FOSS Patents

Patent pool administrator Sisvel just announced that Huawei, one of the largest patent holders in this space, has joined the narrowband Internet of Things (IoT) patent pool that was launched last November. The largest founding licensor was Ericsson.

This is the second Sisvel pool for Huawei to join. More than a year ago, Huawei started its working relationship with Sisvel as an initial licensor of its WiFi 6 pool (as Sisvel president Mattia Fogliacco recalls in today's press release) and is also an Avanci 5G licensor, which shows that the Chinese innovator is increasingly receptive to pool-based licensing solutions. Other major SEP holders may now be more interested in joining this pool, given that Huawei and Ericsson (as well as various other patent holders) already give the licensing program a lot of substance.

Sisvel's updated list of NB-IoT licensors contains several other names I haven't previously noticed, such as KPN, Deutsche Telekom, and BlackBerry. What licensees get is a "one-stop shop" (as Sisvel's cellular IoT program manager Sven Torringer calls it in today's press release) that gives them access to more than two dozen patent portfolios, with Huawei, Ericsson, and NTT Docomo being considered particularly strong in narrowband IoT. It's fair to say that Huawei had a leadership role in the development of the IoT-related parts of the 4G/LTE standard.

Sisvel's press release mentions "new royalty rates for the Cellular IoT patent pool, including for devices with a lower selling price," and that it has "expanded its offering to new product verticals." Previously, as I reported at the time, the rate was "0ドル.66 per unit for LTE-M" and "a distinction [was] made between asset trackers (1ドル.33 per unit) and smart meters (2ドル per unit)." The 2ドル (smart meters) and 1ドル.33 (smart sensors with a selling price above 20ドル and up to 130ドル) price points are still found. But now there is also a 0ドル.35 per-unit royalty rate for smart sensors with a selling price above 6ドル (up to 20ドル) and a 0ドル.08 per-unit rate for devices with a selling price of 6ドル or less.

For makers of low-priced IoT products, this means the value proposition has changed enormously in their favor: far more patents, and far lower rates for certain applications. It looks like two factors have resulted in this more flexible and attractive royalty structure:

• In my commentary on the creation of the pool I already noted that the challenge for an IoT patent pool is market development: it's not enough to offer licenses or to dissuade implementer from infringement, but about adoption of the standard. Patent pool administrators operate in a two-sided market. They have to bring licensors and licensees together, and in order to do so, they have to listen to both sides.

• Sisvel's new NB-IoT royalty rates are definitely reconcilable with Huawei's bilateral licensing terms for such products, which are highly differentiated as I explained last month. The pool is a one-stop option, but it's not the only way to get access to Huawei's intellectual property.

  Huawei is both a major patent holder and a large-scale implementer. If a pool worked for only one side, Huawei would find it hard or even impossible to join.

The rock-bottom rates that the pool now offers for devices with a selling price of 6ドル or less should enable new applications. What comes to mind as potentially the highest-volume and lowest-price NB-IoT application is called printable NB-IoT tracking labels. They are asset trackers in the form of stickers that could, for instance, track an Amazon package.

At a time when policy makers are working on an EU SEP Regulation, it's warranted to put major SEP news such as this one into the current political context. The European Commission's Directorate-General for the Internal Market (DG GROW) clearly underestimated the extent to which patent pools could be part of the solution as opposed to being part of the problem. The favorite pretext (not only of DG GROW but also Apple and its allies and astroturfers) to push for legislative intervention is that IoT SMEs allegedly need a different legal environment. But patent holders are smart enough to realize that IoT patent licensing won't work unless the royalty rates enable IoT product makers to thrive. While I still haven't seen a single SEP enforcement action against an SME as defined by the EU, and the IoT sector is generally not a legal battlefield at the moment, the market continues to find and improve solutions.

Patent pools not only bring licensors together with licensees, but also have to broker a compromise between licensors of very different kinds. Huawei with its high product volume (which is not fully visible to people in the Western hemisphere for purely geopolitical reasons, but is a reality in the rest of the world) obviously has a more balanced take than a non-practicing entity, which is not meant to disparage NPEs but plainly a fact. In order for a pool to unite product makers like Huawei and Ericsson with infrastructure companies like NTT Docomo and Deutsche Telekom as well as with research institutes and patent licensing firms, it has to identify royalty rates that work for all of them.

An EUIPO-led process for aggregate (entire standard) and bilateral (licensor A and licensee B) royalty determinations will cause delay and complicate matters. In the meantime, patent pool administrators and other market actors work out and fine-tune the solutions they offer.

With so much in flux especially concerning IoT, the prudent thing for the EU to do would be to wait and better understand what's going on. Instead of trying to adopt something before the end of the legislative term just for the sake of having some kind of outcome to show (no matter how flawed), they should go back to the drawing board and take note of new developments, such as IoT licensing terms becoming more attractive at a breathtaking pace.

When I reported on Huawei's annual innovation and intellectual property event last month, I summed up the standard-essential patent (SEP) licensing terms for cellular IoT products in a highly condensed format:

• IoT-centric devices: 1% of net selling price, capped at 0ドル.75/unit, for categories NB, M, and 1; category 4+ to be discussed individually

• IoT-enhanced devices: from 0ドル.30/unit for Category NB to 1ドル.00/unit for Category 4+

Given that IoT is basically the "new frontier" in cellular SEP licensing, and in light of the IoT and SME-related concerns that led the European Commission's Directorate-General for the Internal Market (DG GROW) to launch an aggressive rule-setting initiative, a more elaborate discussion of Huawei's cellular IoT licensing terms is warranted.

Huawei provides details on that licensing program in the form of a dedicated web page as well as a two-page PDF that elaborates on the key definitions, particularly of product categories. That effort is laudable, but it also shows that the topic--regardless of whether a particular company makes a licensing offer--is not as straightforward as licensing smartphones or connected vehicles.

