Everyone is all of a sudden very excited about Caller Line Identification, or CLI, at the moment.
On October 1st 2018, new General Condition C6.6 comes into force in the UK obliging certain communications networks in the UK to prevent calls from reaching the called party where the CLI doesn’t have integrity.
Or, rather, that’s the headline that’s going around. Unhelpfully, the Office of Communications (Ofcom) have not made things explicitly clear to network operators on the details. So, let’s unpack it a little bit.
C6.6 Where technically feasible, Regulated Providers must:
(a) take all reasonable steps to identify calls, other than calls to Emergency Organisations, in relation to which invalid or non-diallable CLI Data is provided; and
(b) prevent those calls from being connected to the called party, where such calls are identified.
What I have noticed is that many communications providers think that this is a requirement to block all calls with any irregularity in the CLI.
Well, it is and it isn’t, as often is the case with Ofcom regulation. If you are the first point of ingress into the UK, say the UK termination partner for USA Telecom and they send you duff headers, then there is an Ofcom sanctioned mechanism to assert the header information and carry on as normal. In fact, Ofcom have even designated some ranges in the National Telephone Numbering Plan for that purpose.
If you are the Originating Communications Provider, or OCP, you have a specific obligation – in fact, this hasn’t changed. Old General Condition 2 (renumbered A2 on 1st October 2018) requires adherence to a plethora of standards, which invariably include CLI. Such adherence is likely to feature in interconnection agreements too. So that’s nothing new. What’s happened is an exponential increase in the risk that someone downstream blocks your calls because they don’t like your SIP header.
In a short value chain where a BT Enterprise customer calls a Sky subscriber, this is unlikely to be a problem, especially if the parties are using TDM because the relative difficulty in manipulating the signalling means the CLI by definition usually has more “integrity” from the offset.
Where this gets interesting is in the world of Internet Telephony Service Providers (ITSPs) and those that sit (and switch) in the value chain between a major “Tier 1” network like BT, TalkTalk, Gamma or Vodafone (I refer to these bigger guys as “Lead Networks” later) and the calling party.
Many of these ITSPs have number ranges allocated by Ofcom; I am even aware of cases where their customers have ranges too. This makes them all Public Electronic Communications Networks by definition (as it’s a prerequisite to number allocation) and are therefore in scope of a veritable truckload of regulation. Their Lead Networks will monitor and assert CLI information into SIP headers where appropriate, or where agreed, or not as the case may be. I am sure, in the coming weeks, there’s going to be a flurry of activity between their voice engineers and their customers to fix irregularities before the 1st October deadline.
These same ITSPs will often have their own customers with switching infrastructure; being Cloud PBX operators or application developers, for example. Are these ITSPs the originators or mere transit providers?
In the literal sense, they are transit providers as they are receiving signalling and media from another network and routeing it accordingly, but as the first “fully functional” communications provider in the call path, they are also arguably the OCP.
This is where I think Ofcom consider the General Condition A2/C6.6 responsibility/obligation to comply with the technical standard will bite. But would you block calls? Obviously, this operator has a vested interest to either fix the signalling themselves or get their customer to send it with integrity, otherwise there’s no money to be earned from a call blocked downstream.
But there’s an elephant in the room here.
What if the called party wants to receive calls come hell or high water? We already know, for obvious reasons, that the Emergency Services are in such a category. Granted, your average suburban family don’t want to be asked about PPI over Sunday lunch, but many businesses will place a greater weight on receiving genuine calls even if that means getting the odd spammer.
It’s not just hypothetical. For example, I have worked with a communications provider that serves oil rigs off the coast of Africa (calls routed via satellite with its SIP endpoint located in the UK); you can imagine the state of some of the CLI information in the calls question and the importance of them too.
Equally, your Aunt Jean using ViberOut to call you on your birthday from their retirement home in Adelaide is going to want their call to connect and you’re likely to begrudgingly want it to connect too. Yes, Ofcom have made provision for the internationally originated calls, but with there being so many applications out there and so many routes a call can take through hundreds of UK network operators, I am not convinced that the assertion of a special number process is going to work reliably. If I am right, this will have the effect of an Ofcom own goal in relation to consumer choice and prices.
Which comes to the back to the elephant – the use of “reasonable steps” in the condition. If the communications provider serving the called party is instructed by them to send all calls, then (subject to technical considerations) blocking would be unreasonable. And that’s why I would suggest that only the Terminating Communications Providers should ever block calls under this condition, because they are the only entity in the call path without the information asymmetry regarding the called party. That said, I suspect this is going to be a battle line between communications providers that service predominantly businesses and their cousins that serve predominantly residential consumers.
So, where do we get to?
Subject to all the usual caveats and assessing business models and specific circumstances on individual merits, my starting point is that:
- OCPs have an obligation to ensure they adhere to technical standards and either assert information or discharge a duty of care and advise their upstream customers accordingly. This means application developers and PBX manufacturers all have a part to play, but it’s likely Ofcom will come knocking on the door of the first entity that looks and feels like a traditional network in the first instance. Whilst a literal read of the regulation could lead you to conclude blocking should be performed, when taking a holistic view of the issues, it should be more of a last resort than a first resort.
- Transit operators need to assert the right information when they reach the first point of ingress into the UK if the CLI doesn’t have integrity, otherwise pass the call on through as they have done time and memorial.
- Terminators should block calls with bad CLI information, unless acting upon the instructions of the called party to the contrary.
It’s a bit of a minefield – do get in touch and we’ll be happy to see if we can help.
Oh, and the subject of what constitutes an “invalid” CLI is a whole thesis, let alone a separate blog post to follow!
P.S I am considering creating a blog category called “Pedant’s Corner”, in which the following would be the inaugural post. General Condition C6.6 can be taken as Regulated Providers not having a requirement to take reasonable steps to identify calls to Emergency Organisations where the CLI Data is invalid or non-dialable at any time.
That’s a literal interpretation of the condition and one clearly at odds with General Condition A3 (as renumbered from 1st October) as compliance with that obligation requires, by definition, CLI information to have integrity. I am pretty sure that Ofcom did not intend for that meaning to come across and in any event, every communications provider I know takes emergency services calling seriously and will monitor irregularities and take action accordingly regardless.