Views on the new media market

The myth of open

It seems in the mobile handset ecosystem that there is a religion around open. Open is good. Open will save the day. All your problems will be solved by openness.

Yet, still today large numbers of handsets that are made, and sold successfully around the world are not open. Why?

If you look at the mobile world from the perspective of a handset OEM - everything looks open. Nokia can implement whatever they want on an S40 device. They have access to every API imaginable. Their ability to deliver is limited only by their imagination, and their ability to sell the resulting devices.

Samsung can put whatever the want on the Samsung Hanset Platform - including changing the underlying RTOS as they switch between different chipset vendors. Sony Ericsson can integrate whatever they fancy into their central platform.

What about the ISVs though? They can't deliver to all those closed platforms though. Let us look at who's been successful in the mobile ecosystem so far - who's shipped over 100 million units?

What stands out to me is that few of these solutions have had their success from open platforms. In fact the closest we get is Symbian and S60 - which up until around now has been closed - you needed to be an OEM with a few €millions to spend on software licensing (not to mention a few €100 millions on R&D) to build a handset with these platforms included.

So after 10 years of open platforms the successes continue to be from solutions that are portable, C based engines that deliver on proprietary, closed mobile handsets.

In fact if you are an ISV open platforms look like a bad thing. Why? They increase your cost. In the case of a closed, proprietary platform an OEM will be adept at taking external, portable C sources and integrating this with their platform - using their own engineers, at their own cost. On an open platform as an ISV it becomes your problem - you need to hire people with the right skills and tools and build and test on the platform. Result? You invest more of your time and effort on porting to multiple open platforms and less time and effort on building your unique, innovative solution.

So openness is actually bad for the industry. So what would be a better approach?

Given that everyone who ends up being successful builds a portable, C based engine that is integrated onto devices - how about going back to that? ARM is ubiquitous in the mobile industry - so we can even support a binary format for those who can't or won't give out their source code. While this doesn't help application vendors who want a full featured platform - it does provide a consistent baseline, and reflects what everyone is actually doing.

Why did music industry spokespeople get to skip Economics 101?

The Sunday Times reports that Elio Leoni-Sceti the current CEO of the private equity owned EMI is repeating the lament of "stop the pirates, they are bleeding us dry" that's been periodically heard from the music industry.

Of course he's supporting this with research, which I believe is helpful to quote:

"20% of Britain's £1.3bn music industry comes from digital sources, by volume consumed digital makes up 70% of that market. Of that 90% is illegal, but the 10% of digital music bought legally is growing at twice the rate of pirated tunes"

The very final part of this is interesting - as it hints that as the music industry makes music available legally consumers are willing to pay for it, but the first part hints at the common fallacy that piracy is taking money out of the music industry. Maybe it isn't.

Let us go back to basics here. Let's look at a typical market. The lower the price of the product, the higher demand. So if we draw out what this might look like we see something like below

So what we see is that as we drop the price the number of purchasers increases. This isn't suprising, as anyone who wandered past one of the Woolworth's closing down sales will have experienced - at a price low enough people will by all sorts of unimaginable things!

Now, with this basic lesson in demand curves, what would we expect if the music industry priced it's product as low as possible? Well - we'd expect demand would dramatically increase. The economics is proved right by EMI's research - 90% of music is illegal - so at "free" we find out that demand is 10x the demand at £8 an album. Two great data points for drawing the actual demand curve for music.

The bad news here for the music industry is that this works both ways - as price goes up, so demand goes down - so were the music industry to succeed in "stamping out" piracy demand would drop, and there may be no actual change in revenue.

So what should the music industry do?

Fortunately economics 101 (which the industry execs skipped, as they clearly don't understand the basics) also contained the solution to their problems. How do you maximise revenue in this sort of market?

Price skimming is the answer - capture the maxium that each and every consumer is willing to pay for the product. Of course in the real world this is harder, so it would take a genius to work out how to do this. So in true music industry tradition, let us turn to the brightest, most talented man in the room. The drummer. In this case the drummer in question is Josh Freese - who's come up with a price skimming strategy for his solo album Since 1972. Of course the bad news for music industry execs is that this strategy might just be successful - and the days when a single, undifferentiated product with no additional effort can succeed in generating large revenues are gone - the future of the music industry is one of continued investment to differentiate products and offer consumers what they want to part with their cash. Welcome to the real world folks, I recommend that you swallow your pride, and attend that Economics 101 course - you'll benefit from it, as will your businesses.

In the mean time I'm going to eagerly await Josh's next album

Internet video is still under control

In the midst of all the excitement around new business models and the internet revolution it's easy to forget that the old order continues to exist.