The price matrix has three royalty structures:

1. The simplest scenario relates to IoT-enhanced devices. That categoriy includes certain products such as shared bicycles where a lot of value is in areas other than connectivity. A bicycle can serve its basic purpose--getting from A to B--with or without connectivity, though cellular connections obviously facilitate "sharing economy" services. For such products, Huawei charges a fixed per-unit rate (from 0ドル.30 to 1ドル.00).

2. IoT-centric devices are the ones that basically combine limited other functionality (such as meters, sensors, or GPS) with connectivity, and it's really just that combination that makes them useful. In April I commented on a European Commission document that claimed to know that a certain set of IoT licensing terms (by a Sisvel pool) were not FRAND, but as I noted, the products in question are very simple devices: they measure something, and send the data somewhere. That's why a royalty rate that may seem high to some EC officials can very well be justified in light of the fact that cellular connectivity accounts for a very large part of the overall value of such products. I also thought the EC did not sufficiently take into consideration that such products tend to be in use for far logner periods than smartphones, for which two years are a common upgrade cycle.

   Normally, Huawei's royalty rate for products that fall into that category is 1% of the net selling price, but capped at 0ドル.75/unit, which corresponds to a cap of the royalty bae at 75ドル.

3. For IoT-centric devices that fall into 3GPP LTE categories 4 or higher, Huawei's web page just says: "Contact us for rates." I've asked some people in the industry what types of devices would be IoT-centric (as opposed to merely IoT-enhanced) but meet 3GPP's criteria for category 4+, and it seems no one can conceive of a use case for that combination. Category 4+ devices would be rather expensive while IoT-centric devices such as smart meters tend to be cheap. Therefore, I assume that Huawei can't state a royalty rate because if someone actually created a product that against conventional wisdom makes category 4+ use of cellular connectivity but is a smart meter-like IoT-centric device, Huawei would have to see it to be able to make an offer. It's a theoretical combination, but one that may never come up in practice.

For licensees, the percentage-with-absolute-cap royalty structure for IoT-centric devices such as smart meters is advantageous because it means that a cheap device, such as a 25ドル asset tracker, gets licensed at a low per-unit rate. For IoT-enhanced devices, however, a fixed per-unit fee works because those are expensive enough that it's not necessary to start with a percentage, but the key thing is that a fixed per-unit fee is certain not to capture the value in any other components of such a product. For instance, if one shared bicycle comes with more expensive wheels and tires than the other, but both make the same kind of use of cellular connectivity, patent royalties don't capture a part of the premium price that would be paid for better wheels and tires.

Presumably it helps that Huawei continues to be a major implementer that pays about as much to other patent holders as it collects from implementers. And at the end of the day, what licensees do by taking a proper license to Huawei's SEPs is to support further research and development, which was a key message of Huawei's recent presentation.

The 3GPP device category definitions are obviously not a special feature of Huawei's licensing terms. They were created by a standardization body. Huawei's PDF just provides clarifies, such as that Category NB (narrowband) includes Category NB1 as well as Category NB2.

Huawei's PDF defines a basic IoT device as a "finished, complete, and ready-to-use terminal device that (1) implements Cellular IoT Standards for data transmission but not voice transmission, and (2) has an application/user function that is dependent on data transmission." Certain product types (such as phones, tablets, and PCs) are explicitly excluded. The PDF also makes clear that "Non-Human-Powered vehicles" (such as automobiles) are not included. There obviously are licensing options and rather different terms for those.

The PDF furthermore gives examples of IoT-centric devices (asset trackers and smart sensors) as well as of IoT-enhance devices (smart utility meters, point-of-sale machines, shared bicycles, and connected healthcare devices).

The EU institutions should have more faith in market-driven solutions. Huawei's terms are transparent, designed to avoid patentee overcompensation, and should also suit IoT SMEs. The terms of Sisvel's IoT pool--whether or not everyone at the Commission has understood the rationale--are also perfectly transparent and SME-friendly. And it's been only two weeks since Avanci announced European SME Axxès (highway toll collection technology) as a licensor of the Avanci Aftermarket pool. Solutions are being worked on, and they increasingly address the problem that some EC officials believe to have identified.

Constructive approaches to licensing should be encouraged by the European Commission, but the ill-conceived proposal that is currently on the table threatens to complicate or in some areas even disrupt (by encouraging holdout) the licensing process. As I argued in my previous post, at least the EU Council and/or the EU Parliament should obtain an opinion from their legal services, as fundamental rights and international comity are at stake.

The European Commission's U-turn on standard-essential patent (SEP) pools is remarkable and apparently attributable to lobbying by certain parties, facilitated by DG GROW's mistake of relying on an adviser whose business benefits from complex licensing processes entailing many bilateral negotiations (and disputes).

1. From "promotion" (2014) to skepticism (2021)

2. Article 11 of the leaked draft regulation

3. Distortions and unsupported assertions in draft impact assessment

4. IPlytics' obvious conflict of interest

1. From "promotion" (2014) to skepticism (2021)

In 2014, a study on Patents and Standards ("prepared for the European Commission Directorate-General for Enterprise and Industry") stated the following about patent pools (click on the image to enlarge or read the text below the image):

Promotion of patent pools

Patent pools provide a one-stop solution for licensing a bundle of standard essential patents owned by different entities, thereby aiming to mitigate transaction costs, avoid royalty stacking and create a level playing field. Given these benefits the study has examined the following aspects:

• Strengthening the relation between SSOs and pools;

• Providing incentives to SEP holders to participate in patent pools;

• Encouraging entities such as universities and SMEs to participate in patent pools.

