I spent last week at the IIR NextGen Service Platforms conference in Munich, featuring a combination of material on Telecom APIs, WebRTC and "Legacy Network Evolution". Much of the emphasis was on service innovation for telcos. As well as myself, both Alan Quayle & Martin Geddes were attending and presenting.
By the end of the day, all three of us were muttering under our breaths - and being considerably more vocal on Twitter.
The main problem that irked us:
It isn't just legacy networks that need to evolve in telcos, it's legacy thinking.
In short, much of the day was taken up by presentations and discussions that combined myopia and lame excuses.
"We can't do anything with WebRTC unless we ask the regulators first"
"Our brand would suffer if we launched a service that didn't work perfectly"
"IMS means the end of silos [for communications]"
"French consumers are tired of downloading apps. They'll use embedded RCS"
"IMS will allow the creation of an ecosystem of OTT players"
"Consumers demand QoS"
"People would use more data if it was free - or if someone else paid for it"
"It brings universal reachability"
"3G was about Skype & Whatsapp. 4G will go beyond those silos, to Joyn"
"Everything can be linked to the phone number"
"We can do all the API & developer relationship stuff ourselves in-house"
"IMS is central to WebRTC"
"VoLTE will increase production costs for telephony, but we're making a bet on IMS as a platfom for other stuff"
"WebRTC is just another access"
There remains a worrying disconnect between market realities and the wishful-thinking, blinkered views still endemic in the telecoms world, especially, it seems, among those with a "telecoms academic" background.
- A lack of focus on real customer requirements & behaviour. There remains no understanding of why people like fragmented applications or why they are often ambivalent to QoS.
- A failure to recognise that "the web has won" and that using the web/cloud/apps domain - and business/development models - is the only way forward for telcos.
- A rigid belief in the primacy of network QoS, despite (a) not understanding the limits of network maths/performance/wireless/coordination, (b) not understanding that QoS value is very use-case dependent, (c) not understanding that QoE is driven by many other variables than the network
- Specifically, a rather charming-but-scary belief that users will choose a less-attractive app that always works well, rather than a better/cooler/cheaper app that only works well most of the time
- A belief that usage of services will gain wide adoption merely because of interoperability (usually with a reference to SMS ancient history)
- A complete refusal to engage with areas like behavioural psychology, social dynamics, user interaction models, purpose/Intenet and, above all, design
- An inability to understand concepts such as "intention" and purpose - ie that increasingly the value of communications is why you communicate & what you're hoping to achieve, not the actual transport of bits that ensues
- Ignoring the use of non-carrier networks in the enterprise, 3rd-party unmanaged WiFi, home etc, even though statistically they are becoming much more common, especially for video-type communications.
- Overlooking the realities of the device marketplace - eg convincing Apple to implement anything it disdains, or equally trying to convince manufacturers of unlocked "vanilla" Android devices to implement software, which adds cost/time, yet appears unappreciated by users at point-of-purchase.
- Taking a very traditional telco view of things like "reachability", without considering that people don't always want to be "reached".
- Continued repetition of stupid cliches like "OTT" and "dumb pipe". These are usually a good indicator of cluelessness, unless prefaced with "so-called" and eyeball-rolling. I'm tempted to say that their utterance should be considered gross incompetence and merit instant dismissal.
Neither Alan nor Martin nor myself advocate that telcos turn into "pipes" - but equally, they have to identify the future sources of value in communications services if they are to remain relevant. This is about context, purpose, intention and "softer" things like that - delivered via APIs or rev-share deals, or on a freemium basis. "Interoperability" is not a consumer value in itself. It might help for some services, and it might hinder others. We all presented about things to do with voice or messaging that add value to the basic "telephony" proposition. But instead, we heard about VoLTE & RCS (!) as some sort of magical saviours that will help telcos "beat the OTTs" without any explanation of why that should be the case, and what real-world human use-cases will be impacted.
One speaker admitted that VoLTE (with SR-VCC) will increase the per-minute "production cost" of basic telephony. It's a brave CFO that takes the bet he offered - that IMS as a platform would generate revenues from new as-yet-unimagined services - in spite of the almost-guaranteed decline in telephony value over the next few years.
