No mention of MicroSIMs at the next big SIM conference

Apple's choice of a new format of SIM card for the iPad is one of its most cynical moves in a while.

While it is a standard (and apparently T-Mobile has mentioned it previously and it's indeed been around as a concept since 2003) it's conspicuous by its absence as a topic for discussion at the next big SIM conference in April.

It seems like it's being done to give Apple - and its operator partners - yet another control point. The easiest way to ensure an iPad-specific data tariff (and whatever deal has been done behind the scenes) is to have an iPad-specific SIM card. In theory the device might be "unlocked", but if you can only get microSIMs from specific operators on a handful of tariffs, you're still constrained.

I'm willing to wager that Apple and/or AT&T has made an interesting deal with microSIM suppliers - it wouldn't surprise me if they've bought up a sizeable fraction of the total manufacturing capacity for the next year or so. If it's been around since 2003 with no uptake, there's probably just a couple of dusty production machines sitting in the corner of a factory somewhere. Sure, I'm sure lots of companies will now be revamping their SIM portfolio and forecasts, but it should enable the early iPad operators to keep data pricing under strict control (and more importantly, sell new connections rather than let people re-cycle existing data SIMs) for 6-12 months I'd say.

Contempt for the customer, in my view.

iYawn

So is that it? A touchscreen tablet?

And $130 extra for a version with an embedded 3G module. So much for predictions that HSPA modems will be "as cheap as WiFi" within a couple of years.... And unless I've missed something, there isn't even a normal USB socket so you can use a regular cheap 3G dongle. I guess it would work OK with a MiFi-type modem though.

And what was Jobs thinking with a non-standard "microSIM" format? It's not like it's a small form-factor device that can't take a normal SIM card. So now you can forget about going on holiday and buying a cheap local SIM with data access in most countries in the world - I really can't see the thousands of small stores in India or Portugal or Mozambique or wherever stocking multiple SIM formats.

Sorry Apple, I'm unconvinced. It's vaguely useful & cute, but it doesn't seem to be a new product category to me, and the connectivity options look severely flawed.

(Oh, and before the rabid Apple fanboy contingent jumps on me for criticising their sacred purveyor of gizmos, I should mention that I would have finally got around to getting a 3GS iPhone today, had O2/Carphone not had such a ridiculous requirement to *mail in* for a number port. More on that saga another time. The iPhone makes sense to me, the iPad doesn't).

Customer loyalty, tolerance and contempt

It continually amazes me how much self-delusion I hear from operators and their vendors on the topic of "customer loyalty". Many seem to confuse inertia (ie people being too lazy to switch) with loyalty. Or they confuse services which they describe as "sticky" (ie which have inherent user lock-in) as having the potential for creating loyalty.

They should have a look in a dictionary: "A feeling or attitude of devoted attachment and affection"

This fits rather uneasily with the reality of customers being treated with contempt over things like ludicrous data roaming fees, or (if the hype is to be believed) future attempts at extortion by non-portable network-resident address books.

Roaming fees are a particularly good example: do you really, really expect your customers to respect you and "be loyal" ("feel affection") when they are so clearly and evidently ripped off, often by factors of 100 or 1000 over reasonable prices? The two concepts - contempt and loyalty - are mutually-incompatible, and yet somehow marketing and pricing executives seem happy to overlook the link between the two.

I know very few customers who are truly "loyal" - in the sense that the extol the virtues of a given operator to friends, much less who would self-describe as "devoted" in the same fashion as a footbal fan or an Apple acolyte. Most are, at best, "tolerant" of their provider.

Cross-platform apps are the only way forward

I've been writing about operator cross-network aps for a while, decoupling access from service. We need to see Vodafone 360 on AT&T handsets, and O2/Jajah VoIP on Hutchison 3 phones. It's the only way to build scale against the real Internet application players.

But thinking further, and especially in the light of yesterday's announcement by Nokia about free turn-by-turn navigation, there also needs to be cross-phone applications.

