Post-VW thought for the day: more software will detect tests and reviewers

A very quick post today. 

Everyone is watching the VW scandal evolve - the company appears to have programmed some of its cars' engine management systems to work out when they were being tested for emissions. And then put the vehicles in a special mode to "game" the tests, before switching back to "normal" (more-polluting) modes afterwards.

I suspect that's the tip of the iceberg. And that the iceberg is going to grow.

I'm not an AI or machine-learning specialist. But I bet one of the easiest things for a computer to "learn" is when it's being subjected to a test of some sort. Tests look different to normal use - formulaic and often repeated. Could be a legal or regulatory test, could be a product reviewer, could be an internal acceptance-test, could be a benchmarking exercise - and so on.

Once the system knows it's being tested, it can then react to try to optimise itself to give a different set of test results.

I once lightheartedly suggested (in 2009) that we'd see broadband networks learn to give maximum speed/QoS to speed-test services. They might also be programmed to "look" very neutral when they see something from voip.regulator.com, but react differently on other occasions. Devices or cloud services might recognise the names or IP addresses (or cookies) of prominent journalists or product-reviewers. Self-driving cars will probably be very good at spotting traffic cops.

I'd be willing to bet that a lot of this type of "gaming" of tests and reviews already happens. VW won't have been the first to think about it.

In other words, testing software will be like giving exams to humans. We'll need invigilators of some sort, to make sure they don't cheat. And testers and product reviewers are going to have to be a lot more subtle, and change/obfuscate their work.

That's going to be a non-trivial problem if we want comparable data. And I haven't even thought about how you invigilate a Turing test to ensure an AI isn't "playing dumb".

VoLTE & WiFi-Calling are just excuses for telcos to avoid real voice/video innovation

Five years ago, I started talking about the Future of Voice, and then subsequently started running workshops with Martin Geddes on the topic. At the time, I found it quite hard to find the right people at telcos/service providers to talk to. Few were aware of concepts such as embedded voice/video, new user-interaction models, developer platforms, or a “post-telephony” world where we all had many ways to communicate, choosing the best tool for a given job.

There was no “product manager” for telephony, and nobody with responsibility for communications services innovation. There was no "VP, Voice" - it was just assumed to be an inherent background task owned by everyone and nobody. There might have been folk in the enterprise unit looking at UC and conferencing, international dial people scared of Skype, and a couple of people in the labs wondering what to do about voice on LTE, but that was about it. Oh, and of course at a handful of operators, there was sometimes some hapless soul trying to push RCS, either internally or at GSMA.

Then, about 2-3 years ago, there was a shift. Various people with titles like “Head of Advanced Communications” started popping up, roughly as the tidal wave of smartphones, messenger and VoIP apps, developer platforms and so on started to take off. A few people had heard of WebRTC, some operators were tinkering with their own early “telco-OTT” comms apps (remember T-Mobile Bobsled?), developer tools were being pushed, and there were signs that some actual innovative thinking was taking place. (And a few hapless souls were still pushing RCS, of course).

But in recent months, that glimmer of positivity seems to have dimmed again somewhat. The bulk of telco announcements recently concerning “advanced” communications has been anything but "advanced". It’s just been breathless announcements about VoLTE or WiFi-Calling, as if they actually changed anything. (And, yes, a few hapless souls are still pushing RCS. Although rather fewer – quite a lot of them have finally escaped the Joyn event horizon). See this recent post of mine for an example.

Let’s be clear – VoLTE has four benefits:

• Offers a solution for 4G-only operators with no 2G/3G or MVNO deal for fallback
• Allows simultaneous voice & data on 4G, rather than forcing 3G fallback for data during calls
• Gives faster call setup time (nice, but the sort of minor feature upgrade that would have been quietly introduced in v6.3 for any other voice app)
• Might eventually help with spectrum refarming. This is equivalent to fixed operators being able to sell big old exchange offices in cities. It allows eventual asset sales / re-use. Eventually.

Beyond that, there’s no new revenue, no change to the basic vanilla 130yr-old format of “phone calls”, and ironically for a standard, very little working interoperability with other operators’ VoLTE. It’s an expensive “forced purchase” as the industry was too slow/complacent to come up with something better, and painted itself into a corner. It’s not going to stop people using other communications apps or services, it’s not going to halt revenue declines or reverse "peak telephony", and it’s still going to take years to transition the bulk of people from circuit. (And no, HD voice is not special – it’s been around on 3G for years, and is another minor feature upgrade nobody pays extra for).

