the prevalence of open standards benefit consumers and have a positive effect on the economy. The FRAND regime is critical to the continued success of open standards by ensuring innovators receive a fair return on investment in standardization. Allowing potential licensees to secure sub-FRAND terms does not provide fair return on investment to the SEP holders, and encourages shifting away from interoperability-enhancing standardization. Ericsson warns that requiring courts to set FRAND terms where prospective licensees refuse to be bound by those terms improperly would encourage using litigation to leverage sub-FRAND licensing terms.

Yesterday the Fourth Circuit issued a decision in Jaffe v. Samsung, et al. regarding the preservation of existing U.S. patent licensing rights that various semiconductor companies had through cross-licensing with Qimonda AG, a German semiconductor manufacturer going through bankruptcy proceedings in Germany.  The decision does not state whether any standard essential patents (SEPs) were at issue, but does discuss issues of a reasonable and non-discriminatory royalty (RAND), patent “holdup” and prejudice to an industry if existing licensing agreements do not follow the patents through bankruptcy.

Qimonda AG had been one of the largest DRAM manufacturers, having thousands of worldwide patents, including about 4,000 U.S. patents.  Qimonda had entered cross-licensing agreements with competitors “to avoid infringement risks caused by the ‘patent thicket’ resulting from the overlapping patent rights of some 420,000 patents in the semiconductor industry.”

In 2009, Qimonda filed for bankruptcy in Germany.  The bankruptcy trustee, Jaffe, sent letters to the cross-licensed companies invoking a provisoin of German bankruptcy law to declare those cross-licenses “no longer enforceable”.  Some licensees responded that they would invoke protections under Section 365(n) of the U.S. bankruptcy code that allows licensees to retain their rights under existing licenses.

The U.S. bankruptcy and district courts were then asked to balance the rights of parties with the United States’ foreign comity obligations.  The district court ultimately ruled that it would preserve the licensee’s Section 365(n) rights as to U.S. patents, which ruling the Fourth Circuit affirmed in its ruling yesterday.

Holdup.  The court explained a “holdup” concern based, not on SEPs, but infringement assertions of patents generally after “sunk costs” in developing a semiconductor product:

The problem of the patent thicket is exacerbated by the enormous costs incurred to bring a new semiconductor product to market.  According to one expert, the price of building a new semiconductor fabrication facility can now exceed $5 billion.  These sunk costs could create a classic “holdup” problem if a new product were ultimately found to infringe someone else’s patent, with the patent’s owner being able to extract a substantially higher royalty after the investment had been made than if a license had been negotiated beforehand. Thus, to avoid this holdup premium and enhance their design freedom, competitors in the semiconductor industry have routinely entered into broad, non-exclusive cross-license agreements with each other, sometimes with the addition of equalizing payments (either up-front payments or so-called running royalties) to account for differences in the size and breadth of the respective patent portfolios.

 RAND.  The Qimonda bankruptcy trustee explained that revocation of existing licenses would include a promise to re-license the patents under a reasonable and non-discriminatory royalty (RAND), to enter good faith negotiations and bring any RAND rate dispute into arbitration proceedings before the World Intellectual Property Organization (WIPO).  The court likened this RAND commitment to those common in technical standard setting organizations (SSOs).  In this instance, the licensor would seek to convert the existing non-monetary cross-licensing terms into a pure monetary RAND rate estimated to give the Qimonda estate about $47 million a year.  In contrast, the existing non-monetary cross-licenses had little value to the bankrupt Qimonda estate since it was no longer in the semiconductor production business.

The licensees responded that, among other things, terminating the existing cross-licenses would “destabiliz[e] the system of licensing that has enabled the extraordinary success of the semiconductor industry” and would have further uncertainty if the patents were sold to another entity that itself goes bankrupt or otherwise does not honor the existing license agreements.  The district court weighed the issues and ruled that the licensees could rely on Section 365(n) to preserve the existing licenses to U.S. patents, and the Fourth Circuit affirmed that ruling as “reasonable.”

The issue of preserving licensing rights through bankruptcy has arisen with standard essential patents (SEPs) as well (see our Rockstar post), though this case illustrates the issue is not unique to SEPs.  This decision does not resolve the issue.  The concurring opinion notes that the decision is only whether the district court’s ruling was a “reasonable” exercise of discretion and not whether the court would reach the same conclusion.  Further, the U.S. government filed an amicus brief in this case seeking to reverse the preservation of licensing rights “on the threshold ground that Section 365(n) cannot constrain the operation of German insolvency law in Germany.”  In contrast, current patent reform legislation seeks to confirm that licensing rights are preserved through bankruptcy (see our November post).