Google has a head start here, with its portfolio of apps available on most devices, not just Android. To compete, Nokia has to follow suit - if it's serious about apps, it cannot just confine them to its own hardware platforms, it needs to port them to Android, iPhone, RIM and so on. Maybe, ultimately even Apple's iron will should crumble, and it should export aspects of its flagship UI to other OS platforms (although I'm not betting on it).

Some of the Nokia apps are really good - the SMS client, for example. Some of the SonyEricsson and Samsung ones too. But if I don't want a Nokia or an S-E phone for any reason, why can't I still have that experience, as long as the hardware supports it?

Pre-MWC notes for analyst relations staff

OK, it's the time of the year when I get bombarded by emails and phone calls from a million people inviting me to briefings and similar events in Barcelona for 3GSM (I still hate the term MWC).

In order to maximise the value of both my time and yours, I thought I'd give some pointers:

- Priority goes to meetings with mobile operator execs, strategists and the like, particularly to talk about mobile broadband, new business models and device portfolio
- Next priority areas for me are with vendors of femtocells, smartphones & OS's, connection managers, mobile broadband devices & modems, LTE and HSPA+ radio gear, infrastructure enabling "offload", handset chipsets / platforms
- Of significant interest are handset apps providers that are doing "hard" stuff like VoIP, backhaul vendors, OSS/BSS suppliers, integration services & consultancies, although I'll probably be waiting until nearer the time to see what gaps in my schedule I have available
- I'm not interested at all in mobile TV, content, advertising, and a million small apps companies trying to jump on the social networking bandwagon because they access Facebook's API.
- If company is based in London, or regularly has execs passing through, it seems fairly pointless to go all the way to Barcelona for a meeting
- Generally I find that joint press & analyst events are not valuable
- If you pitch me with "story ideas" I won't be interested
- If your company/client refers to either LTE or WiMAX as "4G" I'm unlikely to take it seriously
- I almost certainly won't be attending anything that's more than 5 mins walk from the Fira, unless it finishes before 8.30am or starts after 6pm. So the Fira Palace Hotel is OK, but anything which needs a mission on the metro isn't.
- Invitations to interesting parties and dinners are always gratefully received...

Lastly, I have a couple of presentation or moderating gigs already scheduled in, on various topics. If you're interested in having a well-known analyst participate at your customer or press event, let me know and we can discuss suitable arrangements. Similar, if anyone is interested in meeting me to discuss Disruptive Analysis' consulting and advisory offerings, please get in touch.

(information AT disruptive-analysis DOT COM)

So what are the capacity hogs? And where are the solutions?

I've been having a pretty forthright debate over on the Forum Oxford discussion site, which ties in with a piece of analysis that Morgan Stanley did on mobile network capacity and broadband usage.

I haven't had a chance to go through the whole MS report with a fine-toothed comb, although I've had a glance at the section on mobile data growth.

My initial observation is that MS seems to be under-estimating short-term growth in mobile data consumption, and over-estimating network capacity. I'm also not convinced that the problem is always necessarily backhaul - the radio network is also creaking.

One thread of the ForOx debate was about "what's causing the congestion?" with lots of theories about video of different types, web browsing, downloads, uploads etc. Also asking about whether it's the uplink or downlink or signalling that's the problem.

I'm actually doing an awful lot of research into mechanisms for mitigating the impact of traffic growth on mobile networks. What are the solutions, and more interestingly, what are the business models around them. I've written a Use Case for the guys at Telco 2.0 on "Managed Offload", and there are various other things I'm pursuing here as well (contact me privately if you are interested in advisory work in this field).

But observation I'd make about this question of "what's soaking up our precious mobile capacity?" is that in many instances it's not apps that are causing the problems. It's concrete and brickwork. Moray Rumney from Agilent gave a great presentation last year which points out where the real mobile capacity increments will come from - WiFi and femtocells, primarily.

I'd also add in extra cost-savings from offload in the transport and core networks. I'm less convinced that traffic-management at the application layer, or things like compression of images/video have such important roles to play, although there will be exceptions.

