Chapter 8: Royalty-Free Patent Policies and the Exclusion Mechanism

Royalty-free patent policies are the foundation of most modern technology standards work. The premise is straightforward: participants commit to licensing their necessary claims at no charge, creating a zone where anyone can implement the standard without worrying about patent royalties.

In practice, these policies are anything but straightforward. The details — scope, triggers, definitions, defensive provisions — determine whether the commitment is meaningful or hollow. This chapter walks through the anatomy of a royalty-free patent policy, using real structures to illustrate the choices and tradeoffs involved.

8.1 The Philosophy: Royalty-Free and Litigation-Free

A royalty-free patent policy creates two things simultaneously: a royalty-free zone and, at least among the participants, a litigation-free zone.

The royalty-free part is obvious — patent holders commit not to charge for licenses to their necessary claims. The litigation-free aspect is equally important, though its boundaries are worth noting. When everyone in the room has made a commitment to license their necessary claims at no cost, the incentive to assert those patents against fellow implementers drops to near zero. The result, among participants, is an environment where companies can implement the standard without the background threat of patent litigation from other members.

This protection doesn't extend to parties outside the tent. Non-participants who hold patents reading on the standard are not bound by the policy and remain free to assert. The royalty-free commitment creates safety among those who made it — not universal immunity.

This is the value proposition that makes royalty-free standards attractive, particularly in the software and web spaces where the technology moves fast and the cost of clearing patent rights for every implementation would be prohibitive.

One practical factor that reinforces adoption of royalty-free policies: familiarity. Organizations like JDF and OASIS offer multiple IPR policy modes, including the ability to adopt established policies like the W3C Patent Policy or the ISO/IEC patent policy by reference. The W3C policy is not a model of clarity — it can be genuinely confusing on a careful read. But it's one of the first royalty-free patent policies, and people know it. The value of a familiar policy is the same as the value of a familiar open source license. You could write a bespoke agreement that's technically superior to MIT or Apache 2.0. Nobody would trust it. But if you say "we're using the W3C policy," the conversation moves to substance rather than stalling on the framework. In standards, where the goal is broad participation, practical trust often matters more than legal precision.

The tradeoff is that patent holders give up the ability to monetize their standard-essential patents — at least for that standard. Industries where companies have made significant R&D investments that they expect to recover through patent licensing tend to resist royalty-free policies. That's why telecom and video codec standards often use RAND rather than royalty-free. We covered that in Chapter 7.

8.2 The Licensing Commitment in RF Policies

As discussed in Chapter 3, a patent policy creates a commitment to license — not an actual license. The license itself is negotiated bilaterally outside the standards body. That general principle applies to royalty-free policies, but with two RF-specific nuances worth noting.

First, there are exceptions to the commitment model. A small number of organizations use actual licenses or non-assert agreements — the Open Web Foundation Agreement, for instance, is a direct non-assert commitment rather than a promise to negotiate. Some policies grant licenses that attach automatically upon implementation. But the commitment model remains the norm.

Second, in practice, the bilateral licensing step almost never happens in RF regimes. It is extremely rare for anyone to seek or offer a formal royalty-free license. When no money is changing hands, there's little incentive for either side to formalize the arrangement — and little practical remedy if they don't. If you've already committed to license at zero royalty, what are the damages? The commitment itself provides the assurance implementers need, and the bilateral license remains theoretical.

8.3 Scope: The First Thing to Get Right

The scope of a patent commitment is one of the most consequential decisions in setting up a standards project, and one of the most contentious to negotiate.

Organization-Wide vs. Working-Group-Specific

Patent policies apply at one of two levels: across the entire organization, or on a working-group-by-working-group basis.

Organization-wide policies are simpler to administer. Everyone who joins the organization makes the same commitment to all of its output. But they carry a significant risk: if the organization expands into new work areas — and organizations almost always expand — participants find themselves with patent commitments to specifications they weren't involved in developing, don't care about, and may not have anticipated.

The expression is that oxygen expands to fill the room. Give a standards organization the opportunity to expand, and it will. If your patent commitment covers everything the organization does, you're signing up for an unknown future.