As for WebRTC, quite a lot of speakers seemed to think its main role will be extending out VoLTE or RCS as a sort-of cheaper/easier softphone alternative. Fair enough, it might be useful for running a VoLTE extension on a WiFi tablet. (RCS - well, let's just say I had a new zombie slide, so you can guess my views. My intro presentation here & WebRTC workshop here). On the positive side, it seems that a lot of telcos are at least trialling or prototyping WebRTC, although most are probably too tied up with getting basic VoLTE deployed and working to worry about a WebRTC Phase 2 extension to it.
In my view, it normally takes the telecoms industry 4 years to spot a good idea, 4 years to implement it, and another 4 years to realise it was too late. For WebRTC, those numbers seem to have been cut to 2 years, which ranks as an improvement of sorts.
I'm off to Atlanta for the big WebRTC Expo & Conference next week, where I'll be moderating a number of the service provider panels. I suspect we'll get to rehash some of the same arguments about IMS again, although hopefully the fervent & innovative web/IT-centric atmosphere will permeate a few telco-academia brainwashed skulls and osmose into the legacy mindsets about service creation....
Great stuff Dean.
ReplyDeleteThe web has won. The world has turned to the cloud. For commodity applications and services with declining ARPU services like voice, the choice is clear: don't rebuild that network, cloud source instead.
MNOs and telcos no longer need to run a voice network to be a voice provider. Their core focus should be on investing in broadband and mobility networks and applications not rebuilding a network that delivers declining ARPU. Voice is still important, but it's a check-list feature. Very soon it won't drive standalone subscription revenue.
Nice sharing. This is going to be a huge problem especially in emerging economies where TELCO's are 'fed' with knowledge from matured markets such as 'reachability, qos differentiation for a fee, bw on demand'. The context can play a key role but the telco guys need time to understand and start exploring. I guess virtualization and the 'ITficiation' or Network will accelerate this.
ReplyDeleteThis comment has been removed by the author.
ReplyDeletePretty disheartening?
ReplyDeleteNo - I see this as amplification of the opportunities! If the old dinosaur operators are unable fathom the magnitude and nature of the telepocalyptic change in the World, then this makes the potential rewards for the smaller, smarter and more agile operators that DO all the bigger!
Couldn't agree more. Telco mentality still permeates our organisation/s. And this stiffles thinking and innovation. Seems like we are not listening to our customers (or are only playing to the traditional customers we know and like) And we certainly are not forward thinking. Its actually really easy. Just listen and watch what our kids are doing and transfer that relevance to the way our business may look in 5-10 years time.
ReplyDeleteDespite the fact that we have top level mangagement saying we are an IT services company and bringing in executive management from IT centric organisations. Telco type people are the senior/middle managers running the show.
What i would like to say is that underneath it all are some people who agree totally with what you are saying.
My entire job is to convince our company to change mentality. Its an uphill struggle but sometimes things get through. WebRTC is a case in point. Easily put together and demonstrated and value could be seen immediately in investing some effort in it. Unfortunately, i still heard exec level people saying "how do we make revenue out of this"
Bah humbug. IMS is here to stay, like it or not (and there is plenty not to like) The fact is that there is no realistic technology alternative for operators that encompasses the mess of regulatory, interoperability, legacy support and yes, QoS requirements (emergency services with WebRTC?). OTT services are not subject to anything like the same level of regulation and of course do not have a century of baggage to carry around - the reason you don't see technology alternatives cropping up for these operators is simply that there are no other credible offerings.
ReplyDeleteWebRTC is great and I expect to see very large-scale support for it but I also expect to continue to see large IMS deployments as networks gradually continue shifting to All-IP.
John - IMS is like a rocket that's run out of fuel. It's still going up, but is feeling the inexorable pull of gravity as it reaches zenith.