Yesterday Adaptix moved to withdraw its complaint and terminate the ITC investigation of Ericsson for infringing Adaptix’s patent alleged to be essential to LTE standards practiced by Ericsson base stations.  The motion does not explain the reason for withdrawal and comes at a time when the parties have been in heavy pre-trial procedures and at the eve of the preliminary conference and evidentiary hearing.  The move does not appear to be based on any settlement, the motion papers saying that “[t]here are no agreements, written or oral, express or implied between Adaptix and Respondents [Ericsson] concerning the subject matter of the Investigation.”  Ericsson’s response yesterday indicates this is some new development that its counsel has not even had time to discuss with its client, stating:

For the first time late last night, Adaptix requested Ericsson’s position on its motion to terminate the investigation based on withdrawal of the complaint, coupled with a suspension of the procedural schedule.  This motion arises in the context of issues that have not been disclosed to the Administrative Law Judge and the undersigned counsel for Ericsson have not been able to reach the senior Ericsson counsel, who are on an overnight flight from Korea to the United States.  Accordingly, the undersigned counsel for Ericsson have not yet been able to obtain Ericsson’s definitive instructions on how to proceed.

In January, Adaptix (a subsdiary of Acacia Research) filed its 337 Complaint alleging that Ericsson’s 4G LTE base stations infringe U.S. Pat. No. 6,870,808, which also had been asserted in several district court litigations (see our January post).  Adaptix claimed that it had not participated in any LTE standard setting process and the patent was under no FRAND or other standard setting organization (SSO) obligations.  The ITC instituted the investigation in February (see our February post).

Some have speculated that Samsung was behind Adaptix’s assertions against Ericsson in retaliation for Ericsson’s enforcement of SEPs against Samsung (see our January post).  Ericsson also had tried to assert in the instant investigation FRAND-based defenses through a novel theory that Samsung was a licensee of the patent-in-suit, participated in ETSI’s development of the LTE standard and failed to disclose the patent to ETSI in breach of ETSI’s IPR Policy (see our July post).

Hopefully we will find out soon what was behind Adaptix’s sudden move to terminate this investigation.  Though this may be unrelated, the move comes at an interesting time when:

  • The European Commission is considering Samsung’s recent proposal not to seek injunctive relief in the European Economic Area for the next five years on its wireless SEPs (see our October post)
  • The Competition Commission of India started its investigation of Ericsson’s licensing of cellular SEPs (see our November post)
  • The ITC’s pending investigation of Samsung’s infringement of Ericsson SEPs includes Ericsson’s suggestion that any exclusion order include an opportunity for Samsung to avoid exclusion by accepting an ITC determined FRAND royalty rate (see our August 15 and August 26 posts).
  • The asserted ‘808 Patent still remains at issue in several coordinated cases that Adaptix filed against Ericsson and other cellular companies pending in E.D. Texas (currently in the claim construction briefing stage).

UPDATE:  Administrative Law Judge Gildea issued an Order that stays the procedural schedule pending the motion to terminate and requires responses to the motion to be filed by Wednesday, December 11.

A few months ago, we posted about patent infringement suits filed by Zenith Electronics (“Zenith”), Panasonic Corporation (“Panasonic”), U.S. Philips Corporation (“Philips”) and The Trustees of Columbia University in the City of New York (“Columbia”), licensors to the MPEG LA Advanced Television Systems Committee (“ATSC”) digital television patent pool, against electronics makers Curtis International (“Curtis”), Viewsonic Corporation (“Viewsonic”), and Craig Electronics (“Craig”).  The complaints accuse Curtis, Viewsonic and Craig, of infringing four alleged standard essential patents (“SEPs”) through their making/selling/using televisions that comply with the ATSC Standards.

This past Friday, November 22, 2013, defendant Viewsonic filed its Answer, which included FRAND-related affirmatives defenses and counterclaims against Zenith, Panasonic and Philips, as well as a FRAND-related Third-Party claim against MPEG LA. 

Implied License Defense.  Viewsonic’s Twelfth Affirmative Defense is “Implied License,” and alleges that:

On information and belief, Plaintiffs’ claims for recovery are barred, in whole or in part, by the doctrine of express license and/or implied license for reasons including, without limitation, that Plaintiffs are barred from asserting claims of infringement of one or more of the Asserted Patents against ViewSonic under the MPEG LA ATSC Patent Portfolio License and the MPEG LA Agreement Among Licensors.

FRAND Breach Defense.  As its Eighteenth Affirmative Defense, Viewsonic asserts that Plaintiffs’ claims are barred by their breaches of alleged FRAND obligations to Viewsonic:  “Plaintiffs do not offer  licenses to the Asserted Patents on fair, reasonable and nondiscriminatory terms and therefore the Plaintiffs are not entitled to any recovery.” 

