Wednesday, December 21, 2005
If I get caught in an Internet cafe during any tropical rainstorms, I may post some observations from remote bits of southern Africa, but otherwise I'll be back in January.
And I might be a mobile analyst, but I take great pleasure in leaving my phone at home when I go in holiday... change the voicemail, put it in a drawer, and head to Heathrow. Connectivity courtesy of Yahoo! Mail is more than enough....
Lastly, a quick shout out to one of the more eclectic "communities" enabled by technology for organising an awesomely bizarre outing on Saturday. You may be unsurprised to read that Disruptive Santa is the one being contrarian and wearing black....
Wednesday, December 14, 2005
During the conference Q&A I tackled SVP Mike Volpi on the topic, and was given a remarkably "scared and blinkered carrier"-centric answer which can roughly be summarised as "broadband is not a human right" and "carriers have to protect their own services on their own pipes, and not have them risked by other people who don't pay them".
Now, I've often seen equipment vendors shrug at their customers' sillier ideas and imply "well, it's their money. If they really want to waste it on something we think won't work, they might as well waste it with us, rather than our competition". However, there wasn't much public eyebrow-raising equivocation of this type. They seem to be serious.
Now, from my perspective there are indeed a couple of reasonable angles to this. Carrier IP networks that are strangled by huge volumes of (usually illegal) P2P traffic are an issue, especially in instances where this could actually break the network. Cellular networks may be especially fragile (imagine a Symbian-born virus which spread using Bittorrent-over-3G) and so I'll buy the notion of packet inspection as a form of pressure valve. Arguably, content filtering should be done in the network too, although personally I think it should be device/terminal-based wherever possible.
But the notion that Internet "service providers" (actually, I see them more as "application providers") like Google or Skype could be forced to pay interconnect fees to carriers (or do co-marketing revenue-share type agreements, as Cisco's Volpi also suggested) is palpable nonsense. The notion that a carrier could explicitly or implicitly degrade Skype of Yahoo!, on the basis that it competes with some of their own services, is horrific to most Internet users, and is likely to stifle the fundamental innovative nature of the Internet.
I struggle to think of a single "cool" and "useful" application or service on the web that was originally developed by a large carrier. They've all come from 2 guys in a dorm at Stanford or 2 ex-Apple employees. Can you imagine the inventors of the next Skype or Hotmail having to negotiate with 327 different carriers and ISPs around the world to make sure their service works properly?
As I said, if IP interconnect is enacted the way the carriers are threatening (largely in the US), it will kill Internet innovation.
Luckily, it won't work. The whole stupid project will fail at several levels - and possibly take the more laudable aspects like P2P traffic-moderating down with it, as it sinks.
The main points of failure:
- All this is scuppered by putting the traffic into a VPN tunnel, originating on a PC or a smartphone. You try filtering something that's encrypted, application-by-application, or URL by URL. Google's supposed WiFi secure access client seems like a poorly-disguised beta for a future version of this.
- In the future, we might build apps out of components more. I reckon XML/.NET based apps and services will be nigh-on impossible to police with deep packet inspection. The same component going across the network could be part of 20 different applications.
- No carriers have a position of "Network Policy Manager", and the complexity of such as putative role in liaising with different departments (network, wholesale, marketing, regulation, legal, HR, large accounts etc etc) makes it unlikely that one will evolve any time fast
- Many apps will just evolve to evade IP interconnect. Someone will make MSN appear to the network as MMS, or Skype will spoof SAP.
- Many operators trying heavy-handed approaches to IP Interconnect will screw up uninintentionally. I'd love to be the fly on the wall when the CIO of a major investment bank rings his carrier account manager and says "I've spent $5m designing a class-leading realtime P2P application to distribute equity derivative pricing information between my traders' desktops and mobile devices. Your stupid network appears to have started blocking it & we've lost $100m through delayed trading. I'm taking my 10,000 users elsewhere, and you'll be hearing from my lawyers and the SEC in the next 30 minutes"
To close off - an open request to the software community. Can you build me a desktop application that will work out (proveably) if my broadband carrier is degrading any particular services? I'll pay you for it. And I'll use it to churn to a carrier that makes a virtue of letting me use my pipes the way I want.
Sunday, December 11, 2005
Microsoft vs. Symbian vs. Palm vs. Linux vs. SavaJe vs. whatever.
I've lost count of the number of presentations I've seen which have basically said "If all phones were based on our OS, then you, the operator, could do all sorts of cool new apps / save costs / increase ARPU / solve world hunger".
Very few of those Powerpoint decks, however, have gone on to point out how the real world works - that there will, clearly, be a mix of OS's, at which point some or all of their "benefits" start to disappear.