ReplyDeleteThere are only two options:
- It's fast enough to enter a low orbit. IMS provides basic VoIP telephony and not much else - maybe a little bit of conferencing & hosted PBX. This is what's happened in (some) fixed networks & may well occur in mobile if the price of IMS infrastructure & VoLTE capex & opex falls far enough. Like fixed VoIP vs. Class 5 switches, expect mobile VoIP to be a 20-30 year slow transition
- Alternatively, the rocket falls back to earth. Its mass is too much for its limited propellant. Weighed down by ridiculous baggage from vendors, 3GPP & GSMA, it fails to reach orbit and breaks up on re-entry. Telcos are hobbled by expensive stuff that barely works, in a declining market (VoLTE), useless stuff that costs time and money and generates zero revenue (RCS), and vendor licencing that prohibits innovation (no freemium services viability).
There is no third way. No IMS trip to the moon. It's either dull mediocrity, or crash'n'burn.
The smart vendors are going to be the ones chasing option A above - cloud/open-source bits of IMS, cheap NFV-based solutions, WebRTC extensions etc.
The smart operators are going to be the ones that:
- stick with CSFB (or GSM900) as long as possible
- outsource basic telephony as it's an expensive "hygiene factor"
- tell customers to BYOvoice & just stick to valuable stuff like content/SaaS etc
- eke out some marginal advantage vs. local rivals with WebRTC enhancements, perhaps integrated with IMS but more likely standalone
- Lobby regulators for less-restrictive rules on universal service, 911 etc.
On that last point, apparently the FCC has no definition of "voice". I'd argue that it probably is meant to mean "voice telephony". Other forms of voice comms should probably be less-regulated.
John - IMS is like a rocket that's run out of fuel. It's still going up, but is feeling the inexorable pull of gravity as it reaches zenith.
ReplyDeleteThere are only two options:
- It's fast enough to enter a low orbit. IMS provides basic VoIP telephony and not much else - maybe a little bit of conferencing & hosted PBX. This is what's happened in (some) fixed networks & may well occur in mobile if the price of IMS infrastructure & VoLTE capex & opex falls far enough. Like fixed VoIP vs. Class 5 switches, expect mobile VoIP to be a 20-30 year slow transition
- Alternatively, the rocket falls back to earth. Its mass is too much for its limited propellant. Weighed down by ridiculous baggage from vendors, 3GPP & GSMA, it fails to reach orbit and breaks up on re-entry. Telcos are hobbled by expensive stuff that barely works, in a declining market (VoLTE), useless stuff that costs time and money and generates zero revenue (RCS), and vendor licencing that prohibits innovation (no freemium services viability).
There is no third way. No IMS trip to the moon. It's either dull mediocrity, or crash'n'burn.
The smart vendors are going to be the ones chasing option A above - cloud/open-source bits of IMS, cheap NFV-based solutions, WebRTC extensions etc.
The smart operators are going to be the ones that:
- stick with CSFB (or GSM900) as long as possible
- outsource basic telephony as it's an expensive "hygiene factor"
- tell customers to BYOvoice & just stick to valuable stuff like content/SaaS etc
- eke out some marginal advantage vs. local rivals with WebRTC enhancements, perhaps integrated with IMS but more likely standalone
- Lobby regulators for less-restrictive rules on universal service, 911 etc.
On that last point, apparently the FCC has no definition of "voice". I'd argue that it probably is meant to mean "voice telephony". Other forms of voice comms should probably be less-regulated.
Dean, I think your metaphor is wrong - IMS is not a rocket and never was (despite the best efforts of everyone involved in standardisation). You, along with a lot of other analysts seized on this interpretation of IMS as a radical overhaul when in reality it was always a case of enabling a (glacially) slow transition to IP which suits incumbents quite well. I suppose there may have been a few IMS zealots at the start who had some kind of grand vision (I remember talk of IMS IPTV for example) but now it has settled into what it should have been from the start, a standard VoIP architecture with extensions for messaging, presence and video with support for all of the rubbish that operators need.
ReplyDeleteIf I were to look for a suitable metaphor for IMS I think perhaps something like the combine harvester might be appropriate - nothing too radical, extending the use of existing technology and increasing the yield a bit on the way.