Working-group-specific policies address this by confining the patent commitment to the output of the working groups you actually participate in. Each working group has its own charter, its own scope, and its own IP terms. You opt in by joining the working group. If a new working group forms to work on something unrelated to your interests, your patents aren't implicated.

This is why organizations like JDF and OASIS allow different working groups within the same organization to operate under different IPR modes — one group might be royalty-free while another, for strategic reasons, uses RAND. The flexibility to match the IP terms to the specific context of each work stream is a significant design advantage.

Getting Scope Right

Each working group should have a defined scope that sets the outer bounds of what the group can work on. This scope cascades into the patent commitment: if the group produces work outside its scope, the patent commitment may not cover it.

Scoping is more art than science, and it's one of the hardest things to get right.

Too narrow, and the group can't adapt as the work evolves. If the scope says "microwave ovens" and the group discovers halfway through that the real product should include a toaster function, they can't do it without rechartering — a process that can take months and require re-executing agreements.

Too broad, and participants face open-ended patent exposure. If the scope says "kitchen appliances that apply heat," a participant with valuable toaster patents may find those patents swept into a commitment they never intended to make.

Companies without significant patent portfolios in the relevant space tend to push for broad scopes — they have nothing to lose and want maximum flexibility. Companies with patents they intend to monetize or protect push for narrow scopes with explicit carve-outs. This tension is inherent in every scoping discussion, and it's one of the places where the negotiation gets real.

The contrast can be stark. When the Alliance for Open Media was formed to develop royalty-free video codecs, the scope of the video coding working group was essentially one line: we are going to create video codecs. That worked because the founding companies were aligned on the goal and had no intention of monetizing codec patents within the group. When the same organization later formed an audio codec working group, the scoping document was a page and a half of detailed carve-outs, because participants had different strategic positions and needed to protect specific aspects of their portfolios.

8.4 Who Makes the Commitment

Patent policies specify who is bound by the commitment. This has two dimensions: which entity within a corporate structure, and what triggers the obligation.

Affiliates and Corporate Structure

Most policies extend the patent commitment to patents "owned or controlled by a member or its affiliates." The intent is to prevent companies from hiding standard-essential patents in subsidiaries or affiliated entities to avoid the commitment.

This seems straightforward until you encounter real corporate structures. A venture-backed startup may be controlled by a VC that also controls hundreds of unrelated companies. A conglomerate's participation through one division doesn't necessarily mean it can — or should — commit the patents of its nuclear reactor division. An entity sitting several layers deep in a corporate hierarchy may have legitimate reasons why it can't bind its parent's entire portfolio.

These situations arise frequently and there's no universal answer. Sometimes you accept that the participant can only commit the patents they directly control. Sometimes you require them to sort out their internal structure before participating. Sometimes you negotiate a side letter that clarifies the scope of the commitment. The key is to understand the issue and address it explicitly rather than discovering a gap after the standard is published.

After-Acquired Patents

Most well-drafted policies specify that the commitment covers patents owned "now or at any time in the future." This means that if a participant acquires a patent after making the commitment — whether through R&D, acquisition, or any other means — that patent is pulled into the commitment if it reads on the standard.

Without this provision, a participant could make a commitment, then acquire a blocking patent the next day and assert it freely. The "now or any time in the future" language closes that gap. It seems obviously necessary, and it is, but it has been the subject of disputes.

8.5 Who Gets the License

A royalty-free commitment can run to different recipients, and the choice matters.

The most common approach is that the commitment extends to anyone implementing the standard — the entire world. This maximizes the value of the standard by ensuring universal access. It also means that non-participants — companies that didn't join the working group, didn't contribute, and didn't make any reciprocal commitment — still benefit from the patent commitment.

Some organizations restrict the commitment to members only. The rationale is that participants should have skin in the game: if you want the benefit of the patent commitment, you should join the organization, contribute to the work, and make your own commitment in return. This also serves as a membership incentive — joining the organization is the only way to get the royalty-free license.

The member-only approach has a practical downside: it limits adoption. If the goal is ubiquitous implementation of the standard, restricting the patent license to members creates a barrier. On the other hand, if the goal is to build a committed ecosystem of participants who all contribute and all benefit, the member-only model can make sense.