But in a pleading that may be counter to having a FRAND obligation, Viewsonic’s Seventeenth Affirmative Defense directly challenges the alleged essentiality of Plaintiffs’ patents:  “On information and belief, one or more of the Asserted Patents is not essential to the ATSC standard.”

 Sherman Act Section 2 Violation.  Viewsonic also asserts four counterclaims involving FRAND/SEP-issues.  In its first counterclaim, Viewsonic alleges that Zenith and Philips violated Section 2 of the federal Sherman antitrust Act by failing to offer a license to their alleged SEPs on FRAND terms.  Specifically, Viewsonic asserts that Zenith and Philips falsely represented to the ATSC that they would license their patents on FRAND terms if their technology was adopted into the ATSC standard.  Then, “having succeeded in using false promises . . .  to get their technology incorporated into the ATSC standard, they now seek to utilize the patents allegedly covering the technology to raise prices [for consumers] and exclude competition” from entities that are willing to accept a license on FRAND terms.  “Unless [the] Court declares that the conduct of Zenith and Philips has violated the antitrust laws, there exists a dangerous probability that they will succeed in monopolizing the relevant market” for products that are ATSC-compliant and which infringe Zenith and Philips’ alleged SEPs. 

 Sherman Act Section 1 Violation.  Viewsonic’s second counterclaim also brings a third-party claim against MPEG LA, and asserts that Zenith, Panasonic and Philips – “horizontal competitors of Viewsonic” – violated Section 1 of the federal Sherman antitrust Act by conspiring “with MPEG LA to reduce or eliminate competition from low price-tier television sellers such as Viewsonic,” specifically by agreeing “not to compete on pricing for the royalty rate for the ATSC patent pool” as well as agreeing to set “a non-FRAND royalty rate for the pooled patents.”  According to Viewsonic, these entities “have used the cover of the ATSC Patent Pool” to further unreasonably restrain trade by, inter alia: 1) “engaging in patent hold up;” (2) “demanding royalty rates that are excessive in light of the fact that some patents in the ATSC Pool have [expired] or will soon expire;” (3) “demanding a royalty rate for products sold outside the statutory period of limitations for infringement claims;” (4) “demanding duplicative royalties for the same products on which royalties have already been paid or are already due from third parties in violation of the patent exhaustion doctrine;” and (5) demanding that Viewsonic “license the ATSC Pool and/or each individual Plaintiffs’ ATSC patent portfolio, including non-US patents, despite failing to demonstrate essentiality of all patents and despite having inadequate safeguards to ensure that the pool is limited to patents actually essential to the ATSC standard.”

FRAND Breach.  Viewsonic’s third counterclaim alleges that Zenith, Panasonic and Philips each breached their agreement with the ATSC by “failing to license their allegedly standard essential patents on FRAND terms” to Viewsonic. In support of this claim, Viewsonic alleges that these entities’ “demands for unreasonable royalties and/or failure to license their allegedly essential patents on FRAND terms was based on their failure to consider” a number of factors, including, but not limited to the fact that the “patents in the ATSC Pool have [expired] or will soon expire, and the royalty rates should decrease as patents in the pool expire.”

Promissory Estoppel.  Viewsonic’s fourth counterclaim is for promissory estoppel, and alleges that, as a result of Plaintiffs’ promises to the ATSC to license their patents on FRAND terms, Viewsonic “invested substantial resources developing, marketing, and selling products that are alleged to utilize the ATSC Standard, with the expectation that [Plaintiffs] would comply with their promises and offer licenses to any essential patents on FRAND terms.”  According to Viewsonic, “Plaintiffs are estopped from defaulting on these promises under the doctrine of promissory estoppel.”

In its Prayer for Relief, Viewsonic requests that the court enter an order “[c]ompelling Plaintiffs to offer licenses to the Patents in Suit on FRAND terms” and to also award it treble damages for Plaintiffs’ alleged violations of the Sherman Act.

What’s Next?

Zenith, Panasonic, Philips and MPEG LA are obligated to file answers to Viewsoinc’s counterclaims/third-party claim and/or file a motion to dismiss.  We will continue to track the case for further developments.

In an order dated November 12, 2013, the Competition Commission of India ordered an investigation into Ericsson’s licensing of cellular patents that are subject to FRAND obligations for certain ETSI standards.  This investigation is based on information provided by Micromax Informatics Limited (“the Informant”)  that had been approached by Ericsson (“the Opposite Party” or “OP”) to take a license under the patents, followed by infringement litigation on those patents.

The order provides a background of the negotiations and litigation between the parties, including Ericsson’s requested royalty rate, an interim royalty agreement and mediation between the parties.  Such mediation did not succeed, which the order indicates was based in part on Ericsson’s refusal to reveal licensing agreement terms under the patents with similarly situated parties.

