While the foregoing represents the canonical view of the question, it appears that there are challenges to this commonly-held wisdom. The first heresy comes from the world of legal practice. This Kat suspects that there are few readers who who have not encountered a clause in a grant for the right of use of a trade secret that provides for a fixed time limit for the secrecy undertaking. This form of clause seems more typically to emanate from the party who is the recipient of the disclosure, and that makes perfect sense. It is difficult to come up with circumstances in which the disclosing party has an interest in limiting the period of time that the recipient is obligated to maintain secrecy of the information. The only situation that I can come up with is where the disclosing party actually intends to compete with the recipient, and the contractual ability to disclose the trade secret might actually confer a competitive advantage on the disclosing party. But such a scenario seems pretty far-fetched.
The argument most typically raised by the recipient in favour of a fixed time limit for the secrecy undertaking is one of mutual convenience. The claim goes something like this:"Because of the rapidly changing circumstances in the industry in which the secret is being used, the contractual term for the secrecy obligation (say five years for purposes of discussion) provides a degree of certainty to both parties. In any event, within five years, the value of the information will be nil. But to avoid any possible dispute that there has been an unauthorized disclosure, the parties agree in advance on the period in which the trade secret will be protected. In so doing, both parties can then go with their business."
"But is not clear that this indefinite term properly strikes the balance between providing incentives to invent and ensuring that the world benefits from the new invention. It may be that after a certain period of time the additional incentive from the prospect of secrecy is marginal, while the costs of maintaining secrcy are not. ...One possible implication of treating trade secrets as IP rights, then, is that the law should provide that trade secrets "expire" after a certain period [footnote omitted]."Lemley argues that there are potential various advantages in providing legal incentives to maintain secrecy of inventions in certain circumstances. From this vantage, what is key is how to best calibrate these incentives. One way to conceptualize this calibration is establish a fixed period of time for protection of the trade secret, after which the value of the secrecy is outweighed by the various costs in maintaining such secrecy. Having raised the point, however, Lemley is very chary about actually proposing how such a time limit might work. He concludes that "[i]t may be too hard to decide on a start date, and therefore an end date, and compelling disclosure of informatiom at the end of the term may also prove problematic."