8.6 Reciprocity and Defensive Termination

Two provisions commonly appear in royalty-free policies that create conditions on the commitment: reciprocity and defensive termination.

Reciprocity

Reciprocity means that the royalty-free commitment is conditioned on the recipient also offering royalty-free terms back. If you want a license from me, you have to offer one to me as well.

This sounds fair, but the details get complicated. Does reciprocity mean "back to me" or "back to the world"? If I license my patents to you royalty-free, and the reciprocity provision says you must license yours back to me, that helps me but not the broader ecosystem. If reciprocity means you must make the same commitment to everyone, it effectively extends the patent commitment to non-participants who license from a participant — a much broader obligation.

How reciprocity is enforced in practice is even less clear. The bilateral license is negotiated privately, and the specific terms are generally confidential. A third party relying on a reciprocity chain has limited visibility into what was actually agreed.

Not all royalty-free policies include explicit reciprocity provisions. Some take the position that it's a perfectly reasonable bilateral negotiation term and doesn't need to be mandated by the policy. Others build it in as a standard feature.

Defensive Termination

Defensive termination allows a patent holder to revoke their royalty-free commitment against a party that asserts patent claims against them. The logic is straightforward: if you sue me for patent infringement related to this standard, I should be able to revoke my commitment to you.

This provision exists in most royalty-free policies and is generally uncontroversial in principle. The details — what triggers it, how broad the revocation is, whether it's limited to the asserting party or extends more broadly — vary across policies and can be significant.

Defensive termination serves an important deterrent function. It raises the cost of patent assertion within the standards ecosystem, because any assertion risks losing the royalty-free licenses the asserter depends on from other participants.

8.7 Comparing Major RF Policies

While the principles above are common across royalty-free policies, the details differ in important ways. Three frameworks dominate the landscape.

The JDF IPR Policy

JDF offers multiple IPR modes that a working group can select when chartering. The royalty-free mode provides a participation-triggered commitment — joining the working group creates the obligation, not the act of contributing. Scope is confined to the working group's charter. Exclusions follow a rolling model, placing the onus on the participant to track the spec and act before finalization. The policy is intentionally lightweight, designed for small to mid-size groups of like-minded participants where exclusions are expected to be rare.

The W3C Patent Policy

The W3C policy, first adopted in 2004 and updated several times since, was one of the first formal royalty-free patent policies. It uses a participation-triggered commitment tied to working group membership. A key feature is its formal exclusion process: when a specification reaches a defined maturity stage (Last Call or Candidate Recommendation), participants have a 150-day window to review and exclude. This is longer than most other policies and reflects W3C's emphasis on giving participants adequate time to review complex web specifications. The W3C policy applies on a working-group-by-working-group basis, and the commitment covers the specific Recommendation produced by that group. One area of ongoing debate is whether the commitment extends to subsequent versions of a specification — W3C has generally taken the position that new versions require new commitments.

The OASIS IPR Modes

OASIS provides three IPR modes: RAND, RF on RAND Terms, and RF on Limited Terms. Each OASIS Technical Committee selects its mode when chartered, and the mode governs all work produced by that committee. The RF on RAND Terms mode is the most commonly used for technology standards — it requires royalty-free licensing but otherwise follows RAND-style mechanics, including a formal disclosure and exclusion process. The RF on Limited Terms mode is more restrictive, limiting the scope of the commitment more narrowly. OASIS uses a formal call for patents tied to defined specification stages (Committee Specification, OASIS Standard), giving participants a structured window for exclusion. The specificity of OASIS's staged process makes it well-suited for organizations that want clear procedural guardrails.

Key Differences at a Glance

Feature	JDF	W3C	OASIS
Commitment trigger	Participation	Participation	Participation
Exclusion model	Rolling	Formal call (150-day window)	Formal call (tied to spec stages)
Scope	Working group charter	Working group	Technical Committee
Multiple modes	Yes (RF, RAND, W3C, ISO/IEC)	No (RF only)	Yes (RAND, RF/RAND, RF/Limited)
Versioning	Configurable	New commitment per version	Tied to specific spec version

8.8 Patent Exclusions: The Safety Valve

Any royalty-free policy that doesn't provide an exclusion mechanism should be a red flag. The ability to exclude specific patent claims from the royalty-free commitment is the safety valve that makes participation possible. Without it, the theory is that companies with significant patent portfolios simply wouldn't join.