The order touches on several issues, many of which have been considered in U.S.-based standard essential patent disputes.

SEPs.  The order describes standardization as “a voluntary process” between market players for “developing and implementing technical standards.”  And “[s]uch technological standards are termed as Standard Essential Patent, when they are patented and for which there are no non-infringing alternatives.”  Thus, “[o]nce a patent is declared as Standard Essential Patent, it faces no competition from other patents until that patent becomes obsolete due to new technology/inventions.”

Patent Hold-Up/Royalty-Stacking.  The order expresses concern about patent “hold-up” and “royalty-stacking”, but does not provide an analysis at this point of how much those concerns come into play here (U.S. courts have raised these concerns with varying views of their actual verses theoretical impact):

When … standard technologies are protected by patent rights, there is a possibility of “hold-up” by the patent owner–a demand for higher royalties or more costly or burdensome licensing terms than could have been obtained before the standard was chosen.  Hold-up can subvert the competitive process of choosing among technologies and undermine the integrity of standard-setting activities.  Ultimately, the High costs of such patents get transferred to the final consumers.  Similarly, royalty-stacking is when a single product uses many patents, of same or different licencors.  As such, from the perspective of the firm making the product, all the different claims for royalties must be added or “stacked” together to determine the total burden to be borne by the manufacturer.

Discrimination/Royalty Base.  The order indicates concern that Ericsson may have violated its FRAND obligations–particularly the “non-discriminatory” requirement–by using the selling price of the entire GSM-capable phone as the royalty base, rather than the price of the GSM chip (an issue that U.S. courts deal with concerning what constitutes the smallest-salable patent practicing unit), stating:

The allegations made in the information and not refuted by OP [Ericsson] concerning royalty rates make it clear that the practices adopted by the OP were discriminatory as well as contrary to FRAND terms.  The royalty rates being charged by the OP had no linkage to patented product, contrary to what is expected from a patent owner holding licences on FRAND terms.  The OP seemed to be acting contrary to the FRAND terms by imposing royalties linked with cost of product of user for its patents.  Refusal of OP to share commercial terms of FRAND licenses with licensees similarly placed to the informant, fortified accusations of the Informant, regarding discriminatory commercial terms imposed by the OP.  For the use of GSM chip in a phone costing Rs. 100, royalty would be Rs. 125 but if this GSM chip is used in a phone of Rs. 1000, royalty would be Rs 12.5.  Thus increase in the royalty for patent holder is without any contribution to the product of the licensee.  Higher cost of a smartphone is due to various other softwares/technical facilities and applications provided by the manufacturer/licensee for which he had to pay royalties/charges to other patent holders/patent developers.  Charging of two different license fees per unit phone for use of the same technology prima facie is discriminatory and also reflects excessive pricing vis-a-vis high cost phones.

The order makes clear that the forgoing were just initial “observations” that should not “sway” the investigation: “Nothing stated in this order shall tantamount to a final expression of opinion on merit of the case and the DG shall conduct the investigation without being swayed in any manner whatsoever by the observations made herein.”

An investigation and report thereon is to be provided within 60 days of the order.

Our prior posts discussed Complainant LSI’s comments and respondent Realtek’s comments in the ITC’s investigation of whether Realtek and Funai infringe LSI’s alleged standard essential patents (SEPs).  These comments were submitted in response to the Commission’s request for information on various issues to aid in its review of the ALJ’s conclusion that Realtek and Funai did not infringe LSI’s SEPs.  Like LSI and Realtek, respondent Funai also submitted comments on merits issues as well as comments on public interest and remedy, both of which addressed various SEP-related issues.

Funai’s Comments on the Merits

            Induced Infringement

Funai contends that, in order for LSI to prove induced infringement of its alleged SEPs, LSI must show both direct infringement by a third-party as well as “specific intent to cause [such] infringement” by Funai.  But (1) the ALJ found that LSI failed to prove infringement based on the claims covering the H.264 standard, (2) LSI’s attempt to show intent based on LSI’s marketing its products as complying with H.264 failed to show “that Funai actually believed that the H.264 processes infringed the ‘663 patent”, and (3) LSI did not show direct infringement because its expert did not read the ‘663 patent claims onto H.264 and simply assumed that users actually performed the H.264 standard even though they could use Funai’s products without practicing that standard.

           Invalidity and Priority Date of the ‘958 patent

Funai also asserts that the ‘958 patent is invalid over a proposal made by another entity to the IEEE when developing the 802.11 standard:

“The Harris Proposal was presented to the IEEE Working Group [for the 802.11 standard] and made publicly available at least as early as November 10, 1997. …  Accordingly, to the extent that the asserted claims are entitled to a priority date of 1998 and not 1996, the Harris Proposal is prior art to the asserted claims. …  [The Harris Proposal] discloses the same digital modulation system that formed the basis for the ‘958 patent.  Accordingly, the asserted limitations of the ‘958 patents relating to the functional elements of the claimed modulation systems … are present in the Harris Proposal.”