That was not the last time the Commission recognized the benefits of patent pools, which also happened in connection with the horizontal cooperation guidelines.

Two years ago, DG GROW started to question the benefits of pools, asking a lot of questions in the description of a webinar. But at least it listened to both sides, and invited people with different perspectives, among them the president of the only major European patent pool administrator--Sisvel--as well as a Europe-based executive of Avanci.

2. Article 11 of the leaked draft regulation

The leaked draft of an EU SEP regulation (see my previous post on that one for a link list) imposes new transparency requirements on patent pools:

Article 11

Information to be provided by patent pools

Patent pools or any entity representing a collaborative licensing group shall publish on their websites at least the following information:

(a) standards [...]; (b) the administrative entity’s shareholders or ownership structure; (c) process for evaluating SEPs; (d) list of evaluators having residence in the Union; (e) list of evaluated SEPs and list of SEPs being licensed; (f) [sample claim charts]; (g) list of products [...]; (h) royalties and discount policy [...]; (i) standard licence agreement [...]; (j) list of licensors [...]; (k) list of licensees [...].

That article is an example--one of many in that draft regulation--of what happens when there is no proper consultation process on specific regulatory ideas. Whoever drafted or contributed to that article must live in an ivory tower.

Item (b) is misplaced because the disclosure of shareholders and ownership structure is a subject of corporate--not patent--law, and determined by national laws.

It is unclear what a pool should say about the "process for evaluating SEPs": when would a pool run afoul of that part of the regulation?

Item (d) is a typical example of regulatory zealots (almost a euphemism) wanting the EU to cut its nose to spite its face: it actually creates an incentive for pools to work with evaluators based outside the EU, and for such evaluators to move out of the EU, especially since their work is not location-dependent and they can find lower taxes and better weather elsewhere. It's also not clear why the disclosure of shareholders would affect entities (be it the pool administrator firms or their shareholders) based outside the EU while the part concerning evaluators is limited to the EU. Apart from that, it is a bad idea because patent pools have good reasons to shield their evaluators from attempts to exert pressure or otherwise influence their views.

To the extent that item (e) wants pools to publish a list of their patents, there are pools that do so already, but as long as the names of the licensors are known, licensees can find out anyway. The bigger issue is that a publication of the "list of evaluated SEPs" will invite validity challenges (and possibly also actions for declaratory judgment of non-essentiality) targeting those patents in particular. Again, whoever came up with or supported that idea is clueless.

Item (f) about claim charts should be left to the way the courts apply Huawei v. ZTE.

Pools tend to be quite transparent about their standard royalties. The term "discount policy" in item (h) is again unclear. Does it include duplicative-royalty policies?

Some pools publish their standard license (item (i)), but why should everyone have to do that? When implementers inquire, they'll get an offer.

While disclosing licensors (item (j)) is non-controversial, item (k) even goes against the interests of licensees: in practice, it is typically the licensees who don't want to be named. Anyone who followed automotive SEP licensing in recent years knows that Avanci has for some time had more licensees than the brands listed on its website. A pool that is successful would like to name all licensees, but some licensees insist on confidentiality.

Article 11 is deeply flawed, but what the draft impact assessment says about patent pools is even worse.

3. Distortions and unsupported assertions in draft impact assessment

In my previous posts and in the first part of this post I've mostly focused on what the draft regulation itself says. But the draft impact assessment, which leaked at the same time and will accompany the legislative proposal, makes laughably off-base claims regarding patent pools.

Before I point out some nonsensical details, let me start with a more fundamental question. Why take issue with the licensing terms (including, but not limited to, royalty rates) of patent pools in the first place? As long as bilateral licensing options exist, pools have to compete with those bilateral alternatives. None of the pools mentioned in the draft impact assessment bars patentees from entering into bilateral deals. Much to the contrary, I could immediately think of multiple examples for pools managed by those firms where such deals have indeed been struck (such as Huawei's agreements with Sharp and Conversant well after those companies had joined Avanci).

A pool's raison d'être is all about transactional efficiencies. If the pool rate or its other terms are too onerous, bilateral licensing will prevail, the pool will be marginalized unless it adapts, and the market--not the Commission--will have solved the problem.

The pools are not the problem, and it's not the pool administrators' fault that some German courts effectively require licensees to take an entire pool license without requiring the patent holder to make a bilateral portfolio offer (satisfying FRAND criteria) as another option. If an implementer is found to infringe a valid and essential patent, and there's a pool license on the table that costs very little and covers maybe a few dozen patent families, a court may reasonably interpret Huawei v. ZTE to the effect that a pool license is what a willing licensee would normally take--and can then determine whether the pool offer is FRAND. Admittedly, if a license fee is high and/or there are huge numbers of patent families in the pool, the market dynamics I described (pool competing with bilateral offers) are lost due to a "license or die" judgment.

If that is what the Commission is concerned about, its draft regulation is unable to solve the problem: there is nothing in there that would prevent the infringement courts from obligating an implementer to take a pool license while not requiring the patentee to make a bilateral offer. There would be a FRAND determination by the EU's trademark office, which the infringement courts can just ignore. There would also be an aggregate royalty determination for the entire standard, which the infringement courts can likewise ignore because they--not the Commission--decide whether to apply a top-down approach to valuation, and because they can disagree with the findings anyway.

The decision makers in the EU institutions must figure out one plain and simple truth: if an implementer takes a pool license, it's normally just because it's more efficient than taking numerous bilateral licenses.

The draft impact assessment is trying to understate Avanci's success. Avanci covers far more than the 60% of (up to) 4G cellular SEPs that the draft impact assessment states on page 9 (PDF page 13). The only explanation for the missing 40% is footnote 54:

"Companies missing from Avanci include among others Huawei, Samsung, Apple, Google, ETRI. Source of % IPlytics."

