"not technologically possible" - Why language matters in a regulatory setting
In a recent reply to CRTC file 8638-B2-201905879 (Application by Bell Canada for Commission approval to allow the Bell Companies to temporarily block certain verified fraudulent and scam voice calls on a trial basis under section 36 of the Telecommunications Act) Bell Canada stated that making this trial an opt-in process that only applied only to traffic terminating to Bell customers was "not technologically possible". This statement was one of the key facts used in the subsequent CRTC decision to allow the application.
This post isn't about the merits of the Bell application - my views on that are already well documented on the public record. This post is about the nuance of language and the implications that can have in a regulatory setting.
There are many things in the world that are not technologically possible - for example, it is not currently technically possible to get to Mars in less than 30 days - the fastest possible time is 39 days with an average time of 162 days. It's also not currently technically possible to drive a land vehicle faster than 655.722km/h - that's the current land speed record. Getting to Mars in 30 days and driving on land at 700km/hm are things that are "not technologically possible".
But in the context of this Bell regulatory filing, the phrase doesn't make sense. Many companies provide call filtering services that are both opt-in and targeted only at terminating traffic - Primus Canada with it's Telemarketing Guard product, Telus with it's Call Control service, and Shaw's Call Blocking service all do this just to name a few.
Reading between the lines, and based on my understanding how these technologies work from my experience building Primus Telemarketing Guard, what Bell Canada is actually saying is that adopting this trial to be opt-in, or adopting it to be subscriber facing only, is ether not technologically or commercially feasible for them. And this is the critical difference - many things are not technologically feasible for a variety of reasons - cost, effort required, regulatory restrictions, etc, but that doesn't mean they are impossible.
Flying cars are feasible today from a technology perspective, but they are not widely deployed for a host of ancillary issues - safety, regulatory, cost, infrastructure, etc. No one would say they are not technologically possible, they would just say they are impractical or unfeasible.
Back to the filing by Bell Canada, I can completely understand if Bell had implementation issues modifying their call blocking technology to work at the subscriber level - perhaps it requires a large network investment, or perhaps it requires excessive cpu or network resources, but those are all solvable problems if Bell was willing to spend the required resources to solve the issue.
What Bell should have said to the CRTC this case is "It is not commercially feasible to limit the Trial to terminating calls or to implement an opt-in procedure, as some Interveners have suggested" instead of not technologically possible. Implying something is impossible leads the regulator to believe there is no other choice - it is the solution as proposed by Bell or nothing. This mis-framing of the issue works to the advantage of Bell, but is a major disservice to the regulator who may not have the domain knowledge required to properly evaluate these claims.
Had Bell phrased the problem as one that is possible, but perhaps not commercially viable, then decision from the CRTC may have been different - they may have denied the application and suggested that Bell undertake a feasibility study to inform the regulator of what would be required to implement these simple consumer safe guards. Instead, Bell misled the CRTC through this deceptive language, and as a result, obtained the regulatory outcome they desired.
We, as an industry, need to call out these misstatements when we see them. To some, there is little difference between technologically possible and technologically feasible, but in a regulatory setting, language really does matter.