Summary

  • Cory Doctorow’s 2019 EFF essay argues that a rival should sometimes be able to connect to an incumbent service without its permission, giving users an ongoing route to leave. The co-authored 2021 EFF report with Bennett Cyphers adds privacy rules and three distinct policy tools.
  • In a 2023 hearing on Canada’s Bill C-294, Alissa Centivany attributed the phrase to Doctorow and EFF. Doctorow was not a witness. The committee later reported a limited amendment, and Bill C-294 ultimately received Royal Assent for a narrow TPM exception involving computer programs and devices; this is not a general platform-interoperability mandate.

An exit is more than an export

For Cory Doctorow, the practical test of competition is not simply whether a user can download a copy of their data. A person may export a profile, address book or archive and still lose the live relationships that made the service useful. In his June 2019 essay for the Electronic Frontier Foundation, Doctorow calls for adversarial interoperability: a newcomer builds compatibility with an established service even when the incumbent does not cooperate.

He uses office-document formats to illustrate the difference. Apple’s iWork suite could read and save many Microsoft Office files, Doctorow writes, despite Microsoft not designing that compatibility for Apple. He treats the history as evidence for a broader claim: compatibility can let a challenger reach people who remain on the dominant service, lowering the social cost of trying another one. These examples and the essay’s claims about market effects are Doctorow’s case for the policy, not independent measurement of what every compatibility layer achieves. Read Doctorow’s 2019 EFF essay.

Portability and interoperability solve different problems in that argument. Portability is usually a one-time transfer of a user’s data. Interoperability can preserve an ongoing connection while a person or group moves between services. The distinction is economic as well as technical: if a rival must rebuild every contact and workflow from zero, an export may offer little immediate relief from the incumbent’s network advantage.

“Adversarial” also describes a relation between companies, not permission to disregard users. The entrant is acting without the incumbent’s consent; the user still needs control over which data move, which agent can act, and how abuse is handled. The term names a challenge to incumbent control. It does not settle the design of a safe interface or identify who should authorize the risk.

The privacy argument belongs to two authors

Two years later, Doctorow and Bennett Cyphers set out a more detailed policy design in EFF’s Privacy Without Monopoly: Data Protection and Interoperability. The paper names three tools: competitive compatibility, or ComCom, which would protect some third-party connections made without permission; back-end interoperability, under which very large firms would maintain interfaces for users to interact across services; and delegability, which would let users authorize third-party software to interact on their behalf. The authors also distinguish these proposals from data portability. The EFF report treats these as related but separate interventions.

The report does not dismiss the risks that come with new connections. It identifies data sharing and mining through new APIs, phishing and sock puppetry in a federated environment, and the loss of some platform discretion to defend a centralized system. It also notes that third-party tools can mishandle information already visible to users. Cyphers and Doctorow argue for informed consent, data minimization, privacy law and enforcement; they contend that trustworthy rules should come from democratically accountable law rather than a company’s unilateral choice.

Their conclusion that the benefits can outweigh the risks is an advocacy position, not a field trial proving that every implementation will do so.

That qualification matters when the proposal enters a different policy setting. A legal exception to a copyright lock can make one form of access lawful; it does not build an API, supply a security model, guarantee consent, or establish a right to cross-platform communication. Nor does an interface itself prove that a third party is entitled to represent every user whose data could flow through it.

A witness cites the phrase

On March 8, 2023, the House of Commons Standing Committee on Industry and Technology (INDU) in Canada heard witnesses on Bill C-294, an amendment to the Copyright Act concerning technological protection measures. The official attendance and testimony record lists Alissa Centivany, Anthony D. Rosborough and others. It does not list Doctorow. Centivany, an assistant professor at Western University, opened by defining interoperability across products, services, interfaces and parts. She said the adversarial form had been coined by Cory Doctorow and EFF, describing it as a new product or service connecting to an existing one against the original company’s wishes. The committee transcript is the evidence for that attribution.

Centivany was offering her own testimony about a bill already before the committee, not conveying a statement from Doctorow. She argued that C-294 did not go far enough and proposed, in descending order of ambition, removing anti-circumvention rules, mandating interoperability instead of permitting it only in limited circumstances, and creating broad immunity for acts aimed at interoperability. The committee heard other views. Colleen Stanley, speaking for the Intellectual Property Institute of Canada, raised uncertainty about the bill’s drafting and warned that circumvention could expose private information or affect safe operation.

Centivany and Rosborough responded that legitimate safety and security interests should be protected through laws designed for those purposes rather than by copyright restrictions. John Lawford of the Public Interest Advocacy Centre supported an aftermarket-interoperability exception while emphasizing the bill’s narrower focus.

These arguments make a hearing useful evidence of contestation, not a test that decides which claims are empirically correct. A witness’s expertise and participation can sharpen a question; neither one makes the witness the representative of all users or gives her authority over every affected tradeoff. Attribution likewise does not transfer the attributed author’s mandate to the person speaking.

What Bill C-294 actually changed

The subsequent institutional record is more specific than the phrase. The committee agreed on March 27, 2023, to report Bill C-294 with an amendment; Report 13 was presented three days later. The House of Commons of Canada adopted the bill on June 14. It received Royal Assent on November 7, 2024. These records establish the committee’s formal recommendation, the House’s recorded vote and the enacted text. They do not, by chronology alone, establish that Doctorow’s 2019 essay caused the amendment or that the committee adopted Centivany’s broader recommendations. The committee study page records the process, and Report 13 records the committee’s text.

The enacted law allows a person, in specified circumstances, to circumvent a technological protection measure protecting a lawfully obtained computer program to obtain information for interoperability or to make the program—or a device in which it is embedded—interoperable with another program, device or component. It limits how communicated information may be used and excludes conduct that infringes copyright; one subsection also excludes conduct that violates another federal or provincial law. The Royal Assent text is the controlling source for that scope.

This is a legal opening around certain program and device locks. It is not a law requiring social networks to preserve cross-platform messaging, expose their back-end interfaces, or let third-party agents act for users. It does not implement the full policy programme in the 2021 report. The records establish an overlap in vocabulary and subject matter, but do not show a direct line of authorship or influence from Doctorow to the statute.

Doctorow’s name therefore appears in the policy journey in a bounded way: first as the author of an argument, then as the subject of another witness’s attribution. The committee made its own decision through a report and the House enacted a specific rule. Keeping those roles distinct does not diminish the force of his idea. It makes clear what the public record can support—and where a claim of representation or adoption would begin to exceed it.

Sources