I'll talk about IPlytics further below. There's no substance there, the number doesn't make sense, and at any rate those five companies do not explain the missing 40% by far and away. In particular, Apple and Google--who acquired virtually of their SEPs--are not known to engage in outbound SEP licensing (they only cross-license), and that generally applies to Samsung with the sole exception of its participation in video codec patent pools. In other words, if a car maker takes an Avanci license, there is only a residual risk of litigation left.

On the same page, there is a Figure 4 that way understates the number of vehicles licensed by Avanci, and it does so by setting an arbitrary, capricious, and either incompetent or disingenuous cutoff date: August 4, 2022. Major car makers, especially from Asia, signed shortly after that date. The text above the charteven mentions Avanci's September 2022 announcement, so why didn't they update their chart accordingly? A picture is worth a thousand words, and that picture is grossly misleading.

The Commission doesn't appear to credit Avanci for the fact that it maintained a consistent pool rate over many years despite adding tremendous value--and it doesn't take that fact (the addition of licensors after the formation of a pool is announced) into consideration when discussing Sisvel. From an industrial policy perspective, it strikes me as odd that the Commission goes not only against Avanci but also against Sisvel, the only major European pool administrator. Other world economies appear to be more interested in the trade effects of intellectual property rights while the Commission would rather have EU companies send their royalty checks abroad.

The draft impact assessment lumps Sisvel--a major pool administrator--together with "specialized patent assertion entities" and incorrectly categorizes it as a former operating company:

"This includes state-owned entities (such as Japanese IPBridge or FranceBrevets), ‘privateering’ spinoffs from large operating companies (e.g. Unwired Planet from Ericsson, Panoptis from Panasonic etc), former operating companies who have ceased other activities to concentrate on patent licensing (e.g. Sisvel), and private companies acquiring patents from a variety of predecessors (e.g. IPcom, Uniloc, etc.). PAEs contribute to further fragmentation of the SEP owners market." (emphasis added)

Give me a break. Sisvel has been a patent licensing firm from the get-go. It never had an operating business.

It's also unclear to me on what factual basis (I believe there is none) the Commission distinguishes between Unwired Planet and IPCom. Why is one a "spinoff" from Ericsson and the other "acquir[ed]" patents from Bosch? But that's just another implausibility that shows some people at DG GROW don't know what they're talking about (and want the Commission as a whole to damage its reputation by proposing to put out such crap).

On page 25 (page 29 of the PDF), the Commission again does Sisvel injustice. It claims--without any factual basis whatsoever--that in the IoT SEP licensing space, "[t]he FRAND royalty expectations are likely to be higher than the market would suggest" and then points to footnote 139:

"For example, Sisvel launched a new IoT pool with 20 licensors who claim to represent about 30% of all SEPs. The price for a smart meter is announced at USD 2, so the total aggregate royalty would be around USD 7 (EUR 6.7 on 15 February 2023). The value of a consumer smart meter is reported to be around EUR 65."

No source is given. They're not even just saying "IPlytics" (which wouldn't be a sufficient answer anyway). There's no evidence of Sisvel "claim[ing]" that the respective pool represents 30% of all SEPs, while it is clear that Sisvel will try to attract more and more licensors to the pool (as I mentioned before in connection with Avanci), so there will be greater value to licensees for the same price.

How can the Commission claim to know what is FRAND?

I would argue that the extremely poor quality of DG GROW's draft regulation and of the impact assessment reflect anything but a good understanding of FRAND.

The alleged value of a consumer smart meter ("around EUR 65") is a sandbagged figure. It's more like twice that amount. But even if we assumed that EUR 65 was correct, and if we furthermore believed the Commission's unsubstantiated royalty-stacking analysis, I can't even see why an aggregate royalty rate of roughly 10% would necessarily be supra-FRAND given that smart meters are very simple devices:

• They measure something, and

• they send the data somewhere.

Cellular technology is a large part of what those devices are about. It's obvious that a cellular SEP royalty rate is, percentage-wise, lower on a connected vehicle than a baseband chipset, as the latter does nothing but implement cellular standards.

In any event, that Sisvel pool--like all of its other pools--competes with the alternative of implementers taking a multiplicity of bilateral licenses.

What is the Commission's agenda? To harm specific (even EU-based) companies? To encourage hold-out? The latter would be an interesting strategy as it would result in more litigation, which the Commission would then point to in order to argue that its proposal must be adopted as quickly as possible...

4. IPlytics' obvious conflict of interest

There is more than one factor that has led to the current mess. The solution is for the Commission to refrain from putting out a premature and ill-conceived proposal in favor of serious discussions of specific proposals with competent people.

In a recent post I quoted the draft impact assessment's footnote 283, which casts serious doubt on the quality of a DG GROW-commissioned study co-authored by IPlytics founder Tim Pohlmann.

It's possible that the Commission has listened to him too much, especially in connection with patent pools. He has a conflict of interest with respect to Avanci and other pools (regardless of whether pool firms are his customers, as some--though not all--are) for a simple reason: because successful pools greatly simplify patent licensing, they reduce the need for IPlytics' services. IPlytics thrives when there's complexity, confusion, and especially a huge number of negotiations and a lot of litigation. That is when parties on both sides of the negotiating table will look for material to underpin their positions.

Earlier this month, Mustafa Cakir ("Çakır" in Turkish) published a YouTube interview with Tim Pohlmann. Around minute 34, the discussion of automotive SEP licensing starts, and Mr. Pohlmann tries hard to downplay Avanci's success and the benefits it delivers. He even expresses, between the lines, unsubstantiated doubts about its royalty rate (a rate accepted by the vast majority of licensees) and about Avanci's ability to put together a 5G pool.

