Some of the privacy-minded readers of this blog might sense some irony in my use of Google’s codification of the corporate ethics it espoused until around 2015. It was replaced by “do the right thing” around that time, and to my knowledge, both sides of a similar moral coin still exist within the Silicon Valley behemoth today.
In his most recent post in a series on Nuisance Calls our founder, Simon, gave a brain dump about reputation scoring (which the eagle-eyed will have noticed I fettered in editing) as a tool against nuisance calls.
There’s a reason for that; it’s because telecoms is a highly regulated industry, with various conditions upon each operator in the UK. Similar conditions exist worldwide, and our friends abroad will invariably find analogues, but for now, we’ll illustrate with the British approach.
I’ve made my position on the anti-competitive risks of specific nuisance call initiatives clear in the past and am currently engaged in a number of very, very lengthy (think longer than the Daily Mail sidebar of shame) Slack threads with Simon and Charles. Simwood has recently taken a quantum leap forward in nuisance call intelligence and action by moving things out of the lab and into production, which makes Ofcom’s recent extolling of BT look like amateur hour.
This evolution has been so significant that the debate has moved from “Can we do this?” to “Should we do this?”. Or, to put it another way, “Are we evil?”
In our lab environment right now, there exists the ability to implement a form of reputation scoring and feed that into the blocking algorithms. Before our haters get all excited and Ofcom’s enforcement team’s inbox is full of half-baked allegations (probably by those it would affect, with good cause), this is not in production.
At the moment, the lion’s share of scrotitude on our network is coming from a single source. Not only is it an order of magnitude more than coming in from BT (who I would’ve assumed before this would be top the league because of their market share as well as IPEX Type A being a known destination for various miscreants), but if one adjusts for market share, you get 130x more scrotitude in from the worst offender than BT.
Some phrases rarely leave my mouth often. This is just such a case: in fairness to BT, considering their market share, they are not the worst offender. The worst two, depending on metrics, are those that regularly wrap themselves up in a shroud of virtue. Those who know me know I abhor hypocrisy more than any other character flaw. Still, it is also possible that incompetence and a lack of investment are to blame rather than a malicious strategy.
Anyway. I digress. Let’s give these two sources of scrotitude a hypothetical name – Plank Telecom. We know that all things being equal, there is a 130x greater probability that a call coming in from Plank Telecom, relative to BT, is dodgy. That could be dodgy in terms of just insufficient attention paid to General Condition C6.6 and CLIs, or calling unallocated numbers, or any of the other metrics we use. And on that, while I would love to tell you precisely all the ways we detect scrotes on our network, this is not intended to be a Hayne’s Manual for fraudsters.
This would suggest that BT (again, mark this date in your calendars) should benefit from the doubt relative to Plank Telecom. But the world is not that simple.
Every range holder in the UK has a monopoly in terminating calls to its numbers. Ofcom reviews the remedies for that monopoly every few years. They are a little lop-sided, as BT has more explicit ones, but the theme of non-discrimination exists. For BT, that is an explicit condition; for everyone else, it is implied within “fair and reasonable”.
If I take action based on the identity of the incoming interconnect, am I discriminating? Have I committed a mental faux pas equivalent to a biased cop seeing a teenager of a particular ethnicity in a certain neighbourhood? On the face of it, yes.
Joe Bloggs on the street may have no idea that his friendly local IT support company from whom he’s plugged his 3CX gets their numbers and connectivity from Plank. In the same way that poorly configured parental controls wipe Scunthorpe from the map, Joe Bloggs’ growing small business suddenly can’t make calls, purely because Plank are the network of choice for scrotes.
These are the reasons we have end-to-end connectivity obligations (be they explicit or at least inherent in the very definition of our industry). The industry side is why we have non-discrimination obligations. It would be a substantial moral hazard for a large incumbent provider to wrap themselves in a shroud of virtue and play God with traffic in the name of nuisance calls while commercially benefitting.
BT would never, would they?
In the same way, the Royal Mail can’t know whether every item of post contains illicit drugs, we cannot know what the contents of every call are. But mere conduit is a defence that only takes you so far. Not knowing is one thing. Turning a blind eye is another. And should our friends in Riverside House start scratching at the surface, there will likely be some surprises.
In the meantime, this debate, as you can see, has moved from being done and dusted technically, but that’s the easy bit. The tricky bit is only just starting as we enter the next market review and the next five years’ rules of our industry.
Oh, and it is possible to make a significant dent in nuisance calls without paying millions for an unfit-for-purpose central database (yes, some are still blathering about that). You’ve just got to actually want to stop the calls, even if it means you forgo margin to stop them getting to the vulnerable in society.