         Funai’s Comments on the Alleged Standard-Essential Nature of LSI’s Alleged SEPs

LSI asserts that “the ‘958 and ‘867 patents are infringed simply by practicing different aspects of the IEEE 802.11 standard” and that “the ‘663 patent is infringed simply by practicing the H.264 standard.”  But, according to Funai, “the [ALJ] correctly recognized that practicing the 802.11 or the H.264 standards will not necessarily result in the infringement of the asserted claims.”  In fact, Funai argues that “the evidence of record does not support the ‘standard essential nature’ of the ‘663, the ‘958, and the ‘867 patents.”  “If, however, these patents are found to be essential to the standards, LSI has failed to satisfy its obligations to license those patents on fair, reasonable and nondiscriminatory (FRAND) terms.”

          The Domestic Industry Requirement

Funai argues that there is no evidence that “any licensee actually practices any of the asserted patents.”  LSI’s “only attempt to satisfy this requirement was based on the wholly unsupported assertion that [a] licensee’s . . . 802.11 compliant products are ‘more likely than not’ and ‘highly likely’ to practice the ‘958 and ‘867” patents, which the ALJ properly rejected.  Thus, LSI  “failed to establish that their licensing investments are related to any articles protected by the asserted patents, which is a critical statutory requirement for any domestic industry finding.”

Funai Comments on Public Interest and Remedy

Funai argues that LSI’s RAND commitments judicially estopps LSI from walking away from such commitments and prevents entry of an exclusion order.  Specifically, LSI has “signed FRAND commitments with the ITU (for the ‘663 patent) and with the IEEE (for the ‘867 and ‘958 patents) and have used the alleged standard essential characterization of these three patents to its benefit.”  Further, the public interest is not served by an exclusion order because “LSI never made an offer that complied with its RAND obligations, and Funai is prepared to accept a FRAND offer.” 

          RAND-Encumbrances on the Asserted Patents

Funai argues that LSI has “concede[d] that it has RAND obligations in connection with the ‘958 and ‘867 patents.”  The appendix to LSI’s letters of assurances to the IEEE includes both the ‘958 patent and the patent application that would issue as the ‘867 patent.  With respect to the ‘663 patent, LSI’s predecessor-in-interest “made a licensing declaration to the ITU” that “obligated it to ‘grant a license to an unrestricted number of applicants on a worldwide, nondiscriminatory basis and on reasonable terms and conditions to use the patented material necessary in order to manufacturer, use, and/or sell implementations” of H.264-compliant products. 

          History of Negotiations

Funai asserts that it “negotiated in good faith with LSI . . . before LSI filed this investigation in March 2012, and has continued to negotiate even during the pendency of this investigation.”  “Yet after [redacted] of zigging and zagging positions by [LSI], Funai is still awaiting an offer that could be considered ‘fair and reasonable’ under any standard.”  The parties could not reach a license agreement after having met at least thirteen (13) different times and exchanging several pieces of correspondence and presentations.     

          Licenses to LSI’s Alleged SEPs

Funai asserts that “[t]he licenses identified by LSI are of minimal value in determining a RAND rate for the ‘663, ‘867 and ‘958 patents” because many, if not all, of those licenses included patents other than the three (3) at issue in this Investigation.  LSI’s expert apparently attempted to “disaggregate the value of the ‘663, ‘867 and ‘958 patents from LSI’s” other licenses but this analysis, according to Funai, “is replete with problems.” 

          Industry Practice for Licensing Similar Technologies

Funai asserts that two patent pools are relevant here: MPEG LA and the Via Licensing Pool.  MPEG LA manages a pool of H.264 patents submitted by many companies, and  “licenses approximately 275 U.S. patents and over 2400 total patents to almost 1300 companies, including Funai.”  Funai argues that the royalty rate from the patent pool (which is redacted) is comparable to the rate for LSI’s alleged SEPs.  Unlike Realtek, Funai argues that the Via Licensing Pool – which “contains over 250 patents. . . licensed at rates between $0.05-$0.55 cents per unit for the entire pool” – provides evidence of a RAND rate for patents essential to the H.264 and 802.11 standards.

          Forums in Which a RAND rate has been Established

Funai asserts that it “has not at this time sought affirmative determination of a RAND rate.”  But Funai asserts that any RAND rate determined for the ‘867 and ‘958 patents in Realtek’s breach of contract action against LSI in the N.D. Cal. “would apply to any Funai products that practice the 802.11 standard using non-Realtek integrated circuits as well.” 