The latest demonstration of this comes from Hulu and Boxee. If you are in London you'd be forgiven for saying "who on earth are these folks?" as both of them have almost no impact over here - as you can't get the content you want (which would be BBC iPlayer and the ITV, Channel 4 and five services on Boxee please)

What's happened? At the content owner's request Hulu has asked Boxee to remove Hulu's content. Boxee has agreed. This was at the request of the content owners it seems. The consumer seems to be the only loser here as this makes rational sense for all the 3 parties. So what is going on - let's look at the perspective from each of the three parties.

Content owners

The main current source of revenue is from cable subscriptions and TV advertising on cable. While they are interested in exploring new models for this (hence the investment in Hulu) at the moment the overall majority of revenue comes from traditional routes. Boxee threatens those - people can replace their cable subscription and set top box with a device running Boxee and get the same content. Thus reducing the revenue stream from the traditional route, with no clear replacement yet. So as a content owner if I want to continue to fund exploration in new business models I need to maintain my revenue stream - accordingly shut down Boxee

Hulu

Hulu's is utterly dependent on it's content - without the content Hulu has no value proposition - so it needs to make sure that it keeps it's content providers (and owners!) happy. While Hulu provides the potential to monetize the internet traffic the problem isn't solved well enough yet. So Hulu needs to comply (though it must be frustrating as this means that they can't explore the possibility of monetizing the content on TV sets).

Boxee

They need to comply - otherwise they risk both lawsuits, and an escalating technical arms race as Hulu engineers come up with an inventive way to "turn off" Boxee, and Boxee's engineers come up with workaround solutions. This would ultimately be destructive for both Hulu and Boxee - as it diverts resource from their respective core value propositions - Hulu's is monetizing video content on the web, and Boxee's is bringing high quality TV experience with internet video.

So - what next? I'd expect that we'll see similar solutions come back in the future - it's in all parties interest to get this to work - I'm sure Boxee would happily license to Hulu and their content providers a solution for a "next generation" set top box.

A look to the future - where the semiconductor companies are taking mobile

Hot off the heels of the launches of lots of shiny plastic from the phone manufacturers it's interesting to have a look slightly longer term, and see what the semiconductor manufacturers are shouting about.

The Semiconductor manufacturers use MWC to talk about their soon to sample chips - those that have been on roadmaps being discussed with handset manufacturers for at least 6 months, so this isn't really news to most of the industry, but it's interesting to see what's made public - as those are the chipsets that are getting real interest and design wins from manufacturers - so chipsets that we can reasonably expect to see in handsets.

A word of caution on the announcements though - the "headline" numbers that semiconductor manufacturers quote are typically without the multiple layers of software sat on top of the chipsets - and as any hardware manufacturer knows, if it wasn't for the software guys the devices would really sing - so expect real life performance to be "toned down" somewhat from the specs when the reality of getting the things working reliably hits.

Also these are devices that will sample during 2009 - so won't appear on handsets until mid 2010 - so don't hold your breath!

Qualcomm

Qualcomm has made massive strides in the european market in the last few years. They have always been a major player in the US market, but traditionally were "nowhere" in the European market. Fortunately for Qualcomm they have been able to take advantage of their IP leadership in CDMA technology to create some great WCDMA chipsets - and have re-used their application processor expertise from the CDMA market to offer a great range of chipsets across the price spectrum. You've probably come across a Qualcomm chipset in an HTC operator branded windows mobile phone in europe.

Qualcomm have a two pronged strategy for europe. On one side they are showcasing more Snapdragon - the high performance chipset that goes up to netbook class devices.

I'm actually more interested in their snappily named MSM7227 which is a low cost smartphone chipset - targetting sub $150 smartphones in 2009 (which could well lead to sub $100 smartphones in 2010) A strong offering that should see their market share grow - as we can see with the surprise announcement of a deal with Nokia.

Texas Instruments

The once invincible king of the mobile chipset market is looking increasingly vunerable. TI continues to suffer from it's lack of investment in 3G solutions early enough that has opened the door for Qualcomm, and with consolidation in ST/NXP/Ericsson the space for TI to maintain it's lead seems to shrink.

The two key announcements are around it's OMAP family of devices - an evolution of the OMAP3 devices to a 45nm process and lower power consumption, and the introduction of a Coretex-A9 based OMAP4.

While TI's customer relationships remain strong I expect to see more promiscuity among it's customers as they look to build a broader portfolio of suppliers. Maybe we'll get a major acquisition from TI in 2009 to rebuild it's 3G position.

ST Ericsson

The other major powerhouse is still digesting the merger - so not much from them, other than a statement around their support for the Symbian Foundation - I'd like to see more around their other platform support - but expect more during the year as they start to execute in the new structure.