Jajah + Telefonica

• Most 3GPP / UMTS operators will need to wait until at least 2011-12 before starting broad migration of circuit telephony to standardised VoIP. In the interim they will have to compete or partner with pre-standard VoIP players.
• Operators expecting to deploy LTE networks need to consider gaining prior experience of mobile VoIP. Simultaneously rolling out a new radio technology and a new voice architecture is a huge risk.
• There is scope for partnership between VoIPo3G innovators and incumbent operators (and other parties), especially on HSPA networks. Initial reticence will be countered by awareness of the threats of outright competition.

Those three bullet points came from the first-page summary of a report on VoIPo3G that Disruptive Analysis published in November 2007.

There has been considerable discussion on the web and various private forums about exactly why Telefonica (specifically its European arm, O2) acquired VoIP provider Jajah for $207m, a couple of weeks ago while I was on holiday. GigaOM has a good article on it here, and I've seen a few others as well.

The most obvious parallel I see is with BT's acquisition of Ribbit. In that case, BT wanted an entry-point to the developer community, and especially enterprise CEBP (communications-enabled business process) marketplace. This area is epitomised by companies like Salesforce.com, as well as a plethora of other firms helping firms manage customer and employee interactions via voice - essentially an evolution of CRM.

Telefonica/02, however, appears to be more focused on the consumer space, especially the possibility of hooking up its existing telephony user base (mostly mobile in Europe, but not exclusively) to web-based and apps-based social networks or other online applications. Jajah provides the voice back-end for Yahoo's IM-integrated VoIP service, and more generally has a presence in mashups and "white labelled" voice for various Internet players.

I can see various synergies here - some immediate, some longer term. But the bottom line has to be that O2 wants to learn more about voice without the "heavy lifting". The value of much of the new telephony concept isn't about "big iron" like IMS or even QoS. It's certainly not about vague 1980s-style waffle about operator-centralised "multimedia", as per the defunct MMtel standard. It's about the social value inherent in having voice as a platform or a web component.

Jajah helps Telefonica start to break away from the stifling "it will all be IMS some day" mantra, or at least leave it to the traditionalists slaving away in the infrastructure dungeon, trying to rescucitate the 3GPP zombie's corpse. That's not to say I expect O2 to just abandon that approach over-night (it's too ingrained into the telco DNA), but taking a first step is a wise move - it allows the company to see what else is out there.

More pragmatically, it also helps with a number of other opportunities:

- Early deployment VoIPo3G, perhaps starting with laptops connected via mobile broadband, or selected smartphones. This will probably run as a "second line" on phones, in addition to traditional circuit - a bit like having Skype or Truphone (or Jajah) on a device today, but controlled by the operator. I've been expecting operator VoIP to appear for some time in this guise.
- An easy off-the-shelf way of testing and playing around with VoIP on LTE for upcoming trials. If Jajah works sufficiently well on best-efforts or "groomed" data connections, it could save Telefonica a large amount of cash paying for unnecessary QoS over-engineering in future.
- Various options for Spanish language communities such as travellers and ex-pats moving between Latin America and Spain
- A way for Telefonica to move into new markets outside its current geographic footprint. In particular, this means that it's yet another operator doing the "unmentionable" and owning web-based services accessible from other operators' devices and access subscriptions. Along with Voda 360 and Orange ON, Jajah is now another example of under-the-floor providers playing over-the-top

What's the story with the Phonejack 'femtocell"?

I'm now trying to catch up with events and analysis over the past few weeks, as I've been on vacation.

One of the more bewildering things I'm trying to get to the bottom of is the supposed Magic Jack "femtocell" as written about in numerous places such as here, here and here.

In theory, it hooks up a GSM phone to a VoIP client embedded in a USB stick. Details on how it works are remarkably thin on the ground, although allegedly it uses the IMEI number as part of its authentication mechanism.