Where he discusses the situation faced by automakers, he focuses a lot on the modus operandi of those companies' patent departments prior to the widespread adoption of vehicle connectivity. Can car makers expect SEP licensors to change their business model only because they otherwise have to learn about cellular SEPs? Automakers incorporated those technologies into their cars, and they've also been able to have existing employees learn about cellular SEPs. In some cases, they've poached IP professionals from companies like Siemens.

IPlytics has only one problem with Avanci: that Avanci streamlines the cellular SEP licensing process for the automotive industry. As I mentioned further above, those who don't join Avanci are largely the ones who don't engage in outbound licensing anyway. Mr. Pohlmann's business would obviously benefit from an environment in which 50+ licensors have to approach approximately 100 car makers to work out bilateral licenses. Every infringement notice and every start of a negotiation process is potentially money in IPlytics' pockets--and every time a car maker takes an Avanci license, that opportunity is diminished.

Avanci is just an example. IPlytics has the same problem with any pool that is--or likely will--be successful. Then there were monks who copied books in a type of room called scriptorium and didn't welcome the arrival of the printing press. It's just that the decision makers didn't really listen to them, and DG GROW didn't exist at that time.

This is a follow-up to yesterday's post on why actual numbers of narrowband IoT-related standard-essential patents are hard to come up with, and keyword searches are inherently unreliable in this context. Tim Pohlmann, the founder and CEO of IPlytics, reacted on LinkedIn, first under his personal and then posting the same statement under IPlytics' corporate account. He "doth protest too much"--and he contradicts himself:

The actual report purports to "paint a robust picture of the NB-IoT landscape" (emphasis added) and to "provide meaningful, actionable insights." The headline is "Who is winning the IoT SEP race?" (on the website: "Who is leading the IoT SEP race?") On LinkedIn, however, Mr. Pohlmann wrote the following, which contrasts nicely with his marketing claims:

"Our reports are not published to provide the truth about the winner[s]. We publish reports to show[]case what can be done with the data we provide. We do not care who ranks first[,] second or third."

IPlytics wants to have it both ways. They make bold claims and promises in public, and their reports have been used not only in licensing negotiations but also by litigants. Once you challenge the methodology and question the results of a particular report, they distance themselves even from its very headline ("Who is winning..."), downgrade the piece to a "showcase," suddenly discover humility, and place the emphasis on a disclaimer ("our data analysis and report results are limited to the approach of a patent declaration-based keywords search").

I would recommend to litigants whose adversaries proffer IPlytics reports to show to the court what Mr. Pohlmann publicly stated on LinkedIn in response to my criticism. Toying around with a database isn't necessarily admissible evidence.

Sadly, the inconsistencies don't end there. Just two examples in the NB-IoT context, comparing last week's report to its 2020 predecessor (Who owns patents, SEPs and develops standards for smart home technologies?) in the specific context of NB-IoT portfolio size:

1. Huawei--of which I said yesterday that it and Qualcomm indisputably deserve the two top spots--was not even listed among the top NB-IoT SEP holders in the 2020 report, and now it is (credibly) the number one.

   What was IPlytics doing and thinking two years ago? The complete absence of Huawei from the list of leading patent owners should have been more than enough for the report to fail any plausibility check before it was published. IPlytics should have asked themselves whether their methodology was sound given that such an outcome was inexplicable. Instead, they went ahead and published their report anyway. How "robust" and "actionable" is that?

2. For Nokia, Figure 4 of the 2020 report indicated approximately 2,300 patents from 500 patent families. In the 2022 report, however, those numbers are down to approximately 400 patents from 150 patent families (Figure 1). How can Nokia possibly have lost 5 out of 6 assets--and more than 2 out of 3 patent families--during a two-year timespan? Again, how is that "robust" and "actionable" short of a patent cliff of unprecedented proportions in the field of wireless communications technology? And how is it responsible to create such confusion around a key long-term growth area for a publicly traded company like Nokia?

I emailed Mr. Pohlmann before German office hours on Thursday, and received a reply a few hours later: a detailed email that for the most part sidesteps the issue I raised (the above discrepancies between the 2020 and 2022 reports). Only one paragraph really addresses my question, and Mr. Pohlmann (with whose company I signed an NDA a long time ago) authorized publication of his email, so let me quote:

"The smart home report from 2020 used a much simpler approach. Here only TS [technical specifications] that mention NB-IoT and LTE-M were identified to then connect this given list [of] TS to declared patents. So this report uses a keyword search in TS documents that have declared patents associated. The data approach of both reports are quite different. Also I must say that the number of declared patents has very much increased since July 2020."

The "much simpler approach" in 2020 did not dissuade IPlytics from publishing it (and declaring "winners") at any rate. If they say in 2022 that they've improved since the 2020 report, what will they tell us in 2024?

The last sentence about a massive increase in the number of declared patents (by the way, we're talking about subsets of the 4G standard, which is already quite mature by now) still doesn't explain why they published a list in 2020 (instead of immediately identifying an issue) without Huawei--the actual number one--among the top 10 patent owners. Mr. Pohlmann is a frequent speaker at SEP conferences, which in addition to his own company's patent database provides him with plenty of opportunity to get a feel for the market.

It is unfortunate that it is so hard to determine portfolio sizes in the narrowband IoT context. There are some companies who may only have a 1% or 2% share of all 4G/LTE SEPs, but their research may be disproportionately focused on parts of the standard that are relevant to its narrowband subsets (LTE-M and NB-IoT). Someone who is an average-sized fish in the larger pond may be a bigger fish in the smaller pond. Someone who filed for patents before those narrowband subsets were defined will not have used certain keywords in the claims or specification of a given patent application, but it may later read on NB-IoT and/or LTE-M anyway (false negatives under a keyword-based approach), while someone else may just have thrown in some keywords at a later stage, which doesn't guarantee that a given patent actually maps to the relevant specification of the standard.

