In politics or regulatory realms, principles often follow practices and action precedes theory. So it is in India, where the Telecom Regulatory Authority of India (TRAI) has ruled that zero rating is a violation of network neutrality.
Now TRAI is trying to actually define what network neutrality actually means. “Putting the cart before the horse” might come to mind, but that is a bit too harsh. The problem with network neutrality, all along, is that it is a concept devilishly hard to define, much less understand.
So it is paradoxical to hear TRAI chief executive RS Sharma say “he is open to the idea of internet content being provided free of cost or at discounted rates, just like toll-free phone helplines.” Huh?
That sounds like zero rating.
"We have no objection in general if someone decides to provide content free, or at a discounted rate, if the same is made available to subscribers of all mobile operators," Sharma said.
Such comments are one reason a new TRAI consultation paper on net neutrality is seen by some strong network neutrality (no zero rating) proponents as reversing or modifying the initial ban on zero rating.
For the cynical, the new interpretation is an attempt to wiggle out of a tight place. Sharma has said that such subsidized content “does not violate net neutrality” and is not in variance with TRAI's ban on zero rating.
One possible way of squaring that circle could be that sponsored data, or zero rating, could be deemed lawful if available to all mobile subscribers, not just subscribers of a single mobile provider.
TRAI, like all other regulatory bodies, faces the challenge of defining and implementing policies that some believe are harmful, not helpful, and difficult to codify, in objective terms, as rules.
As a matter of science or engineering, it is difficult to guarantee that “every bit will be treated the same,” in terms of delivery delay. Some would argue the obvious point that some apps actually require predictable delivery and low latency (interactive videoconferencing and voice being the best examples).
Nor is it easy to reconcile the fact that the purpose of lawful content delivery networks is in fact to ensure low-latency, predictable packet delivery. So non-neutral packet delivery is part of the existing fabric of consumer Internet access.
Nor have we ever been able to settle the question of how to manage network traffic without some forms of packet shaping, really.
Rationing or prices actually are the only two ways to manage traffic on networks, argues consultant Martin Geddes.
And one problem with bans on zero rating is that doing so precludes the development of “quality of service” differentiators. That, in fact, was precisely what proponents had in mind in requiring “best effort only” Internet access as the only level of lawful service for consumers.
But network neutrality rules essentially try to “protect competition in the app market” from ISP market power in a way that is not fact or science based, Geddes argues.
To be sure, regulatory harmonization always poses a big choice. Even when applying “the same” regulations to all contestants in a market, regulators must choose between “more” or “less” approaches.
Should markets be harmonized up or down; in the direction of more rules for all contestants, or fewer rules for all contestants; allowing robust competition or restricting it; applying legacy rules to new providers; protecting incumbents or letting innovation reign?
In the past, the question of how to regulate OTT voice and messaging has turned on precisely such questions.