As yet, the whole thing seems to be based on press releases rather than demonstrations or any hard detail on the underlying architecture. If it works as billed, I can see numerous pitfalls or open questions:

- what frequency band(s) does it work with?
- does it need a separate SIM in the phone, and does Magic Jack supply these? Do you need to switch SIMs when you leave the house? Or is the company hoping to sign some sort of roaming deal with operators?
- how does it deal with SIM-locked GSM phones? Does it "spoof" particular operators' network characteristics, or have its own network ID?
- what are the legalities around its use of licenced spectrum, even if the power is supposedly low enough to avoid interference?
- how many end users can work out how to use the "manual network select" function on their phones... or can be bothered to switch back and forth when they're at home.
- how does it deal with phones that are configured to look for 3G first, then fall back to 2G?
- what's the control, authentication and security mechanism? Does it emulate an MSC and HLR somehow?
- it looks like the PC it's attached to is a fundamental part of the device. What happens when it's in standby or hibernate mode?

The Engadget article linked above has a comment from someone who claims to be the device's inventor. He says "As far as licensed spectrum is involved,who gave somebody the right to sell spectrum in my house?You own your own cellphone,you own your own magicjack device and you own the air in your house.The licensed carriers owns the right not to be interfered with.Our device does not interfere."

While that's pretty contentious, it fits with some of the rhetoric I hear about "open spectrum". It's also worth pointing out that the various iPod-connected FM transmitters were initially thought of as illegal in some places, yet regulators such as Ofcom allowed them if they operated at sufficiently low power.

Other posts try to work out what's going on - one suggests that it looks to the phone like a network it's roamed into, but it "pretends" that it's authenticated with your "home" network and just replies "OK" to the handset.

That said, I'm extremely doubtful that this will fly, technically, legally, commercially or in terms of user experience.

EDIT: I see that Andy Tiller of ip.access has a detailed take on it - and actually got a chance to try the device at CES.

Actually, now I think more about it, one possible killer app for this is in markets with deregulated or light-licenced guard bands. In particular, if others follow the Netherlands' lead and try to have licence-exempt bits of GSM spectrum, all sorts of things become possible. Especially if you can do some clever things with connection management apps on smartphones - perhaps running 2G voice (via the Magic Jack) in parallel with 3G or WiFi data.

Back from vacation - thoughts on mobile in Sri Lanka & the Gulf states

I got back from my extended break this morning - looking through my news emails, there's clearly been quite a lot going on while I've been away, so it'll take a while for me to process my thoughts on CES, Nexus One and so on.

Before I get back to "the business in hand", though, I want to jot down a few thoughts and observations on things mobile I've seen while on holiday. I was primarily in Sri Lanka, plus a couple of days each in Oman, Kuwait and Dubai on the way there and back.

Now, I generally like to separate work and down-time, to the extent I've had my voicemail saying that I'm not picking up messages, for the past three weeks. So I certainly haven't been doing report-grade research while I've been away. But at the same time, I've certainly been observing people, adverts, shops - so I have a few anecdotes about mobile and wireless that might be of interest to a wider audience. I should point out that I tend to travel around a lot, so I usually see a lot more than just a capital city and its high-end malls, or a plush beach hotel. In a lot of developing, you see a very different side to "mobile" on a local bus between two small towns, versus the lobby bar of the local Hilton.

First - Sri Lanka. I spent most of my time outside the capital Colombo, mostly in a variety of beach and hill towns. The most obvious thing I noticed was the differences between Sri Lanka and its neighbour India, across which I drove last April. There is much less conspicuous mobile advertising - few of the house walls painted in corporate colours so common in India. Where there was display advertising, or point-of-sale material outside various shops, it was 95% dominated by simple brand advertising for the various local mobile operators' prepay SIMs tariffs. As is common in many countries, you can buy credit for prepay anywhere from a florists' shop to a market stall, to one of the numerous dedicated communications/Internet facilities.

I bought a local SIM on the Dialog 3G network plus some credit to use a smartphone as a web browser. While coverage is OK and purchase was easy, getting online was definitely not. I needed to use a PC first to go online to get the settings for the right access point for my phone, and then find a number to SMS to a short code to purchase blocks of Internet time (1 hour etc). In fact, I asked two or three phone shops about which SIM was based to access the web on my phone, and got blank stares. The provisioning / access-control system seemed pretty clunky too - sometimes I got my allotted access time, sometimes I got "unlimited" access, sometimes it seemed to charge me adhoc amounts. It was cheap, whatever, though - nominally about £0.10 (20 rupees) for an hour's access, cheaper than the £0.01p per minute in the many Internet cafes.