IPlytics should be lower-key in its PR and marketing communications. That will lead to greater consistency. It's always better to underpromise and overdeliver, especially when there is the possibility of those reports being used in litigation or relied upon by policy makers and regulators.

The Internet of Things (IoT) is a tremendous growth area in multiple respects: for product makers, for standard-essential patent (SEP) holders, and for patent analytics companies looking to acquire new categories of customers. But in the midst of a gold rush, not all that glitters is really gold.

Last week there were two IoT SEP-related announcements: Sisvel's new narrowband IoT pool, which started with 20 licensors, most notably also Ericsson; and just one day earlier, IPlytics released a report on who is supposedly winning the IoT SEP race. As an IP and antitrust commentator, I wish to help my readers avoid being misled. That's why I criticized a ranking of German patent litigation firms last month, and suffice it to say I received a lot of positive feedback from people who appreciated it. And there are serious issues with that IPlytics report, which I believe someone has to call out.

Prior to this one, I publicly criticized IPlytics only once, and that was when Tim Pohlmann interviewed the president of an Apple astroturfing operation claiming to represent small app developers and IoT companies while actually working against the interests of small innovators, especially in the App Store antitrust context. When that interview took place, there was enough information out there already to know that ACT is not a legitimate representative of whom the claim to speak for, and IPlytics should (or must) have known that.

Now IPlytics has been acquired by RELX (congratulations!), a company known for such services as LexisNexis. I have no problem with RELX, and I have no position at this stage on whether WIRED is right that RELX's practices represent a threat to data privacy.

That recent IPlytics report on IoT patent holdings is nothing that decision makers, whether in the public sector (such as competition enforcers and policy makers) or in companies (patent holders, pool administrators, or implementers), should rely on. I have problems with its methodology and with some of the results.

If it wasn't free, I'd have to say: caveat emptor!

First, the methodology makes it a clear Daubert case (those familiar with U.S. litigation know what I mean, and the others can figure).

The report mentions a "keyword approach" three times. On the penultimate page, there is a disclaimer:

"We used a keyword approach without additional filters for the active or granted status of a patent or patent family. We want to highlight that other keywords or additional selected filters might result in other ranking positions and shares. Further, we refrain from making assessments of the technical relevance of patent portfolios."

A "keyword approach" cannot make up for a problem facing anyone (not just IPlytics) who will undertake to evaluate the narrowband IoT patent space: there are no databases of NB-IoT- or LTE-M-specific declarations. The reason being that NB-IoT and LTE-M are simply true subsets of the wider 4G/LTE standard, and that's where the declarations and concomitant FRAND pledges were made. Those narrowband standard-setting efforts started when 4G/LTE was already far along, and what they did was to select those parts of LTE that they deemed suitable to task.

Keywords are a poor indicator, and actually they are no indicator when many patents were filed before it was possible to know or even just predict the relevant keywords. Many 4G/LTE patents read on narrowband IoT, but when the applications were drafted, the keywords simply weren't floating around yet.

The sad truth is that one would have to get down almost to the level of claim charts to really find out who owns how many narrowband IoT SEPs. That's obviously not going to happen.

Second, there are some implausible results. And some of those implausible results are plausibly attributable to the methodology issues I just outlined.

I don't have a plausibility problem with Huawei and Qualcomm coming out on top here. Those two companies have powerful portfolios and long-standing strategic interests related to IoT. It's hard to think of any wireless technology where they wouldn't be very strong. My problems are mostly at the level right below Huawei and Qualcomm:

I would normally assume that companies like Ericsson, Nokia, Samsung, and LG are roughly on an equal footing (Ericsson probably being the number one among them)--to give you a ballpark figure, I would expect each of them to hold around 10% of the relevant patents. Those are companies that clearly have not only a general interest in IoT but also participate quite actively in the related standard-setting. But in IPlytics' narrowband IoT ranking, Nokia is roughly at a level with a company like Lenovo, which has a rather different focus and business model.

Below companies like Ericsson and Nokia, I would see players like Sony and NTT DoCoMo, yet they presumably own more relevant patents than a number of companies listed ahead of them.

Maybe IPlytics did its best and it's not good enough (yet). So what are the alternatives?

The first alternative in a context like this is simply the Socratic approach. In his Apology, Greek philosopher Plato praised the key quality in Socrates not to think he knows what he doesn't know. "I know that I know nothing."

It would be terrific to have a set of numbers that would enable us to get an idea of what the licensing costs for the full stack might be. That would enable top-down royalty determinations, which are so very popular. But we may have to live with the fact that, at least at this stage, we just don't have the data at hand for that. And when data is unreliable, it's better to just accept that fact. Things get worse--not better--if we act in defiance of that realization.

There are bogus medications for incurable diseases. They raise false hopes and have adverse effects--as do unreliable rankings, even those produced with the best intentions.

I wouldn't underestimate the ability of the market to figure out solutions and royalty rates, especially in narrowband IoT, where different standards are competing with each other for "design wins."

And I would strongly recommend to use common sense. While one can't assume that companies' contributions to a standard-setting process always correlate to a given player's portfolio strength, we should at least look at who really participated in standardization and take that into account, as we all know that those who sit at the standard-setting table will also ensure that some of their technologies are included and will strategically file patents so they read on the relevant standard.

Is more transparency needed in the narrowband IoT context? Absolutely. It should be a priority for the industry to come up with something better, ideally in 2023.

To be clear, I am not taking a position here on any other IPlytics report than the one I specifically mentioned, nor on their raw data or anything else they offer.