Funai cites to the decisions in Microsoft and Innovatio to support its asserted RAND royalty rate (which rate is redacted).  Funai relies on Microsoft’s holdings that, “for a portfolio of H.264 patents, a RAND rate was between 0.555-16.389 cents” and, for patents essential to the 802.11 standard, a RAND rate was between “0.8-19.5 cents.”  Funai also cites to Innovatio’s holding setting a RAND of approximately “0.5 cents per patent average rate” as support for its proffered RAND rate.

          Alleged Abuse by LSI of Their RAND-Encumbered Patents

Funai argues that LSI has “in several ways attempted to gain undue leverage over or constructively refuse[d] to negotiate a license with Funai” with respect to LSI’s alleged SEPs. 

Holdup.  First, LSI filed the ITC investigation prior to negotiating a license setting a RAND royalty rate.  LSI waived its right to seek an exclusion order from the Commission “[w]hen LSI declared its patents essential to a standard.”  “Because money [through a RAND license] can compensate LSI fully for any purported infringement, injunctive relief should be unavailable to LSI.”  But “by coming to the Commission, LSI is trying ‘to extract holdup value by exercising market power that it would not have had absent the inclusion of its technology into the standards.’”

Royalty Stacking.  Second, LSI “does not account for the scores of patent holders that also are entitled to RAND rates for their intellectual property.”  According to Funai, “[t]his results in royalty stacking, which effectively overvalues the standard-essential patents, a phenomenon that RAND policies are designed to avoid.”  For the 802.11 standard alone, “[n]inety-two companies have submitted letters of assurance for patents covering [that] standard.”  Forty-eight companies have declared patents essential to the H.264 standard.  “If every holder of a RAND-encumbered patent for these standards was entitled to the royalty rate demanded by LSI, the total licensing fees for a client would be cost-prohibitively high.”

         An Exclusion Order Should Not Issue

Funai argues that the ALJ was wrong in concluding that an exclusion order should issue if the Commission finds infringement of LSI’s alleged SEPs.  The exclusion order requested by LSI would apply to Funai’s downstream products that incorporate the allegedly infringing chips.  However, “excluding downstream products potentially expands the scope of an exclusion order and increases the risk of interfering with legitimate commerce.”  The ALJ and the Commission are required to consider the value of the infringing article compared to the value of the downstream products into which they are incorporated.  The higher the value, the more likely an exclusion order including the downstream products will issue.  Relying on its expert’s testimony, Funai contends that, contrary to the conclusion of the ALJ, “the value of the allegedly infringing integrated circuits remains miniscule compared to the value of the Funai Downstream Products.”  This weighs against the issuance of an exclusion order.

         Funai Comments on Bond

Funai argues that the ALJ properly rejected LSI’s contention that it is entitled to a 100% bond.  According to Funai, LSI “offered no evidence in their post-hearing brief concerning the proper bond (likely because such evidence would have undermined its argument that its offer to Funai complied with RAND).”  “Because [LSI] is required to affirmatively prove bond, in the absence of evidence from [LSI], the proper bond value is zero.”

Should the Commission determine that an exclusion order should issue and a bond is required during the Presidential review period, Funai contends that “the bond should be no more than 0.555-16.389 cents for the ‘867 and ‘958 patents, and no more than 0.8-9.5 cents for the ‘663 patent, as found in the Microsoft” case.  Funai contends that these ranges are also consistent with the RAND rate set by Judge Holdermann in the Innovatio case.

What’s Next?  On November 14, 2013, the parties filed reply comments with the ITC, but those comments were filed under seal.  The public versions of the reply comments were recently made available and will be the subject of a separate post. 

We previously discussed the comments filed by complainant LSI in the International Trade Commission (ITC) investigation of whether Realtek and Funai infringe LSI’s alleged 802.11 and H.264 standard essential patents (SEPs).  The ALJ’s initial determination found the SEP patents were not infringed but otherwise rejected RAND-based defenses.  The Commission then decided to review the ALJ’s determination in its entirety and, as part of the review, requested comments from the parties and third-parties on various issues, including RAND obligations, the history of license negotiations among the parties, other licenses for the patents-in-suit, and industry practice for licensing similar technologies.

A summary of the comments filed by respondent Realtek is provided below.

Realtek’s Comments on the Merits

In its merits comments, Realtek addresses LSI’s contention that its ‘958 patent, an alleged SEP,  is entitled to an earlier priority date of a parent patent based on the technology disclosed in the parent being incorporated into one portion of the 802.11 standard.  Realtek argues that “[a]fter [LSI’s predecessor-in-interest] filed the ‘958 patent application, the [IEEE] adopted the 802.11b standard, which refers to a formula in the ‘182 patent,” the parent patent.  “In 2002, Complainants amended the ‘958 patent application to incorporate the [‘182 patent application] and claim priority to the 1996 date … to try to make the ‘958 patent relevant to the new standard.”  According to Realtek, “[t]hat desire is not enough to justify a new filing date.”

