Frustratingly confined to my sick-bed throughout this year's Mobile World Congress, there were one or two small consolations to be taken from missing the greatest telecom show on earth: I could get a cup of tea whenever I wanted one; and I could switch off the media noise and watch the garden birds preparing for spring.

But most of all, I could, for once, see the show from the outside looking in. Instead of burrowing my own topic-specific tunnels through the forest of booths and their caffeinated sales folk, I could see what matters to Joe and Jane Public.

And from that perspective, nothing, really, could be plainer. To the world at large, it's all about apps and devices. The big stories as viewed from my sick-bed were, in no particular order:

From this point of view, the network (and its supply chain) is little more than an afterthought: annoying when it fails, but otherwise invisible. SDN, NFV, NG-OSS and all the acronyms that keep those inside the network industry awake at night are of no more interest to the world outside than, say, new kinds of road surfaces, or the latest gas turbine.

No doubt that sounds a little unfair to those who toil heroically to keep their networks relevant in an end-user environment that is changing at frightening speed. The fact that telcos have been able to cope (just) in an environment where they control less and less of the device (and app) value chain is little short of miraculous.

But the world can be an ungrateful place, and viewing MWC from the outside in raises a larger and even more uncomfortable question in my mind. If telcos over the past 20 years have steadily lost control over the devices connected to their network, as well as most of the applications that run on them, what will they be selling to Joe and Jane Public in, say, 10 years' time?

It's hard not to follow the logic of the past ten years and answer: nothing but a connection and the bandwidth that goes with it. TV, telephony, messaging, and all the rest, on this reading, will have gone the way of every other app -- disappeared into Whats App, Facebook, Amazon Prime, Netflix, and other start-ups yet to be born. No longer will these apps and services be called, absurdly, "over the top" -- they will constitute 100% of what customers value.

There are, of course, no sure bets about the future. Many telcos are still deeply embedded in the business of providing services, and won't easily abandon a core source of value and stickiness. In fixed networks, triple play is still the default package in many countries. In mobile networks, a different kind of triple play (voice, messaging, bandwidth) rules. But for how much longer?

The question then will be whether telcos can develop enough ancillary capability to earn more money from their consumer customers, not just directly but indirectly. Ancillary capability means not only the services themselves, but the service enablers, perhaps paid for not by the users but by content and applications providers, or by enterprises. Things like security & parental control, guaranteed delivery, QoE, analytics, and other goods that network operators are well positioned to provide.

Small wonder, in light of that list, that network neutrality -- or rather, its absence -- matters so much to the telcos. With so much at stake, we have not, I think, heard the last of that debate.

Whatever the outcome, a great deal is at stake. If telcos can't uncover new sources of revenue (or persuade regulators to give them special breaks) the relentless focus on cost-cutting will continue. But if they can, new vistas will open up for both them and their suppliers. One way or another, service innovation -- or its absence -- will be the key to the future.

— Graham Finnie, Chief Analyst, Heavy Reading