Last year, patent pool firm Sisvel was on a roll in terms of settling infringement actions and, generally, striking license deals with major implementers. This year, Sisvel is primarily making news with new pool initiatives. In the summer, Sisvel launched a WiFi 6 pool, the most notable contributor to which was Huawei, and a formula for the computation of license fees that encourages early adoption called Licensing Incentive Framework for Technologies (LIFT).

Just this morning by European time, Sisvel announced "the launch of its Cellular IoT (C-IoT) Patent Pool consisting of 20 patent owners."

The initial licensors are a large and diverse group:

• mobile telecommunications pioneer and infrastructure maker Ericsson is particularly powerful (and on the winning track against Apple in Mannheim, as I reported yesterday);

• like Ericsson, Datang builds mobile networks;

• network operators like KDDI, NTT DoCoMo, Orange, and Telefónica;

• ETRI and Langbo are examples of renowned research institutes;

• non-practicing entities (NPEs) like the Unwired Planet/Optis Wireless/Optis Cellular group; and

• MediaTek and Sony Group Corporation (through its subsidiary Altair Semiconductor) supply chipsets to (not only, but also) IoT implementers, which makes them particularly interesting contributors to a pool that licenses at the end-product level: they obviously understand the IoT supply chain and have determined that licensing at the end-product level is workable, at least when patent pools provide transactional efficiencies.

The standards covered are LTE-M and Narrowband IoT (NB-IoT)--the same two standards with respect to which Huawei granted a license to chipmaker Nordic Semiconductor in the spring. A simplified way to explain the typical fields of use for those standards is that LTE-M has the benefit of being compatible with existing LTE network infrastructure, while NB-IoT requires hardware capable of DSSS modulation. On the bottom line, LTE-M is more powerful than NB-IoT, a fact that is also reflected by the respective royalty rates: for any NB-IoT product, the rate is 0ドル.66 per unit for LTE-M, a distinction is made between asset trackers (1ドル.33 per unit) and smart meters (2ドル per unit).

The relevant applications are characterized by low power usage--they may even come with batteries that last ten years--because they transmit small amounts of data and only sporadically, such as once per day or week, or maybe just on an "as needed" basis. Another characteristic is that data must travel over longer distances than the ones that could be cost-effectively covered with a WiFi network. For instance, if cattle needs to be tracked on a large farm, one would have to set up a huge number of WiFi access points, each of which would need access to the power grid. In that case, it makes a lot more sense to use the 4G cellular standard and transmit data via existing base stations.

There are many small companies, especially startups, that serve those needs with highly specialized (vertical) products. Bilateral licensing between dozens of standard-essential patent (SEP) holders and a large and growing number of mostly rather small companies would come with relatively high transaction costs. That's precisely the pattern that favors the creation and operation of patent pools: instead of countless one-to-one deals, pools enable many-to-many transactions.

As I already noted above when discussing the initial group of licensors, it is very meaningful that companies like MediaTek and Sony have concluded that a pool licensing at the end-product level (as opposed to the component level) is the way to go. While Ericsson is a far bigger cellular patent holder than Sony and MediaTek, it has a long-standing policy of licensing only at the end-product level, so no surprise there. When a consensus is built around a licensing structure and the parties who agree to it have such a diversity of perspectives on the IoT business, and even chipmakers participate, it suggests to me that the market may very well able to work out IoT SEP licensing.

When LG joined the Avanci automotive SEP pool earlier this year, the fact that LG is a supplier to the automotive industry made its accession particularly interesting. And now that we're appproaching the end of the year, there isn't even room for debate: the Avanci model (through which Sisvel actually licenses its own patents to car makers) has been universally accepted. Even a tireless tire maker named Continental has given up opposing that which the market has accepted to do. What LG is to Avanci (a licensor situated higher up in the relevant supply chiain), MediaTek and Sony (because of Altair) are to Sisvel's narrowband cellular IoT patent pool--and from Day One.

Today's announcement further reduces to absurdity the way in which Apple and its notorious astroturfers try to leverage the question of IoT SEP licensing only to devalue SEPs in general, particularly with a view to smartphones. At this stage, we are not seeing widespread IoT patent litigation; in fact, one would be hard-pressed to find even one lawsuit targeting an IoT startup. This is not the time for regulatory intervention as market dynamics may take care of the problem. As I'll discuss further below, this new pool is about boosting adoption.

There is a major difference between this pool and most other SEP pools (such as Avanci, which I just mentioned): this is an emerging market. Car makers had adopted 4G when Avanci started; it took years to solve the licensing question, but the demand was there, except that some though that hold-out would be profitable. Sisvel's president, Mattia Fogliacco, is quoted in today's press release as follows:

"The IoT market has been in development for about a decade, however only now we see clear early signs of widespread roll-out. The LTE-M and NB-IoT technologies are an obvious pick to connect products and services, and with this pool, available already in an early phase of technology adoption, we will remove a lot of questions and concerns by giving easy, transparent and reasonable access to the patents that the creators of the technology hold." (emphases added)

So this is a pool that is even more about market development at this stage than just monetization, though the latter could--all going well--reach a very interesting scale over the years. In the same vein, the pool's program manager, Sven Törringer, said:

"[W]e are convinced that the offer we bring to the market will not only reduce friction and concerns, but rather boost the interest in the adoption of the LTE-M and NB-IoT standards in IoT products, allowing for swift investments in this sense."

Sisvel's press release mentions competing standards: "LoRaWAN, Sigfox, WiFi, MIOTY, Bluetooth, various mesh network technologies, etc, and combinations thereof." When implementers choose between standards, patent licensing plays a major role in addition to technical considerations. This new pool needs design wins (decisions by implementers to use those standards) while many other pools are all about courtroom wins from the get-go (if the relevant standards are already in widespread use and rampant infringement needs to be remedied).