Realtek’s Comments on the Public Interest and Remedy

          Alleged RAND-Encumbered Nature of LSI’s Alleged SEPs

Realtek  contends that, as a result of LSI’s declarations to the IEEE, LSI is obligated “to offer Realtek licenses to both the ‘867 and ‘[958] patents on reasonable, and non-discriminatory (RAND) terms and conditions” and also “have a duty to negotiate in good faith with Realtek.” 

Citing to the USTR’s recent disapproval of the exclusion order in Samsung v. Apple, Realtek argues that “the use of patents in standards may allow patent holders to gain undue leverage and engage in patent hold-up to extract excessive royalties from implementers of standards or exclude them altogether.”  “Such conduct harms competition and consumers by ultimately raising prices for products.”  Thus “standards organizations such as the IEEE…require that patent holders commit to license patents essential to a standard on RAND terms” in order “[t]o address the concerns over patent hold-ups.”

Contrary to LSI’s contentions, Realtek argues that the first license offer made by a holder of alleged SEPs must be on RAND terms.  “[A] patent holder cannot satisfy its RAND obligation simply by making any offer.  It must be a good faith offer that is not unreasonable, because an unreasonable offer violates a patent holder’s RAND obligations.” (citing Microsoft v. Motorola ).

History of the License Negotiations Between LSI and Realtek

As with LSI’s comments, most of the license negotiations comments are redacted.  An initial negotiation apparently occurred in 2002-2003 (nine (9) years before LSI filed its complaint with the ITC that initiated the investigation) and a second negotiation beginning in 2012.  Early on, Realtek apparently requested a “detailed claim analysis” showing LSI’s infringement contentions.  In response, LSI “suddenly terminate[d] discussions with Realtek,” which Realtek interpreted to mean that LSI was “abandon[ing] its assertions against Realtek.”  “After the 2002-2003 discussions, there was no further contact between [LSI] and Realtek until March 12, 2012, the day [LSI] filed [its] Complaint in this Investigation, when Realtek received a [redacted] letter from [LSI].”  The parties further exchanged correspondence, but could not reach a deal.  Realtek asserts that LSI “remain[s] indifferent to [its] RAND obligations to Realtek up to and including their most recent licensing demand.”

         LSI Licensing of Its Alleged SEPs

Nearly all of Realtek’s summary of LSI’s SEP licenses to other entities is redacted.  Realtek provided a comparison of LSI’s offers to Realtek and to other licensees, also redacted.  Realtek contends that this comparison shows that LSI’s offer to Realtek violate LSI’s RAND obligations:  “[W]hen [LSI’s] offer to Realtek is compared to the licenses . . . that [LSI] ha[s] successfully negotiated, the unreasonableness of [LSI’s] offer becomes even more apparent.” 

          Licenses to Patents Involving Similar Technologies

Realtek argues that “[t]he record does not include sufficient information to establish whether there is an ‘industry practice’ for licensing patents ‘involving technologies similar to the technologies in the ‘958 and ‘867 patents.” 

Realtek asserts that the recent opinion in Innovatio sets “a royalty rate of less than a dime for each Wi-Fi chip sold for the patentee’s entire Wi-Fi portfolio for patents that were of moderate to moderate-high importance to the standard.”  Realtek argues further that the court in Microsoft “set a royalty of 3.471 cents per unit on Motorola’s entire 802.11 standard essential portfolio of patents that were of very minimal value to the standard.”  Relying on these two decisions, Realtek argues that “[a]ny royalty rate on an individual patent within [LSI’s] portfolio would presumably be far less and would need to factor in, among other things, the value of the patent to the standard.” 

         Forums in Which a RAND rate Has Been Established

No RAND rate has been adjudicated for these patents, though Realtek is seeking “a determination of a RAND rate for the ‘958 and ‘867 patents . . . in [its] Northern District of California” case against LSI.  The trial to determine the RAND rate is set for February 14, 2014.

Realtek argues that “[t]he overall value of the 802.11 standard must be divided among all standard essential patent holders resulting in a necessarily low value of the average standard essential patent.”  Citing the Microsoft and Innovatio decisions, Realtek argues that “[u]nless it is shown that a patented technology made significant technical contribution that had no reasonable or viable alternatives, the presumption should be that the value of that patent is low.”