That said, handset-based web access definitely seems an unusual exercise in most of the country, although I saw a fair number of shops selling Huawei HSPA dongles for mobile broadband, as well as (a few) outdoor adverts. Anecdotal evidence suggests that some of the upper echelons of the business community in Colombo have BlackBerries, but it's definitely wrong to assume a wide use of "the Mobile Web" among the general population.

A very telling anecdote came from a driver I hired for a day-trip. He laughed and shook his head when I asked if people were "accessing the Internet on a mobile phone". He has a $20 Nokia handset, but bought a $50 PC (an old Pentium 3) for web access. He pointed out that 'normal' people (well, men anyway, it's quite a segregated society) use Internet cafes, and some schools have PCs, but was completely unconvinced by the notion of massmarket mobile web usage. He said that it might appeal to the Colombo business folk - although amusingly he suggested that anyone who cared about mobile Internet use would probably be affluent enough to have a personal driver, so they could sit in the back seat with a laptop rather than struggle with a smartphone. Also, home access to the Internet has high levels of parental supervision, which would be difficult to apply to anonymous prepaid mobile access.

I saw absolutely no advertising that suggested that handset-based web access was seen as important, nor much point of sale material in a couple of stores I peeked into. I saw nobody on the buses, trains, streets, or local markets or restaurants doing anything identifiable as Internet access, although obviously SMS usage was rampant. Everywhere, however, were shops selling low-cost second-hand and refurbished PCs - Colombo even has a dedicated shopping mall just for PC and software/accessory shops.

I've spoken before about the oft-repeated myth of "the next billion" Internet users supposedly accessing the web on their handsets first, rather than PCs. I've often wondered where they might be hiding, outside of Japan and a couple of bits of India. Well, they're not living in Sri Lanka, that much seems probable. (Of course, if you live there and have more concrete observations, please let me know).

One thing that was interesting, though was the pervasiveness of WiFi. Many of the hotels and guest-houses I stayed in had free wireless attached to ADSL (mostly low-ish speeds of 1 or 2Mbit/s) , as did many of the other more tourist-oriented locations like restaurants. Coupled with the Internet cafes, the growing prevalence of tourists with netbooks may well be having a drip-feed effect on the local awareness that "The Internet" is primarily a PC-based phenomenon.

One last comment that springs to mind - looking in some of the phone shops in Colombo (handsets are normally bought unlocked and used with any SIM), I was conscious of continued Nokia dominance... but also a growing number of clones and outright fakes. The very-unsubtle "Nokla" brand with an L rather than an I seemed quite common, as did fake E71's branded "TV Mobile" which were clearly GSM sheep in 3G wolves' clothing.

The much-wealthier Gulf countries were different, although except in Dubai I still didn't get the impression that smartphone or mobile web usage was that common. Adverts for 3G dongles were pretty common though, as were PC-based broadband users in numerous cafes, with the Kuwaiti operators loudly trumpeting 21Mbit/s HSPA+ speeds. Oman had a high density of Internet cafes, especially in smaller cities outside Muscat, where I suspect that levels of affluence and home-broadband ownership are much lower. It was also notable that many Internet cafes combined games rooms with billiards/snooker tables - essentially becoming (again, mostly male) social hubs in a country where there's obviously no "bar" scene.

Ironically, like Sri Lanka, Dubai also had some less-usual handset brands. Stockists of the flashy/vulgar Vertu and Porsche phones were everywhere....

(And a note for the mobile broadband commentators - notthat many embedded-3G notebooks in the swish Dubai mall electronics shops, either).

Note: I'll probably remember some other anecdotes over coming days. I'll add them to this post as edits. And once again, this is just some observations made while on holiday - I'm sure there's a lot more rigorous and detailed research out there, but I always like to notice a few things that stand out.