WiFi-Calling is no better. It’s a slightly better implementation of a 10-year old idea, basically UMA v2.0. It gives better indoor coverage for some users in some areas. It covers for a lack of cell-sites or sub-1GHz frequency bands. In other words, it’s window-dressing, not something substantively different. There’s probably 4-500m+ people doing some sort of voice/video communications over WiFi anyway, using 3^rd-party apps. It can in no way be described as “advanced communications”.

Some (fortunately few) are talking about ViLTE – which is VoLTE’s ugly video-calling sister. It’s pointless. The last thing to do with video is to “call” someone like a phone-call, unexpectedly and interruptively. There isn’t even a legacy user-base to pretend to migrate, and it’s clearly not as functional / cool / integrated / well-designed as the 100 other video-chat apps and APIs available, even without the fact that WebRTC means that all apps can integrate video if they need it. I'll skip over RCS as I'm sure you've got the picture by now - but read this if you're uncertain.

And this is the problem. All of a sudden “advanced communications” means VoLTE and WiFi-calling, with a side-order of irrelevant video/RCS. That’s just a convenient excuse not to do any proper innovation. They both just deliver plain-old phone calls, but on different networks. Yes, it’s nice to have better indoor coverage, but covering up for existing deficiencies is hardly worth a press release. It’s like adding a bagel function to a bread-toaster* and claiming a major step forward in cooking technology. Only at least people still think bagels are cool.

In my view, VoLTE and WiFi-calling are “make-work”. They make telco engineering and core network groups look busy. They give an excuse to vendors to try and finally sell their IMS infrastructure – albeit in NFV-based versions at lower cost. The policy vendors get a look-in too, so they can finally prioritise something with network QoS. And there’s the nice comforting mythology of ViLTE and RCS on the horizon to continue the gravy-train. 

And it gives an equally comforting mythology of “level playing fields” to take to regulators. That's nonsense, too. (See here)

Meanwhile, genuine innovation in voice, video, messaging, contextual comms, APIs, developer platforms, enterprise communications, CEBP, WebRTC, cool mobile comms apps, social voice, personal broadcasting, telemedicine, IoT-integrated comms and 101 other areas is carrying on regardless. But on the Internet, or on mobile, or in enterprise cloud-based comms.

But telcos and vendors, with their nice warm VoLTE/WiFi-Calling comfort blankets, can delude themselves they’re doing something “advanced” because they’re spending money and doing “stuff”. But it’s simply an excuse for failing to make hard choices. It’s “going through the motions”.

CEOs and CFOs should call their bluff. If it's just "phone calls" they might as well outsource the voice infrastructure in entirety. And telecom regulators should ignore the protestations about so-called "OTTs", when telcos are doing nothing to try to compete or meet modern customers' communcations needs and purposes. It's the Internet and app providers that are employing a "design" mindset here and need protection, not vice-versa.

Now this is not true of all telcos, nor all SP business units / teams though. There’s still a lot of interest in doing cool stuff with WebRTC, a number of interesting mobile apps by telcos, some interest in contextual communications and developer APIs. Telefonica TokBox, Orange Libon, Telenor appear.in, WebRTC platforms from AT&T and NTT & SKT, Comcast's Xfinity Share, Swisscom iO and various others. For many of these it's still early days - but that's the type of trial-and-error, agile, customer-centric approach that's so desperately needed.

But usually, those initiatives are done by the telcos' more peripheral units – labs teams, enterprise arm, international opco's, TV/content business, standalone developer-platform units, internal MVNOs, so-called “digital services” groups and assorted other teams of free-thinkers unencumbered by legacy mindsets or GSMA/3PPP/ATIS/ETSI doctrine. Often, they have interal battles with the legacy fiefdoms that don't want to risk cannibalisation - or being made to look over-resourced and slow. Politics wins too often. There's also various MVNOs and smaller MNOs, from Truphone to Google Fi, that are trying to do something different as well.

Something similar is occurring in parts of the vendor space too. GenBand has its Kandy PaaS business which focuses on WebRTC for enterprise apps. Ericsson's Labs team is working on the OpenWebRTC mobile stack, and assorted non-telco uses of voice/video. Metaswitch is repurposing IMS as its cloud-based open-source platform Clearwater, encouraging tinkering and developer innovation. 

But plenty of other vendors keep recycling the tired old marketing lines on their ghost-written "content marketing" blogs or webinars about "How VoLTE and WiFi-calling & RCS will help you beat the OTTs". It's cynical clickbait, and either self-delusional or aimed at deluding their customers. Not one has any case-studies - or even a decent argument - about winning back users from WhatsApp, Snapchat, Talko, Wire, Periscope, Slack, Skype & Skype4Business or the 10001 other cool services.