Realtek contends that the record evidence demonstrates that “the ‘958 and ‘867 patents necessarily comprise, at best, a small fraction of the overall value of the 802.11 standard.”  This is because “there are estimated to be over 3,000 patents related to the 802.11 standard” and “the IEEE 802.11-2012 specification . . . includes very little discussion of either of those technologies.”  Realtek also argues that “based on the number of letters of assurance (LOAs) submitted to the IEEE for the 802.11 standard, a conservative estimate is that there are at least 203 declared essential patents and 46 then pending patent applications for the 802.11.”  Therefore, “at best, [LSI’s] patents represent only a tiny fraction of the 802.11 standard essential patents and do not represent a significant portion of the overall value of the 802.11 standard.”

Realtek argues that the ex-ante value of the ‘958 patent to the 802.11 standard is low because several alternatives, including at least one with better performance, were available at the time the relevant 802.11b standard was being developed.  Realtek contends that the ex-ante value of the ‘867 patent is low because alternatives existed at the time the 802.11b standard was being developed and the portion of the standard to which the ‘867 patent is relevant only applies to an optional feature on a small subset of products. 

Realtek then cites to expert testimony in its Northern District of California case against LSI on the appropriate RAND rate to argue that “[c]onsideration of the Georgia-Pacific factors indicates that the outcome of the hypothetical negotiation would have been pushed toward the lower end of the bargaining range”  because of the availability of alternative technologies. 

Relyin upon Judge Robart’s opinion in Microsoft, Realtek argues that LSI’s expert’s reliance upon the Via Patent pool as a comparable is misplaced given Judge Robart’s conclusion that there are “numerous flaws inherent in the Via Pool,” including the fact that it has “not been very successful in attracting licensors or licensees.”  “Given the poor participation in the pool, it has not achieved a primary purpose of RAND commitments – to encourage widespread adoption of the 802.11 standard – and thus has lower relevance as an indicator of a RAND rate.”

         Undue Leverage or Constructive Refusal to Negotiate a License

Realtek argues that LSI’s “conduct is the very definition of seeking undue leverage and reveals that they constructively refused to negotiate with Realtek.”  Specifically, Realtek contends that LSI “attempted to gain undue leverage and engaged in patent hold up against Realtek by failing to offer Realtek licenses to [the] ‘958 and ‘867 patents on RAND terms before initiating this investigation and negotiate a license in good faith with Realtek even after this investigation began.”

Realtek argues that, if an exclusion order is issued in this case, “then RAND obligations would be rendered meaningless.”  Failing to require SEP owners to make a RAND offer before initiating litigation or to negotiate in good faith afterwards would leave SEP owners “free to extract higher royalty rates with the threat of an exclusion order or discriminate against certain licensees by never granting a license before their products are excluded from the United States.”  Realtek asserts that this result “would harm the public by suppressing competition and ultimately rais[ing] prices for consumer products.”

Realtek also argues that the LSI license negotiations were not really negotiations at all.  Realtek cites to Judge Whyte’s summary judgment ruling in the N.D. Cal. action that held that LSI’s initial correspondence to Realtek in 2002-2003 did not even contain an offer.  “By initiating this investigation and seeking injunctive relief before even making a license offer to Realtek, [LSI] imposed undue pressure on Realtek before the parties even started license negotiations.”  This is the type of undue leverage “that RAND terms were created to prevent and that the USTR, DOJ, USPTO, and district courts” have warned against.  Thus, quoting Judge Whyte,  “‘[LSI’s] conduct [against Realtek] is a clear attempt to gain leverage in future licensing negotiations and is improper.’”

Realtek also asserts that LSI constructively refused to negotiate with Realtek because it “never intended to give Realtek any legitimate offer to license the ‘958 and ‘867 patents, let alone an offer that reflects a true RAND royalty rate for” those patents.  The rates proposed by LSI “would far exceed the revenues Realtek receives from the sale of its products” and “would end Realtek’s business altogether.” 

                        Realtek Proposed Bond

Realtek asserts that any bond should be set at zero since LSI does not make any competing products nor does its licensees.  Realtek also argues that LSI’s bond request must be rejected because LSI has “only asserted against Realtek patents that they declared standard essential and committed to licensing on RAND terms” and LSI’s  “failure to enter any evidence of a reasonable royalty or to claim they are unable to determine a reasonable royalty is disingenuous, at best.”  “I[t] further undercuts any arguments they may make that they have negotiated with Realtek on a good faith basis or that they have offered Realtek RAND terms.”

Realtek also argues that any bond should not exceed a RAND rate, because any more would give LSI amounts in excess of what they agreed to take under their RAND obligations.

What’s Next?  Respondent Funai’s comments will be the subject of a separate post.  Also, on November 14, 2013, the parties filed reply comments with the ITC, but those comments were filed under seal.  The public versions of the reply comments were just made available and will be the subject of a separate post.