In an IP(IMS) world, the mobile device will drive convergence because services shift to the edge of the network – and devices are at the edge of the network ..

In an IP(IMS) world, the mobile device will drive convergence because services shift to the edge of the network – and devices are at the edge of the network .. A long title but one which encapsulates the content of this post. I covered this topic last week at the Informa IMS world conference in Monaco and also briefly today in my panel discussion at the Cambridge Wireless network.

Following on from my previous post about services in a post-IP world , at IMS 2.0 world forum there was a panel called Establishing collaborative operator device specifications for IMS handsets. This panel was unique because even before the panel started, there were two hands asking questions. The second was me .. The first was Usama Mansur of UAE based operator Du . It turned out, we both had the same question .. And the question was:

The most important people were missing on the panel ..

With all due respect to the folk at omtp , if we are talking of devices and IMS, we want to hear from the big four : Nokia, Motorola, Sony Erissson and Samsung. Where are they?

Where indeed ..

I alluded before that devices are going to play a far more significant role in a post – IP future. I will use Motorola and Nokia to highlight the points here ..

Let us look at a basic principle:

We all agree that the network is becoming IP based. Also, most devices are now supporting multiple ‘networks’ (i.e. modes of access)– for example Cellular, WiFi, Wimax(in case of Motorola), DVB-H(in case of Nokia) and so on. Thus, the device is becoming customer centric(because customers don’t care what network they connect to) and network agnostic(because customers can connect to many networks).

We should not be surprised by this.

This phenomenon is natural fallout from an IP network i.e. in an IP world, the network becomes dumb. The services shift to the edge of the network and the device sits at the edge of the network .. Hence my belief that the devices are going to be very significant because they put the customer at the centre of the equation and become a logical driving point for richer services across network types(access mechanisms).

Thus, the device manufacturers are looking beyond Cellular(and beyond IMS).

Note the change in terminology from device manufacturers : Nokia calls itself an Internet company. Motorola speaks of ‘Seamless mobility’.

Today we have a device accessing a network – tomorrow we will have a seamless network(not necessarily cellular) being accessd by a device.

Here is what it means for me, as a customer.

When I start up the device, I am offered a choice of the best network I want to connect to(and that could be a WiFi network, A Wimax network or a cellular network) and as I use the service, the optimal network is being accessed for me seamlessly by the device.

The device thus becomes very important – the network less so.

This could change the dynamics of the industry. Currently, devices are subsidised. In future, that subsidy may have to increase to retain the best devices(best being defined here as ‘what customers want’). Note that the best devices may also be totally non subsidised(maybe ad funded?). A non subsidised device(with greater connectivity) could be actually cheaper in the long run because you would save on the operating costs. But note that ‘cheaper’ is not the critical issue here .. I believe customers will pay for a device which is more connected (in contrast to one that is not). And the customer is at the centre of this convergence because as customers, we want more choice and more connectivity.

In a sense, while we debated ‘Convergence’ – it has already happened at the device level i.e. networks have converged within a device.

This connectivity is also impacting applications and services.

In a creation driven world, the predominant requirement of the client will be to communicate. Yes, UI is important – but not at the expense of communication. Web 1.0 world was based on consumption of content. Web 2.0 is based on creation of content. As creation of content becomes more important, ubiquitous (cross network/cross access type) connectivity becomes critical because without seamless mobility/converged networks/Internet ecosystem(call it what you will .. ), we cannot share user generated content!. Thus, services will become seamless across networks.

Nokia presentations at IMS world talked of Movial – a company that has an interesting application that runs on SIP alone(and quite well as far as I can see). Similarly, a Motorola white paper called: Motorola Seamless Mobility

Connectivity Architecture(pdf) gives the table below distinguishing between IMS/non-IMS/session and non session based services.


Thus, coming back to IMS, my simple conclusion is: Device manufacturers are seeing a much wider and a richer ecosystem(which we, as consumers want). They are seeing beyond IMS – and to convergence and communication.

All comments/feedback welcome


  1. edsalo says:

    Hi Ajit,
    I agree with you that services will be driven to the edge of the network. However in your post you forget that application servers are also at the edge of the network. Therefore it is also possible that services will continue their transition towards client-server architectures in which the intelligence is in the actual server and this server talks to standard, fairly dumb, client devices. In this case the devices are merely “browsers” of content that is processed by and sent out from the servers. The only reason to put intelligence in the device would be to work around QOS issues in the actual network and improve the user experience.

  2. OIM says:

    I fully agree with Ed. And I would add that to keep client devices affordable the intelligence should still be put in the server.