What started life as a crusade against the pesky Internet (or Over-the-Top) communications services seems to be heading, slowly but surely, for over-the-topness itself. Will Joyn become just one more free, more or less over-the-top service, initiated and controlled by a download app and available for use via any telco? Recent evidence suggests that if it succeeds, that's what it will look like.

Does it matter? Maybe not, but this development is a far cry from the initial conception of RCS\Joyn as OTT-killer and restorer of visible 'telco-grade' communications services,

What's more, vendors and others are starting whisper "IMS-less networks". If you take away Joyn as an application which relies on IMS (that is supposed to be its foundation) and delay the introduction of VoLTE (4G voice over IP via IMS) then the business case for telcos' huge investment in IMS network infrastructure might start to look shaky.

So instead of taking the giant leap into penury in these difficult times, at least some telcos are likely to look instead at avoiding or at least 'virtualising' the necessary underpinnings to support Joyn - perhaps even replacing IMS-supporting hardware with chunks of software on standard, high-powered servers in the cloud (network function virtualisation - to give it this year's buzz-name). If some telcos also decide to rely on so called 'circut switched fallback' by using the 2 and 3G network to make calls and just leaving LTE for data for an extended period (perhaps until virtualised IMS becomes more viable) then it doesn't look good for IMS investment in the short term.

Let's step back.

Joyn was developed as a branded services-set making use of the GSMA's RCSe (Rich Communications Services - enhanced) specifications. It was to be the basis of the mobile telcos' next generation, interoperable and extensible communications services - SMS, if you like, with next generation bells on.

Joyn services were to be adopted by users in preference to OTT services like WhatsApp because of their interoperability (between carriers), their extra resilience in use (they wouldn't fall over as telcos allege OTT services do), and because they would come embedded on all phones (not just smartphones). They would 'just work' as an early slogan had it.

Revenue generation plans were always a little hazier. It was thought RCS/Joyn services would probably be freemium, but certainly the idea was that in the long run they were there to give visibility to the telco AND to monetise the network and IMS infrastructure on which they were to rely.

Despite much criticism and a long, long run-up, Joyn services at last appear to be getting traction. But far from being what was originally inscribed on their tin, they look increasingly like surrendering everything just to get into the market. To beat the OTTs Joyn-weilding telcos have had, essentially, to become OTT-like.

For instance: last month Korean telco Korea’s SK Telecom claimed to have signed up one million customers for its RCSe Joyn.T service, but in fact Joyn.T is being offered free (for life) for any SK customer on a flat-rate plan. Sounds very OTT/freemium to me.

SK has also made ‘messages’ data charge exempt and the messages themselves free to send. Nothing wrong with this, but it's not really an example of the telco services model triumphing - why wouldn’t SK’s customers welcome a free (in fact almost subsidised) Joyn.T as one more interesting free messaging option for their smartphones? What is there to lose?

In Europe too, there is now at least some Joyn adoption. Spain is big on Joyn and Germany is catching up.

Germany's DT had earlier 'apparently' indicated that it was finding it more complex than expected to integrate Joyn into its network architecture and that there was therefore an 'indefinite' delay. Apparently this was a case of lost in translation as in English usage, 'indefinite' has come to mean 'open-ended' or even 'permanent'. Its use in this context indicated that Joyn had been "permanently delayed", whereas the German translator wanted a word which meant "uncertain of the exact date"... you get the gist.

DT has now launched its service in Germany with Vodafone so far the only other Joyner in that market.

Meanwhile in the US, Metro PCS is showing the way to what might become the IMS-less approach, and vendors are coming up with solutions to enable it.

Acision, for instance, is using its Ascision Cloud environment to enable its telco customers to launch "rich services while maximising their revenue-generating potential" from the cloud. Acision says it's getting operators to launch Rich Messaging fast and with limited up-front cost and - crucially - without the need to deploy an IMS Core. It's been offering RCS as a key service on its Acision’s consolidated messaging platform, Acision Broadband Messaging Service Centre (BMSC), since last year.

The US mobile telco MetroPCS has gone one step further. It's in cahoots with cloud-based video, voice and chat services, Jibe, to use its cloud-based version of the GSMA's Joyn under RCS 5.0. But here's the killer - MetroPCS will completely detach Joyn from its own network by making the smartphone app which initiates it (on the user's phone) available to any user on any North American network! In other words, you get to joyn (sorry) MetroPCS's services no matter what network (or device - it's also available across WiFi on tablets and smartphones) you're on.

So Joyn, in this version, has come full circle. It is, for all intents and purposes, another free(ish), over-the-top application, which is not tied to 'the network'

Given the likely IMS liberation of Joyn it's little wonder then that some observers (who often like to remain anonymous) make a powerful case that goes like this: why would you invest huge amounts of money to gold plate a network (with IMS) so it may support the very part of your market (carrier voice) which is no longer growing in volume or revenues and will even decline over the coming years?

Most/all of the big network operators are adding LTE frequencies to networks which are happily supporting 2/3G switched voice networks. Since the voice network volumes are not growing, one strategy may be to stick with circuit switched fallback to the older network when voice calls are to be made.

And since Joyn looks like it's just as happy (if not happier) going over the top as an app-initiated service (not embedded on the handset and trapped on top of IMS) there doesn't seem to be a need for IMS there either.

IMS network anyone?