This needs to change. Yes, VoLTE and WiFi-Calling have some value for some operators, mostly because they're forced into it. If they can reclaim spectrum, great. But they should NOT be excuses for inaction elsewhere. They do not redefine communications. They do not open up new revenue streams, or significantly help loyalty. They are, at best, strengthening the walls of the final core communications fortress, so telcos can defend 10% of their former territory against the invaders. Actually, the analogy is flawed - perhaps "liberators" is better, given the alternative are welcomed by users with open arms. The GSMA's so-called "Network 2020 Green Button Promise" is a pre-eminent example of this woefully narrow vision.

VoLTE and WiFi-calling should represent maybe 20% of operators' activity in future communications, not 80%. ViLTE & RCS should both be zero %. The bulk of effort should be on genuine innovation - or else acquisition / partnerships with those who can do it instead.

Yes, this post is confrontational and will no doubt put a few noses out of joint, including those at some of my own clients. But this is important - there's no value in rearranging the telephony deckchairs, when there's a vast iceberg of contextual communications, design-led apps and WebRTC hoving into view. Making a phone call on WiFi isn't going to help.

Apple Upgrade - will carrier activation be done with an eSIM or Apple SIM?

The iPhone 6s still has an ordinary SIM slot. Next year, I predict that the iPhone 7 will still have an ordinary SIM slot too. 

But Apple's new Upgrade installment plan that allows people to get a fresh new smartphone every year might change the SIM in the longer term. It's US-only to start with.

There is also a possibility of a removable Apple SIM as part of the Upgrade plan - a similar concept as the one in last year's iPad. That's more likely for next year rather than this year.... but at the time of writing this post (11th September 2015) Apple still hadn't put up the full details of how "carrier activation" would work on the web, so it's *possible* that there will be a surprise coming much sooner (maybe even tomorrow, when pre-orders start).

There is also a possibility that the iPhone 7, next year, will have an embedded "eSIM", although it would also need to have a proper SIM slot too, or at least two versions to choose from.

The Upgrade programme makes a lot of sense - most US carriers now have some form of installment plan for smartphones, rather than a bundled-in "subsidy". That's actually been part-driven as an effect of changing accounting rules, which restrict the bundling of products and services in reported revenues. See this article here from last year, and my own 2007 post where I mused about the economics of phone subsidies, here.

The Upgrade plan allows customers to get their iPhone finance plan from Apple, rather than through AT&T, Verizon, T-Mobile or Sprint. (For Apple, it's actually a lease underwritten by Capital One). The one-year cycle allows Apple to potentially increase revenues, while running this through the Apple stores removes the risk of last-minute switching to Android devices, with a nudge from a carrier salesperson who's been incentivised to pitch Samsungs.

But a couple of things are left unsaid at the moment. When Apple talks about setting up and "activating" your phone on the carrier of your choice, what does that actually mean? 

For some users, it will just mean taking the SIM out of your existing 5s or 6, and popping it into a shiny new 6s. But what if you want to switch to a new carrier - or perhaps a new plan? Will the Apple store keep a stock of SIMs from all the carriers and activate them on the spot, like an independent phone shop? Or.... will it perhaps use Apple's own SIM, with a remote-activation setup menu as on last year's iPad? [EDIT: it has been pointed out to me that Apple already stocks some SIM cards in its US stores. Question remains though, which plans? Including prepay or just normal multi-year contracts?]

[My take on the Apple SIM last year was posted here. As I argued at the time, it has not been a big deal. Other carriers have not signed up, beyond a provider called GigSky, which appears to work with major global carriers like Digicel and Vodafone on a roaming basis]

In theory, and with the cooperation of the carriers, the new 6s could work with an Apple SIM card as well. Rather than walking out of the Apple store and then going to a carrier store for a SIM separately, it would be easier to just remotely download and activate a "profile" on either a blank (Apple-branded) SIM, or in future, to an embedded SIM chip inside the body of the phone. Ideally, Apple would love to do away with the design compromises from cutting a slot in the device, reclaiming space and removing a clunky mechanical component.

However, various scenarios here would require some complex behind-the-scenes processing, like porting your existing mobile phone number, and also maybe dealing with contract termination fees. This is why I have my doubts that it will happen this year, and even next year will have some headaches. They're not insurmountable for Apple - but they probably are insurmountable for others, for anything other than IoT-type deals like Samsung's Gear S2 watch with eSIM from last week.

