Question 4 of 7: Protecting an Idea

Question 4: Protecting an Idea (15 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Part A — Working with others before ideas are finalised

Before anything is filed, the only protection available is contractual and procedural, and it has to be in place before the first substantive conversation. In practice this means four things. Execute a written non-disclosure agreement with every external party — supplier, contract manufacturer, prospective customer, university collaborator — before any technical disclosure, with the confidential subject matter, the permitted purpose, the term and the return-or-destroy obligation all stated; a mutual NDA is normal where both sides will disclose. Keep dated, contemporaneous records: a bound or electronically timestamped design record, signed and witnessed, which establishes conception and diligent reduction to practice. Settle ownership in writing at the outset, because in Canada an invention made by an employee in the course of employment is generally the employer’s, but a contractor’s invention is generally the contractor’s unless the agreement says otherwise — and a joint-development agreement should state who owns foreground intellectual property, who owns improvements to background intellectual property, and who may license what. Finally, disclose in stages: share the problem and the interface requirements freely, and the enabling detail only under agreement and only to those who need it.

Part B — Challenges facing a new designer

Inadvertent public disclosure destroys novelty. A conference paper, a trade-show display, a crowdfunding page, a thesis in an open library or a sale can each be a public disclosure that defeats patentability. Canada and the United States allow a twelve-month grace period from the inventor’s own disclosure; most of the rest of the world, including the European Patent Convention, does not, so a disclosure that is survivable in Canada has already forfeited Europe and much of Asia. New designers routinely learn this after the fact.

Cost and cash flow. A Canadian patent is not the main expense; national-phase entry in several jurisdictions, translations, prosecution and the maintenance fees over a twenty-year term are, and they fall due long before the product earns anything. A portfolio wider than the business can maintain is a slow drain that eventually lapses at the worst moment.

Enforcement is the real cost, and it is asymmetric. A granted patent is a right to sue, not a right to be left alone. Litigation costs are substantial and largely independent of the size of the parties, so a right the holder cannot afford to enforce has limited deterrent value against a well-resourced infringer.

Freedom to operate is a separate question from protection. Owning a patent does not confer the right to practise the invention if it falls within a broader earlier claim. New designers frequently conflate the two and discover the problem only when a demand letter arrives.

Choosing the wrong instrument. Filing a patent on something whose value is in appearance, or keeping as a trade secret something that is fully disclosed by the product itself and reverse-engineerable in an afternoon, wastes the protection budget entirely. This is the choice Part C addresses.

Part C — Five options for protecting an idea

The five statutory and common-law instruments available in Canada, with what each actually protects, are set out below. The distinguishing question in every case is what is being protected: a functional principle, an appearance, a source identifier, an expression, or simply the fact that nobody else knows.

The Canadian frame matters in two specific ways worth stating. Canada has been first-to-file since 1989, so the date a complete application reaches CIPO ordinarily decides priority, not the date of invention — the laboratory notebook is evidence of many things but no longer of entitlement against a faster filer. And the twelve-month grace period under the Patent Act applies only to disclosure made by the applicant or someone who obtained the subject matter from them; it is a rescue, not a strategy, and it does not exist in most export markets.

Choosing between the instruments: patent or trade secret?

The patent-versus-secret choice is the one that is genuinely quantitative, because it is a comparison of two income streams with different risk structures. A patent yields a finite, reliable annuity and then stops; a trade secret yields a perpetuity that survives only as long as the secret does.

Given. A licensing income of 150,000 CAD per year attributable to the idea, a discount rate of 8 per cent, a 20-year patent term with 3 years of pendency consumed before the royalty begins, and an estimated annual hazard of the secret leaking or being independently discovered of 15 per cent. Find. The present value under each route, and the leak hazard at which the two are equivalent.

1. Value the patent as a finite annuity. Seventeen years of enforceable royalty remain after pendency, so $$PV_{\text{pat}}=R\,\frac{1-(1+i)^{-N}}{i}=150{,}000\times\frac{1-1.08^{-17}}{0.08}=1{,}368{,}200\ \text{CAD}$$ giving $\boxed{PV_{\text{pat}}=1.368\ \text{million CAD}}$.
2. Value the trade secret as a decaying perpetuity. If the secret survives each year with an annual hazard $\lambda$, discounting and survival compound, so $$PV_{\text{ts}}=\frac{R}{i+\lambda}=\frac{150{,}000}{0.08+0.15}=652{,}200\ \text{CAD}$$ and $\boxed{PV_{\text{ts}}=0.652\ \text{million CAD}}$. At a 15 per cent hazard the patent is worth 716,100 CAD more — it more than doubles the value of the idea.
3. Find the hazard at which the two routes are indifferent. Setting $R/(i+\lambda^{*})=PV_{\text{pat}}$ and rearranging, $$\lambda^{*}=\frac{R}{PV_{\text{pat}}}-i=\frac{150{,}000}{1{,}368{,}200}-0.08=0.0296$$ so $\boxed{\lambda^{*}=3.0\ \text{per cent per year}}$, which is an expected secret life of $1/\lambda^{*}=33.8$ years. The secret must therefore be the kind of thing that stays secret for a third of a century before it beats the patent — which is precisely the Coca-Cola case, and precisely why it is the exception everybody cites rather than the rule anybody follows.

The practical rule that falls out of the arithmetic is the one to state: patent what the product itself discloses, and keep secret only what it does not. A cyclone geometry is visible in any unit sold and would be reverse-engineered in weeks, so its hazard is very high and it must be patented. A syrup formulation, a heat-treatment schedule or a tooling parameter leaves no trace in the shipped product, so its hazard is genuinely low and secrecy can win. Most products need a combination of instruments rather than a single one — a patented mechanism, a registered appearance, a trade-marked name, copyright in the firmware, and secrecy over the process settings.