IMS and IP Communications, Two Different World Views?

Telcos are faced with a tough choice by some of their vendors, “IMS, what choice do you have?

If telcos’ services do not work reliably they suffer through public humiliation.  Remember when AT&T had an exclusive deal on iPhone in the US?  Jon Stewart from the Daily Show said about owning an iPhone in NYC during that time, “we sacrificed being able to make a phone call…

To be clear, IMS did not solve AT&T’s network coverage problems in NYC; building out a decent radio network solved that.  But it does highlight that Telcos cannot simply offer ONLY the best effort experience of say Skype, where voice gets garbled or the line goes blank and then the voice comes back again but a couple of sentences have been lost.  BUT for some customers and some conversations this is clearly acceptable, given the volume of traffic Skype represents internationally, over one third of voice traffic and near 100% of consumer video telephony traffic.

Also there are lots of scenarios where a call starts on 4G and then needs to drop to 3G. Unless a telco makes the simplifying assumption that 4G works everywhere, which as a Verizon customer is not yet the case on its network.   This complex network scenario IMS solves so communications just work.

The fundamental issue is: do telcos build out a platform that supports one or more than one quality of service / communications model?  Given the risk of public humiliation some vendors are betting on telcos making the simplifying assumption of one model with full-IMS across all calls.  But does this make economic sense?  It ignores that customers are prepared to accept no guarantee of QoS in many situations and trade experience for price.

As Orange Libon has clearly demonstrated, the fall in phone call minutes is because of substitution as Telcos have simply not offered competitive alternatives as customers’ communications behaviors have changed.

Let’s review the differences in assumptions between full-IMS and the many IP communication platforms.  Yes, IMS is IP Communications and a telco could implement a light-weight version of IMS that follows most of the IP Communications assumptions.  This is not a clear cut division, I use the terms to indicate two clusters of assumptions.

IMS and IPBusiness people will pay for some / all of their communications to work with more certainty than best effort, e.g. calls to customers, while most people and most calls do not require a rock-solid guaranteed QoS, rather good enough QoS.  Customers have moved on and learned to accept the IP Communications assumptions.  Here is a refrain I hear every month or so:

  • Do I need to deploy an IMS? With RCS I can just deploy a client and app server, and internetwork with other operators through the app server implementing the RCS NNI.  And why not do the same with VoLTE, and even make it the same client.  Hence save all the politics, delays, expense and complexity.  Some examples of politics and expense include:
    • The IMS “owner” is quite restrictive with services not under their management. Their focus is PSTN replacement, and they are very conservative about trailing any other service.
    • An IMS platform has about 8 vendors involved and the per subscriber per service charging structure simply stops the telco moving beyond the current service as the business case keeps getting rejected.
  • Roaming is an issue.  When customers roam on other networks or other operators’ customers roam onto my network how can I support VoLTE, RCS, etc.?  But if I just enable data roaming and have all communication services through the client, then the roaming issue goes away for my customers, and I’m not going to invest in IMS just for other operators’ customers.
    • I’ll lose out on roaming voice revenues, but over half my customers now bypass roaming voice while roaming anyway.

IMS is focused on the legacy communications model with an expensive and complex platform.  That’s what telcos keep telling me when they look at the invoices for IMS, which is encouraging operators to seek out simpler IP communications solutions.  But those IP Communication solutions have risks. The biggest being public humiliation.

Can IMS adapt to the new market reality of IP Communications?  Take the choice of AMR-WB (Adaptive Multi-Rate WideBand) in IMS, when Opus is a much better codec from an implementation perspective as it easily runs in a virtualized environment, while AMR-WB generally requires DSPs.  The lack of progress in IMS on Opus clearly demonstrates the vested interests of the standards process.  I raised Opus at the IMS World Forum last year to a mute response.

Should operators simply ‘give up’ on IMS and hope IP Communications are good enough for all their customers, clearly no if they want to retain business customers and avoid public humiliation.  BUT operators do need to adopt a model that spans both sets of assumptions as that is what customers want, else loose more and more customers to other service providers.