A notable mention goes to Broadcom who announced a win in Nokia for 3G devices - more pain for the team at TI.

Myriad - Purple Labs + Esmertec

Last week there was an interesting development in the low end platform space.

Esmertec - the listed Swiss Java runtime company acquired Purple Labs - the french private equity owned group that acquired the browser assets of OpenWave and the software parts of Sagem last year.

This is a real sign of the consolidation and cost cutting in the low end mobile platform market - the combined company has a full suite of software products necessary to make a low cost handset and the scale to manage the challenges of operator customisation needed to get the product wins with operators in this space.

This is a bet on growth in the low end space - the featurephone market, a market that's been largely neglected by the major vendors in their announcements (it's not as glamorous as the smartphone space), but in the case of Nokia & Samsung is where they are seeing most of their growth coming from.

The opportunity for the Myriad group is to enable new manufacturers to enter this growing space, though the challenge is that the mobile business has always been about scale - the companies with the volumes can get the best price for components and secure their supply chain - they win. Traditionally this has been the big 5 OEM manufacturers.

Might this change going forward?

Maybe - the growth in features and hardware form factor in the "smartphone" end of the market (large screens, touch, web browsing) may be splitting the mobile market in two - the "high end" components with processors moving toward ARM Cortex-A8 and above and large VGA+ size screens with touch, and the "low end" that has ARM9 processors and still QCIF/QVGA screens. Will there continue to be economies of scale for manufacturers producing both?

Late Monday at 3GSM (OK, OK, but it's still snappier than MWC)

Late monday had more announcements on handsets from MWC - here's my take:

HTC

Our Taiwanese friends kicked off with a slew of Windows Mobile handsets - after all the hype around Android and the HTC Dream (or T-Mobile G1) it's easy to forget that HTC are the biggest Windows Mobile manufacturer and the folks in Redmond haven't gone away, and aren't likely to go away soon. Nothing too ground breaking from HTC - sensibly incremental evolutions of last years models to continue to appeal to their core business customers of Windows mobile handsets.

A particular mention goes to the HTC Touch Pro2 - which has one notable innovation, when you put it face down on a table it behaves, and looks, just like a speakerphone. Great innovation here - I particularly like the design where you place the handset face down on the table - so you don't get embarrassed by incoming SMS messages during a conference call being visible. Well done to HTC in continuing to focus on the core market and understanding what their consumers really need.

LG

A surprise here - LG seems to be committing to Windows Mobile in a big way - ahead of the other platforms (LG has previously dabbled in Symbian with S60 devices, is rumoured to be launching an Android handset). I'm surprised at this because Windows Mobile doesn't seem to have a strong following outside of the corporate/enterprise market - and with Microsoft's Exchange ActiveSync protocol being licensed to everyone it's hard to see Windows Mobile getting growth, without a more radical change in the platform to appeal more to consumers. I hope that LG can focus on this and start producing some innovative handsets - but we'll see what the real impact of this is in 2010.

Acer

The ubiquitous netbook and laptop manufacturer announced some - predictably - Windows mobile phones. Given past history of PC device manufacturers producing mobile phones I'd expect that there won't be anything stunning here. Nothing to see, move on.

Asus-Garmin

Some GPS phones here - one linux based and one windows mobile based. I have to admit I'm confused by this - I don't really understand the segment that they are targetting with these devices.

So far the folks notable by their absence are Motorola (remember them?).

Monday at 3GSM, sorry MWC, what are people coming up with to beat the recession?

One of the challenges of the mobile phone industry is that it takes 12 months to take a phone from "concept" to "launch"

This means that there isn't much time to react to market changes - so the phones that we will see being launched this week were dreamt up in the heat of early 2008 - which if you think back was boom of Web 2.0, and iPhone fever (why don't we have a touch screen device?) - so some of them may look a little dated. But the phone industry moves to it's own beat - announce products in February, get your final QA done over the summer (by all those students looking for summer work testing devices) and then get volumes ramped for Q4 and the Xmas rush.

So what do we have so far?

Sony Ericsson

Music and movies - W995 is the lead here, with a music and movie store integration. Idou with Symbian (S60 here again) and lots and lots of megapixels.

So Sony Ericsson are targetting folk who are after a music and movie player that isn't iTunes integrated (is the strategy here lower cost than Apple?) - and people who want more megapixels. Unfortunately SEMC seem to be suffering from platform anxiety, with in their product line up now central platform (the Walkman phones), Symbian (Idou), Windows Mobile (Xperia) and rumoured Android devices. While this is great news if you are an engineer with integration skills it limits the scope for SEMC to invest in core product improvements or gain economies of scale from any one of these platforms. To me this looks like SEMC weakening their strategic position and focus, which is unfortunate as they have been comparatively strong in previous years.