Of course, as we've seen in such contexts as automotive (where Avanci is now a one-stop shop) and video codec patents (where things worked out better in the beginning than they have more recently), standards typically benefit from a single pool bringing everyone together. I'll keep an eye on who else will join this one.

IAM was first to report (paywalled) on an announcement I've now found on Nordic Semiconductor's website, according to which Huawei has extended to Nordic Semiconductor a component-level license to Huawei's standard-essential patents (SEPs) on what both companies consider fair, reasonably, and non-discriminatory (FRAND) terms. As a result, Nordic's customers are covered ("automatically licensed and protected" according to the announcement).

Just last week, the breadth and depth of Huawei's patent portfolio was on full display at a corporate event with distinguished guest speakers. Huawei clearly positioned itself not only as an innovation powerhouse but also as a company with a very balanced perspective, knowing the perspective of licensors as well as what it's like when the shoe is on the other foot.

In today's Huawei-Nordic announcement, the first sentence of the first quote is instructive:

"'Huawei owns a leading portfolio of LPWA SEPs for LTE-M and NB-IoT, a subset of the 4G standard, which creates great value for IoT,' says Huawei’s Head of European IPR Department, Zhang Xiaowu."

LPWA means low-power wide area cellular technology. They're not talking about baseband chips for smartphones or telematics control units (TCUs) built into cars. Note in the quote above that this is merely "a subset of the 4G standard." Obviously the royalty rates are lower in such scenarios, though the announcement doesn't state any specific amounts. It would be discriminatory to apply equal terms to unequal conditions, meaning that products making use of only a subset of a cellular standard--and making limited use even of that subset, such as by sending only small quantities of data--had to pay the same license fees as, for example, Apple.

About five months ago, Nordic announced another SEP-related agreement with Nokia. This makes Nordic a bit of a pioneer--perhaps even thought leader--in its market segment. While the terms of either deal weren't disclosed other than what one can deduce from the announcements, it's clear that Nordic's agreements with Huawei and Nokia couldn't structurally be more different:

• The headline of the announcement on Nordic's website specifically talks about "component-level licensing." Nordic itself is the licensee; its customers are indirect beneficiaries. There is no indication of any need on the part of Nordic's customers to reach out to Huawei in order to obtain those benefits.

• The Nokia deal essentially makes Nordic an intermediary that obtained from Nokia an offer that Nordic's customers can elect to take. If they don't take it, they have to seek a bilateral license, and if they did neither, they'd obviously expose themselves to potential infringement litigation.

So there's a diversity of licensing models, and Nordic's two publicly-known deals with major cellular SEP holders are heterogeneous. That's why I'm not buying the part of their headline where they call this "a big step towards industry-wide component-level licensing." The accurate way to assess the situation is that some SEP holders--here, Huawei--are prepared to grant component-level SEP licenses, but others--such as Nokia--are not. The announcement notes that "Huawei and Nordic were able to conclude the agreement through a transparent and amicable discussion within a short period of time." If anyone ever were to claim that Nordic was an unwilling licensee, or that Huawei wasn't a constructive licensor, this would be one of the transactions they could point to as an example of their willingness to make things work. That said, Nordic has validated not only one licensing model--but two different ones. It has a clear preference for the one it announced today, but the Nokia deal is and remains in full force and effect. If they persuaded Nokia to switch from end-product licensing to component-level licensing, then they would be in a stronger position to claim that the industry was going in that direction.

Where I definitely agree with Marianne Frydenlund--Nordic's senior vice president Legal & Compliance--is that "[l]icensing in cellular IoT is a comparably new practice in the industry, calling for flexible solutions." Flexibility, of course, is never a one-way street.

Talking about flexibility, the aforementioned two models--Huawei/Nordic and Nokia/Nordic--are not the only ones. For example, last year Huawei announced a license agreement with a Volkswagen supplier (in that case, about all of Huawei's 4G SEPs, not just a subset of the standard) under which the supplier is licensed, but specifically with respect to Volkswagen cars--while Nordic today announced that all of its customers, not just one, are licensed.

Then there are license agreements that involve "have-made" rights. That is a way for end-product makers to provide a certain degree of legal protection to their suppliers. While patent exhaustion works in the other direction, a similar effect can be achieved contractually.

I'm a pluralist: if it works, it works.

The IoT licensing landscape is in its infancy--but, as we see, it's evolving pretty fast. That's why now is not the time for massive regulatory or legislative intervention. It's all too easy to see that some (particularly Apple and its traditional allies, as well as Continental, which prefers litigation that is going nowhere, and combative speeches at conferences, over pragmatic licensing-focused solutions) would like to use IoT for a pretext to urge governments to put a thumb on the scales of SEP licensing. The U.S. government ultimately decided not to support those organizations' SEP devaluation campaign.

In light of today's announcement, I recommend the following:

• Policy-makers and regulators should give the market a breathing space to work things out. There is progress, and as Conti's own submission to the EU Commission notes, there's virtually no SEP infringement litigation at this stage targeting IoT end products. "If it ain't broke, don't fix it." IAM's Joff Wild wrote on LinkedIn this morning (in the specific context of this license agreement) that "others in [Nordic's] position may well be able to secure similarly interesting opportunities, without any threat of litigation hanging over negotiations. That, in turn, may be something that European regulators could reflect on as they ponder the future of SEP licensing on the continent."

• Licensors should keep it that way and refrain from litigation unless they really encounter an intolerably unwilling licensee. Otherwise there could be unintended consequences, such as regulatory action or legislative measures.

• Implementers should follow the example of Nordic's agreements with Nokia and Huawei, and act constructively. SEP holders can't wait forever to get paid: they need licensing income to finance the next innovations.

Today's announcement is a very important one, and hopefully the IoT industry won't have to wait another five months until the next one of this kind.

Follow @FOSSpatents

Share with other professionals via LinkedIn: