The fierce debate over net neutrality continues unabated. On the one hand, advocates of net neutrality believe that without regulation, a few large network operators will dominate the market, discriminating against consumers and content providers to the detriment of affordable services and innovative content.

Opponents, however, believe that absent regulation, multiple competing broadband networks will flourish and support a plethora of innovative content and services. They contend that net neutrality regulations will only stifle innovation and curb necessary investment.

The FCC's Open Internet Rules, scheduled to go into effect next week, provide general guidelines for broadband providers, but leave much to interpretation. Regardless of pending lawsuits on both sides, some form of net neutrality will be implemented. Without more specific guidance, implementation will inevitably result in regulatory confusion and additional lawsuits, as both sides argue over what types of behaviour violate net neutrality.

To protect providers from uncertainty and clarify Internet users' rights, net neutrality rules must clearly define what constitutes a violation, how broadband providers can comply and what happens when a violation occurs. Stakeholders including industry, regulatory and public interest must reach accord on three pivotal issues.

First, regulators must define what measurable performance standards and network management practices violate net neutrality. Justice Stewart's famous test for obscenity, "I know it when I see it", will not suffice here. Because they may perceive the same behaviour as either harmful or innovative, opposing interests are unlikely to agree ex post facto on what constitutes a violation.

Ideally, efficient measurement mechanisms will inform consumers, allow regulators to monitor with minimal intrusion, and be easily and objectively reported by third parties or broadband providers. Such regulations will require two types of measurable information: (1) disclosure of network management policies, performance characteristics and service terms, and (2) a set of network performance measures.

On both counts, the fundamental questions are: Who will do the measuring? Where will the information be reported? And how will the information be used?

The answers will depend on the type of data being measured. For example, whether providers are meeting disclosure requirements may be best measured by consumer complaints, as envisioned in the FCC rules. Network performance measures may be more efficiently reported by carriers themselves, making costly third party monitoring unnecessary.

According to FCC sampling, carriers largely live up to their self-reported broadband speeds. Whether this information is actively monitored by regulators or reviewed in response to complaints would likely depend on the type of information and where it is reported.

Second, enforcement and corrections must be clearly defined and proportional to the seriousness of the violation. Broad regulation of a wide set of behaviors that might result in consumer harm would only stifle innovation without remedying an underlying market failure or addressing consumer harm. Instead, regulations must define both the violations and the appropriate, measured response to discourage negative outcomes, while protecting broadband providers who inadvertently violate a rule.

Once a violation is identified, the remedy must also be proportionate to the violation in order to correct the perceived harm. A graduated response with early notification would allow providers an opportunity to correct improper behaviour without disproportionately penalising providers and chilling innovation.

For instance, if a broadband provider is intentionally slowing broadband speeds undisclosed, regulators could increase monitoring and require the provider to increase disclosures. If the provider's behaviour continues to violate net neutrality after appropriate disclosures and heightened monitoring, changes in network operation would be necessary.

Third, recognising the constantly and rapidly evolving nature of the Internet and its content, today's rules may not be applicable five years from now. Therefore, regulators should limit this first phase of net neutrality regulation to an initial review period, two to five years, to assess the evidence of existing harm and potential for harm in the future.

If violations, complaints or new concerns emerge, further oversight and enforcement or other rule changes may be warranted. The dynamic nature of the Internet, in short, limits the likelihood that any static set of rules will be appropriate over the longer term, absent the opportunity for periodic review and modification.

Stakeholders on all sides of this debate now have a unique opportunity to shape the implementation of net neutrality regulations. It will be up to them to develop efficient and effective mechanisms for disclosures, define what will be considered inappropriate behaviour, and determine how these rules will evolve.