Let's face it, heterogeneity in mobile phone OS is permanent. At the bare minimum, Nokia will continue to champion Symbian, Motorola will push Linux, HTC is making a good living with Windows Mobile, and assorted proprietary OS's continue to make traction because consumers don't care.
It doesn't matter if 2008's coolest phone runs on FORTRAN or has a dozen monkeys with abacuses inside it. People will buy it, the way they bought the Moto RAZR, because it's cool. And therefore operators will clamour to sell it, irrespective of their internal goals to reduce the number of OS's they support.
So - OS diversity is a baseline. Most manufacturers recognise this, and most mobile operators as well. Even the most proscriptive, NTT DoCoMo, which develops its own software stacks, dual-sources from Symbian and Linux.
As a result, the focus of the action is moving a little further "up". I am seeing a revitalised set of attempts to compete at the "application platform" layer. Clearly, Java and BREW have been around for a while. DoCoMo has championed iMode around the world with a measure of success. Most of these can work "on top of" multiple OS's (although I'm not aware of any BREW-on-Symbian phones).
I see various new attempts by the Java and GSM communities to create essentially a "BREW clone, but without having to deal with Qualcomm". The OMTP and the approach of firms like SavaJe exemplify this. I see Qualcomm itself use its success in 3G chipsets and possibly its UIone interface solution to subtly push BREW towards wider acceptability anyway. I see DoCoMo getting greater traction with iMode (and maybe FOMA in the future?), and the Koreans seemed very enthused about the possibility of exporting WIPI. Arguably, even higher-level software layers like Macromedia Flash Lite, and Surfkitchen's & Action Engines' UI tools could be counted in this category too. If it didn't have more pressing concerns, I would have expected RIM to be playing harder here as well.
To some extent, this is the promise of J2ME, but updated with better functionality, security, less fragmentation and better end-to-end control by operators.
Where does this leave the underlying OS's? I'd argue that - except for specific usage cases - they are heading towards commoditisation. Clearly, they're not commodities in the way that suppliers of oil or coffee beans are - the investments and switching costs remain very high. But I see the debate remaining very restricted to internal bean-counting and engineering debates inside handset manufacturers, in a similar fashion to phone chipsets. I see no real reason why Symbian vs. Windows Mobile should get more coverage than Texas Instruments vs. Infineon.
Do you know what chipset is in your phone? Or what its differentiators are from its competitors? Is there a dedicated blog site extolling the virtues of Agere-based handsets? Sure, these matter if you're building phones. But so does the type of battery, the audio chip and a 100 other things. Add the OS to the list.
What are these "specific usage cases" I mentioned? Well, obviously Nokia's range of high/mid-end devices is one of them. Then, clearly, the Internet is stuffed with enthusiastic evangelists of one OS or another, who would gladly beat each other to death with the device of their choice. Certain groups of developers have natural leanings towards Microsoft or Symbian too, for example in the enterprise. But the Holy Grail of the OS vendors, to have mobile operators standardise on specific OS's, seems to have run out of steam. It's also much easier to test and develop new solutions on smartphones before pushing them downmarket - there's a reason why so many of the DVB-H mobile TV trials are on Symbian-based phones.
So, operators are increasingly happy to standardise on application layers, and have these run across multiple OS's. Sure, they might prefer fewer OSs - but we're maybe talking reducing from 10 to 6, not down to 2 or 3.
But fundamentally, we're moving to the "not-quite-lowest-common-denominator" position, abstracted a layer from the OS.
What this means is that OS's are not major differentiators for manufacturers. I have a feeling that Symbian may end up as a loser here. It's notable that Panasonic has shifted allegiance from Symbian to Linux this week, and that the Koreans I spoke to a couple of weeks back had tested - and rejected - WIPI-on-Symbian.
Microsoft's position is more obscure, but its focus on IT-type applications and more corporate-aimed devices may mean it runs parallel to the more consumer/content-centric WIPI/Brew/Java/i-Mode debate.
Linux seems to be a potential long-run winner (recent developments in APIs and standards are helping), but it "isn't as easy as it looks" seems to be a familiar refrain in my research. A newly-revitalised SavaJe is an outside bet, but my general belief is that all this means proprietary OS platforms still have several years' breathing space in front of them.
[Oh, and any VCs interested in funding my MPMMAOS (massively-parallel mobile monkey abacus OS) programme, please contact me at the usual Disruptive Analysis address]
Tuesday, December 06, 2005
Owing to various client and travel commitments, I haven’t had a chance to fully write up a lot of what I learned in Korea, about innovations like WiBro wireless broadband, and DMB satellite mobile TV, as well as some of the more general trends in mobility.