A telco has a delicate balance between full-IMS, IP Communications and WebRTC.  The move to virtualization will help to operationally find a balance across all three, which will change over time.  This will be one of the themes we’ll be discussing at the IMS World Forum pre-conference workshop, “A New Wave of Communication Service Innovation: IMS and WebRTC” on the afternoon of Monday 28th April.  Note its the day before the main conference – so no turning up on Tuesday and asking about the workshop!   The MOST innovative vendors Metaswitch, OpenCloud, Solaimes, and Tropo will explain how they help telcos manage the real-world situation they find themselves in where they need to take more control over their future communication services given the slowness and vested interests of standards.

IMS World Forum Pre-Conference Workshop: A New Wave of Communication Service Innovation: IMS and WebRTC

IMS may not be everyone’s ‘favorite cup of tea’, but it’s there in most networks.  Put simply, for any operator migrating their services to IP, and in particular voice, what choice do you have?  But rather than complain about architectural correctness, it’s time to get over the silly academic architectural wars and focus on services and customers. In the limit that is what will define business success, not what’s hidden away in the network, but the services customers experience every day.

IMS and WebRTC: What does it mean and where is it going? – Alan Quayle
This section provides an overview of WebRTC’s current status.  The current status of IMS deployments based on a global survey Alan runs every 2 years.  And reviews how the latest developments in IMS, IP Communications, and WebRTC create a range of unique services and capabilities.  We’ll also quickly review the importance of taking an ecosystem based approach in developing new services and businesses.

Bringing IMS into the 21st Century with Open Source Software (Project Clearwater overview and demo)Metaswitch
Traditional IMS implementations based on proprietary hardware and 1+1 fault tolerance are costly, inflexible and lacking in scalability.  Project Clearwater is IMS re-imagined for the cloud.  Borrowing design patterns from well-known Web-scale services, Clearwater employs stateless SIP processing elements in conjunction with proven Web-scale storage repositories to deliver massive and dynamic scalability, active-active geo-redundancy, and extraordinarily low operating costs.  Clearwater is perfectly aligned with the objectives of Network Functions Virtualization, and enables network operators to compete far more effectively with over-the-top voice, video and messaging services.

Bridging the Network DivideOpenCloud review on the underlying plumbing necessary for these services to just work
For most operators IMS will be deployed alongside the legacy network which holds decades of investment in consumer and enterprise services. This section looks at adaptive signaling solutions to enable operators to take legacy services forward into IMS.  We also look at harmonized legacy-IMS service layers to reduce the barriers of service innovation.

Rich Communications and WebRTC – Juan Mateu, Solaiemes
WebRTC is the perfect tool to evolve the telco communications. WebRTC makes telco comms (Voice, RCS, UC) available ubiquitously and also allows connect number-based telephony with OTT identities.   Demos will include: Voice & RCS WebRTC clients and WebRTC Telco Service Portal created with WebRTC-Telco GW SDK.

Apps in Your Calls – where IMS differentiates! – Tropo
Tropo Ameche helps operators to differentiate from other service providers by using their core IMS capabilities to build a new class of apps in your calls.  Ameche is the first ever PaaS (Platform as a Service) dedicated to communication networks. The Ameche Platform and APIs are subscriber-oriented, allowing developers to weave apps and services directly into the mobility and real-time communication experience. Ameche enhances communications by enabling developers to mash-up telco capabilities with popular cloud services such as SalesForce, LinkedIn, and Google Apps. Ameche also extends mobility by leveraging RCS (Rich Communication Suite) and WebRTC (Web Real Time Communications) to embed communications and create a new form of contextual communication experience.

Wrap-up discussion and Q&A on the new wave of innovation
Fun and frank discussion by the innovators on what IMS, IP Communications, and WebRTC mean for the services telcos deliver to their customers.

Addendum.  I’ve also included below Schumann Sebastian’s presentation from last year’s IMS World Forum which explores this IMS / IP Comms duality from a technology perspective.