Either way, whether it's this year or next, the Upgrade plan gets customers used to the idea that you "activate a carrier" on an iPhone bought from Apple. And given that there is no subsidy or payment plan from the carrier, there is no justification for 2-year contract plans, either.

If you have an unlocked iPhone, you'll be much more amenable to getting a rolling and cancellable 1-month contract (already popular in the UK and elsewhere, but less-so in the US) or even a full pay-as-you-go prepay account. You might even choose to go for a data-only SIM, and "bring your own voice".

In that scenario, it actually doesn't really matter (for now) which SIM model Apple uses:

• Swap out your existing SIM & put it in the new iPhone, in the Apple store
• Buy a SIM from a carrier store & put it in the iPhone
• Buy a SIM sold in an Apple store & put it in the iPhone
• Get a removable Apple SIM supplied with the iPhone, activate a carrier if it's chosen to be on the menu, or else take it out & buy a separate SIM as per the options above
• Have an eSIM inside the phone and activate a carrier of your choice, if they're on the menu. (If not, then there's probably a version with a proper SIM, as some countries' operators won't all be eSIM-ready anyway).

Long, long term (maybe 2020 to coincide with 5G) we might get to the "promised land" (or dystopia, depending on your viewpoint) of fully virtual SIMs, but don't hold your breath.

The problem with the eSIM / downloadable Apple SIM type model has always been getting carriers to agree to be involved. I've been skeptical that the model had legs, because of this. But the installment / upgrade plan - and Apple's footprint of own-brand stores - seems to be a victory thought up by a clever game-theorist.

One of the carriers will likely agree to in-store activation on Apple SIM / eSIM - or at least, agree to having their SIMs stocked in Apple's retail outlet. It saves customers a second shopping visit. And then the other carriers may be forced to follow suit. Given that at least *some* people will be able to activate their 6s "on the spot" by simply swapping out their existing SIM, there's even greater incentive to use the Apple Store as a point of decision, if you're trying to capture people ready to churn.

Apple has essentially flipped the cellular sales model on its head - rather than a Verizon/AT&T salesperson having the power to convince a user to switch to a Samsung in a carrier store, it's now in a position to convince users to switch to a different carrier, in an Apple store.

The interesting line on Apple.com is this: "Because the iPhone Upgrade Program isn’t tied to a single carrier, you don’t need a multiyear service contract. If you don’t have any carrier commitments, you’re free to select a new carrier or stick with the one you have. A Specialist can answer questions and help you set your iPhone up the way you like." Note the absence of the word SIM, and the phrases "select a new carrier" and "set up your iPhone". That's rather significant - it implies the SIM is available in-store, in some fashion, for the carrier to be "selected".

So in many ways, the actual SIM mechanism is irrelevant here - it's the retail footprint that matters. Lots of carriers are worried about the eSIM / Apple SIM meaning they "lose ownership of the customer", but the truth is more prosaic: it's the physical store that's the point of control / decision, because it plays to the human psychological need for instant gratification. Even online purchases are clunkier - unless you have same-day delivery, there's an in-built lag for activation anyway. Of course, it's also important that other device vendors don't have a similar retail presence.

Now obviously, the Upgrade plan isn't actually needed here. Nothing stops people from walking into an Apple Store and buying an unlocked phone at full retail price & getting a SIM card however they want anyway. But in the US at least, that's still very much a "minority sport", because of the price tag involved. It's just not how people buy phones, when they're accustomed to an apparently "free" or cheap handset. The monthly plan - and upgrade cycle - might change that, as it alters perception. It's also a clever lead in to some form of programmable SIM card, when it's worked out the various kinks and practicalities.

I suspect that Apple isn't 100% sure how customers will take to this. And it's probably ironing out various kinks and complexities with any sort of remote activation. It may want to wait until all the carriers have back-end systems capable of handling it, rather than risking relationships by jumping the gun with just one or two. Again, the game-theorists are probably trying to work out how to avoid one or more carrier stopping selling iPhones entirely in their own stores, which would have definite negative impact, in the short term at least.

My view on programmable SIMs / eSIMs is that business model is pretty much unworkable, except, perhaps, for Apple. Let's see what happens either next week, or next year.



I've been doing a lot of work thinking about SIMs, eSIM, programmable SIMs, multi-IMSI and so forth recently. Over recent months, I've done a variety of private consulting projects and presentations on my thoughts on SIMs. I've been looking at the announcements, and also considering what the commercial, technical and regulatory implications of various evolution paths might be. If you're interested in a workshop on this, please get in touch via information AT disruptive-analysis DOT com