However, I want to highlight a few things that appear to differentiate the Korean approach to wireless versus Europe and much of the rest of the world.
Perhaps the most obvious is just how much “Korea Inc.” is pulling together on WiBro. This is represented by a combination of the government (through 2.3GHz spectrum licencing), SKT and KTF in services roll-out, and Samsung and (to a lesser extent) LG on network infrastructure and devices. I got the distinct impression that WiBro is being set up as a showcase for all these firms to establish and extend their global position in the coming major transition to WiMAX. As a sidenote, it will be interesting to see if any of these companies can lock down a significant IPR position and patent portfolio, around the practicalities of implementing a WiMAX network. This could, for example, give Samsung a much stronger position in the wireless infrastructure market than has been the case in the cellular domain in the past.
Next is the Korean attitude (and reverence) towards technology. It’s very fashionable for many European and US companies to downplay the importance of underlying enablers. “Oh, we don’t want to focus on the nuts and bolts, it’s the content / services / applications / user experience that’s important”. Nonsense. In many cases – and the Koreans get this – if you bring the technology (and get it right, and get it early), the rest will follow, even if it’s not obvious to begin with. Both science and technology (especially IT & computing) have always worked this way, and it reflects badly on many western telecom companies that they are trying to present a fuzzier image now, largely in order to please mass-media and relatively tech-phobic populations. It was very notable that a slide from KT listed as key success factors “network, terminals and services”, and not brand or user experience.
[It’s also notable just how tech-savvy (and science/tech-educated) Koreans are. More generally, a similar story applies in China and India as well – just look at the science student population in Asia, and the attitudes of their peers, versus the trendy sneering about science and engineering geekiness in countries like the UK.]
So, in Europe, the idea of fast, open, mobile broadband would be viewed as scary. “An IP pipe??! But what about commoditisation? Can’t we just bury our heads in the sand instead and hope it goes away?” Operators are intensely defensive. “and, er, we’re really scared of Skype, so we’re going to price this stupidly high to skim the market and deter VoIP, and try and create some sort of allegedly value-add service to flog to our customers instead of giving them what they really want”. In my view, this is a surefire way to get hit much harder by the IP steamroller in a couple of years time.
But in Korea, not one of the presenting companies at the iMobicon conference in any way implied this was a major concern. The attitude seemed to be more like “sure, some people will want to use Skype or something else competitive. But to be honest, we’re not going to be able to stop them, so instead we’ll compete ourselves, creating our own VoIP and other IP applications that are better, easier to sell and use, and integrate with users’ other services”. This is much more of the “if we can’t beat them, then let’s join them….and then beat them at their own game” philosophy. Oh, and let’s have LOTS of bandwidth while we’re at it.
Another key observation is the aggression of the major Korean operators in extending their reach overseas. The SKT/Earthlink MVNO deal in the US is likely to be just the tip of a new iceberg. I got the distinct impression that the Korean operators feel they could be more successful in exporting mobile content platforms than DoCoMo has been with iMode. They characterised the iMode business model as being too inflexible, and seemed to suggest that a variant of WIPI and extra application layers could have wider applicability.
This is also reflected in the Korea operators focus on service brand (eg for their mobile music services like MelOn, or segment-focused service plans), rather than their overall corporate brand and image. This is extremely different to the Voda-style “it’s all got to be bright red with the same logo” approach to branding.
It’s also interesting to see an apparent inversion of the usual timescales for service development and investment horizon in Korea. Service trial and deployment schedules seem hugely compressed. WiBro commercial trials are scheduled for Feb and March 2006, with full switch-on intended for around June. No unwieldy 9-month trials and 4-month subsequent fiddling about with final service definition and launch.
Conversely, capacity for things like network backhaul is put in with a view to the long term. Korean cell sites typically have metro ethernet fibre connections. No short-term, build-it-incrementally “oh, just put in an E1 or two, and maybe upgrade to microwave or fibre some other time” mentality. This is one of the reasons why it is now simple to overlay WiBro, or future terrestrial mobile TV transmission on the cellular infrastructure.
Lastly, a few more “snippets” that I don’t have time to delve into now:
- Koreans seem happier with “chunkier” handsets then I expected, emphasising function over small size.
- Lots of slide form-factor devices and “swivellable” screens for horizonal viewing
- Somewhat surprisingly, there was much more emphasis on integrating WiBro and DMB into handsets, rather than WiFi, which was conspicuously downplayed. I think this fits with my general view that integrating WiFi into phones is much more difficult than expected, as it is not inherently a “service-oriented” technology, but is usually privately controlled.
