The notion that new services (not cost reduction alone) will justify the New IP is very attractive to operators and is gaining ground every day. If we could only increase network agility, the theme goes, we could develop new services that would drive new revenue. It's a matter of faith that these services exist and that network modernization would realize them, but even operators are confused about the details for both technical and regulatory reasons. Interestingly, there's been a pathway available all along.
Broadly, new network services fall into three categories: the service of connection that's "in" the network, hosted services that are "on" the network and connection-point services related to the point of service delivery. Obviously we have all these service types today, and we can offer new in-network services like on-demand bandwidth with current network technology. Consumers and enterprises already have firewalls and NAT and the other connection-point services that everyone talks about. It seems, in fact, that two of our three categories aren't really "new" at all.
What does that leave us? The hosted service option.
Operators have long envied the way that OTTs could jump out to address new opportunities and quickly gain traction and revenue without the enormous infrastructure deployments with which operators contend. Operators could go after the same opportunities now, but they'd face some significant barriers that OTTs never dealt with, mostly from regulators.
In much of the world, incumbent telcos are prohibited from offering OTT services from their core business entity; they have to build a subsidiary. Further, many regulators would then require that subsidiary to offer regulated services to other OTTs under the same terms they were offered to the subsidiary. This means telcos could not leverage their current networks to differentiate their OTT offerings, and that means telcos would start from behind in an OTT competition without a clear path to catch up.
We could argue, as some have, that relaxing net neutrality rules to let operators compete more freely in the on-network OTT space could help, but that would likely be a long and bloody lobbying fight. Not only that, but all it would do is help operators be OTTs, which doesn't get the network itself modernized. What we need is a way of linking OTT services to networks so that operators can differentiate technically without spending all their time in court. If we could do that, we could fund the New IP and also gain some insight into what services it might offer.
The path to the New IP may have already been established, for regulatory purposes at least, with CDNs. Technically speaking, CDNs are caches on the network combined with address management that's in the network, so they bridge the two worlds for operators. From a regulatory perspective, they're immune from neutrality rules because they're part of the delivery network for content. If operators can define other useful delivery-network services like CDN, they have a pathway to invest in the New IP in a way that can differentiate their offerings and increase their revenue.
The CDN Surge
Like IoT, CDN encompasses surging demand for data and devices. (Source: Statista)
Following CDN's example, operators can define the cloud and big data elements of new services in the same way they define cache points. Like cache points, these new elements wouldn't be "on" the network in a traditional sense of having a URL you can click, but would be accessed as a part of a service, which in the case of CDN is content delivery. This indirect access is the key to keeping a new service and its infrastructure out of the regulatory trap of net neutrality.
An example is probably the best way to visualize this, so let's look at IoT. You could surely put a bunch of sensors and controllers directly on the Internet, in which case they'd look like websites and everyone could access them. Apart from the obvious (and much-discussed) security and privacy issues this would create, it makes it difficult to monetize the sensor deployment. But suppose we had a "Thing Delivery Network" or TDN?
A TDN would collect sensor data, analytics tools and control and policy elements into a cloud application that would live inside the carrier network. Access to the TDN would be through a web service that would receive a request and return a response. Do you want to know where your partner might be stuck in traffic? You could inquire based on a registered partner name. The TDN would validate the security/policy credentials you and the registered partner offered, then do a database query to identify the sequence of sensors that reported the partner nearby. That creates a virtual track than can be laid onto a map and you have your partner's progress in a convenient form.
This doesn't destroy the openness of IoT to innovation, because new application elements could be added to the TDN providing that the application provider paid an access fee (to cover deployment cost) and complied with the security/policy limits set by regulators. It does provide a pathway to network operators to earn revenue without demanding that every sensor and controller pay a 4G/5G phone bill. It also shows that a CDN model can be applied to new opportunities, and that TDNs could pave the way for profitable operator participation in the New IP.
— Tom Nolle, President, Founder and Principal Analyst, CIMI Corp.. Follow him on Twitter @CIMICorp. Special to The New IP.