Why Exclusions Exist

When you participate in a royalty-free working group, you're committing to license your necessary claims at no charge. That's a meaningful concession. The exclusion mechanism gives you a way to protect specific patents that you're unwilling to commit — either because the spec evolved in a direction you didn't anticipate, because you have a change in business circumstances, or because you have particular claims you need to preserve for other purposes.

In practice, exclusions are rare. Most participants in royalty-free organizations never exclude anything. The shared expectation — sometimes explicit, sometimes not — is that participants are entering a royalty-free zone and will behave accordingly. Exclusions are widely considered the nuclear option.

How Exclusions Work

Exclusion mechanisms come in two forms.

Call for patents. The organization issues a formal notice — typically when a specification reaches a near-final draft — declaring that participants have a defined window (usually 30 to 60 days, often settling at 45 after the usual negotiation) to review the spec and declare any exclusions. This is a specific trigger with a specific deadline.

Rolling exclusion. There is no specific call for patents. Participants have a standing obligation to declare exclusions before the specification is finalized. There's no formal trigger — the onus is on the participant to track the spec's evolution and act in time.

Both approaches have tradeoffs. A call for patents gives participants a concrete deadline and a piece of paper that prompts internal review — particularly valuable in large organizations where the standards team may not know what the patent team holds. A rolling exclusion is lighter on the organization and works well when participants are engaged enough to track the work themselves. For smaller, like-minded groups where exclusions are expected to be rare, rolling exclusions reduce administrative burden. For larger or more adversarial environments, a formal call for patents provides better discipline.

What You Must Disclose When Excluding

When a participant makes an exclusion, they typically must identify the patents or patent applications being excluded and the portions of the spec they believe are implicated. The level of detail required varies across policies and is itself a negotiation point.

For issued patents, most policies require at minimum the patent number and the relevant spec sections. For unpublished patent applications, the requirement is necessarily different — the application isn't public, so the policy may ask for a description of the subject matter rather than specific numbers.

Some organizations are moving toward more rigorous exclusion requirements: specific claim identification, mapping to spec sections, grouping by patent family, and even a good-faith attestation that the excluded claims are indeed necessary claims. The motivation for these enhanced requirements is experience with participants who have abused the system by mass-excluding hundreds of patents without meaningful review.

When Exclusions Go Wrong

Exclusions done in good faith, narrowly focused on one or two specific claims, with engagement and communication with the working group, are manageable. The working group can evaluate whether to design around the excluded claims or accept them.

Mass exclusions — where a participant searches their patent database by keyword and dumps hundreds of patent numbers — are a different matter. Standards bodies are not equipped to process that volume. It's highly disruptive, it undermines the royalty-free commitment, and it's generally regarded as bad faith. But if the policy allows it, the organization has limited recourse.

This is an area where policies are actively evolving. Newer versions of IPR policies are adding procedural requirements specifically designed to raise the cost of bad-faith exclusions — requiring more detailed disclosures, good-faith attestations, and in some cases limiting the scope of what can be excluded against existing spec material that has already been through a prior exclusion window.

Versioning and Exclusions

One of the most contentious questions in exclusion practice is whether a new participant can exclude against material from prior versions of a specification.

If a working group produced version 1.0 under a royalty-free commitment, and a new participant joins to work on version 2.0, can they exclude claims that read on the version 1.0 material that carries forward into 2.0? Reasonable positions exist on both sides.

The argument for allowing it: the new participant only committed to the working group when they joined. They shouldn't be bound by past versions they had no hand in developing.

The argument against: the new participant joined with full knowledge of the existing spec. They had the opportunity to review it. Allowing exclusions against entrenched, widely-deployed technology is massively disruptive to an ecosystem that relied on the royalty-free commitment.

Policies handle this differently. Some are explicit that the commitment covers only the current version under development. Others use formulations like "substantially similar purposes" to extend coverage to related prior work. Many are silent, which creates ambiguity that surfaces at the worst possible time.

Common pitfalls that apply across all patent policy types — scope creep, versioning gaps, withdrawal mechanics, and more — are covered in Chapter 10.