- The WiFi hotspots in places like airports and the convention centre automatically tried to log my PC on, proactively looking for a certificate on my PC, rather than waiting for me to use the connection manager & sign on manually.
- Korea probably benefits from having a few very dense urban areas, which makes it easier to roll out new services more rapidly
- One conference participant (I forgot who) said that they had trialled implementing WIPI on top of Symbian – but decided against it.
- LG reckoned they have their handset development timescales down to as little as 3-4 months in some cases.
One last comment that underscores the “techiness” of Korean mobile…. Samsung’s convergence roadmap goes rather further than just “basic” mobile TV and FMC function. Their presenter gets the prize for the first cellular presentation I’ve seen which mentions nanotech and biotech as being in line as future handset technologies….. put me on the waiting list for a DNA-locked phone please.
I've been a big believer in the upsides of the "broadband operator + MVNO" model for a while - not much legacy PSTN to worry about, someone else deals with the nasty complicated radio bit of the cellular service, and so you, the operator, can optimise least-cost routing and clever IP-based services to your heart's content. You don't have to be over-focused about shoe-horning traffic over a cellular network, just to fill up your available capacity that you've got with your expensive spectrum licence and network build-out. And at the same time, you have a nice shiny new IP network, modern and flexible IP billing system & (hopefully) a customer service function that doesn't get scared by words like "firewall" or "PC".
In theory, combining fast IP pipes and resold & well-branded cellular should be a match made in heaven for a service based on dual-mode cellular/WiFi phones, for example. And, unlike an operator such as BT with Fusion, Virgin comes with a big existing mobile customer base.
However, there are a couple of flies in the ointment. For anyone out there doing their due diligence on this deal, you might want to consider the following:
First, most MVNOs (especially in Europe) have an overwhelming proportion of pre-pay customers. Many will be young people who use someone else's broadband & whose parents / flatmates / university residence halls' managers may be unwilling to change to benefit someone else's bundled quad-play
Secondly, many of the favoured approaches to dual-mode services - not just UMA (which regular readers will know I tend to criticise anyway) but many SIP-based alternatives too - need boxes to be integrated fairly tightly into existing cellular networks. These usually require direct access to elements in the cellular core such as MSCs (switches) and HLRs (home location registers - essentially the database that tells the network where you are). My understanding is the most MVNOs' deals don't give full access to these network elements, and that "technically deeper" MVNO deals like BT/Vodafone are relatively unusual in this regard. I don't know what the precise network-side ins & outs of Virgin's MVNO set-up are..... but I'd advise any M&A guys to take a close look at things like "Access to the 'A' Interface" before putting up any persuasive Powerpoint slides featuring WiFi/cellular handsets & (yawn) "Seamless Roaming".
On the face of it, nothing particularly unusual, except the very heavy bias towards data services, especially texts, and to a lesser degree downloads.
What I want to know is exactly how the accountants and management work out the value of each element. Is it based on that element's included allocation in the bundle? Is it based on the actual usage of those allocations? And exactly how do you work out the relative implicit value of say, actually using 800 texts vs £7 of downloads vs. all 150 voice minutes?
This advert & bundle has reminded me of a question I first thought of a couple of years back.
How exactly do carriers work out the % ARPU attributable to data services, when most of those services come pre-bundled at a non-specified cost per message/download/video minute in a plan? Is there any consistency between mobile operators? Are these numbers audited, and if so, to what rules? Do the rules vary by country, or change over time?
I'm not getting at 3 specifically here; this applies to all operators in a similar fashion. Maybe everyone shouldn't be so surprised that mobile data ARPU percentages keep creeping up, as they're such an important and closely-watched metric.... and (perhaps) so easy to "fiddle".....
Friday, December 02, 2005
However, the thing which really caught my eye was the small KTF-branded repeater dangling from the ceiling by a cable, in a small downstairs bar in Seoul. (I think it was one of these , but unfortunately I didn't have a camera with me at the time). Nothing flashy, not even a large venue with 1000s of people (I reckon there could have been a max of 150 in the venue at any one time). Just a pragmatic installation to help people make calls where they want.
I keep hearing stories that European operators are going to be much more aggressive with these types of self-install solutions to improve coverage in retail or public facilities, but I see precious little action. I'm not sure whether there are clear technical or regulatory reasons for this, or whether it's just down to apathy or business model uncertainty. And although I'm hearing a lot more "chatter" about mass deployments of picocells and residential-grade "femtocells", offering extra base station coverage (and extra capacity), I'm pretty sure the time horizons are quite far out.