Samsung

Samsung's focus is on touch - with a range of devices from the Ultra Touch, Omnia HD, BeatDJ and BeatDISC products - their TouchWiz UI is appearing on more devices - this is going to be interesting to watch - while Samsung is promiscuous in it's platform usage it seems to be continuing to focus on it's own platform for innovation then porting solutions across to others I'm going to be interested to watch how Samsung fares during 2009.

Nokia

3 major announcements from the Finns - the E55, a "half qwerty" type device, the E75 - a qwerty slider device that joins a stable of devices including the E71 and the N95 - the most interesting thing here is that Nokia has come over all "qwerty" - finally a reaction to the popularity of Blackberry?

An even more curious launch is the services launch - Ovi Store. Which is interesting because it doesn't take advantage of all those qwerty keyboards! The business proposition for content providers looks solid though - 70% revenue share to them, and with Nokia's reach I would expect Ovi Store to be a winner.

Comparing Google with Vodafone

I've been comparing Google and Vodafone recently - it's interesting to contrast the strategies of two major players - one in the internet and media space, and one in the telecoms space. There is a symbiotic relationship between the internet and telecoms companies - it's the telcos that carry the traffic for the internet companies - and the telcos have a lot of influence over default settings for mobile customers - they can drive traffic from their portals and handsets to brands. Let's look at some headline stats

Q4 08	Vodafone	Google
Revenue	£10.4bn	£4.0bn
Operating margin	29%	32%
Market capitalisation	£72bn	£83bn

Now, not withstanding there are some key differences in the market environments:

• Vodafone operates in licensed, regulated markets with significant capital expenditure on network build out
• Google operates in unlicensed, unregulated markets with minimal capital expenditure required to reach new consumers

There are interesting parallels

• Google derives its revenue from selling advertising to people who want to get their message to other people (business to consumer)
• Vodafone derives its revenue from selling network access to people who want to reach other people (consumer to business or consumer to consumer)

Looking at the financial performance and the relative valuations the market clearly expects greater growth from Google than from Vodafone. So how do Google go about this?

Google need to

• Drive more & higher quality web traffic - connecting people to the businesses they want
• Deliver better targetted advertising - connecting businesses to the people they want

Vodafone need to

• Drive more connections between people and businesses
• Drive more connections between people and other people

Google do a great job of this - they offer superb tools like Google Analytics to help people optimise their websites and drive more, better traffic.

What about Vodafone? How do they help me as a consumer connect with people? They provide me with a phone number. That's about it - they don't provide me with tools to connect with people, they don't enhance my experience, as a business they don't provide me with information to help me understand more about the people who are calling me. They don't seem to do much of anything short of provide the network.

Let's imagine a world where the management of Google were installed in Vodafone, what sort of services might be provided?

• The ability to install "analytics" on your phone number to get details of when, and where people call you from
• "Analytics" for your website to see where mobile customers on Vodafone's network are and when they access your site
• Information to show you when people you are sending an email are available to call - so that you can swap a low revenue email for a high value phone call

Overall Google seem to have spent much more time and effort understanding what drives the organic growth of their business - Vodafone seem content with geographic expansion and defensively maintaining market share in europe.

On the internet nobody knows you are a dog

The most recent identity scandal is the faking of the Dalai Lama's twitter account. This raises an interesting question of identity and authentication on the internet. Peter Steiner had this right 16 years ago. Identity on the internet is an interesting challenge. It is trivial intially to create a "fake" identity on the internet. It's harder to maintain a fake identity - though it's still possible. For many applications identity doesn't really matter - you are who you say you are. For other applications, typically those involving financial transactions it's critical that your identity can be authenticated. When thinking about identity it's useful to identify three categories - and be clear that there are different cases that suit these:

• anonymity
• who I say I am
• authenticated for financial purposes

In many cases anonymity is fine - Google doesn't need to know who I am when I search, and BBC iPlayer doesn't need to know who I am to show me what I last watched - auto generated, transient identities here are fine. These solutions are great as they avoid the need for users to create and remember usernames and passwords. I'd like to see more solutions like this.

Who I say I am is the typical internet identity - when you create a hotmail account, a facebook account or a twitter account your identity is created by the sum of what you say. It may bear no relation to who you actually are - that's fine, and your identity increases in value and credibility the more you invest time in it - whether it's the "real you", or your online identity that's created to represent something that you want it to represent. Some of these may allow financial transactions - pre-pay mobile phones is a good example here - in some countries it's possible to get mobile phones without any formal identity - you pay cash for the SIM card, and get a mobile phone number as a result.

Sites like ebay, paypal have a financial identity - for them fraud is a reality, and has a real cost - so they need to work harder to tie back the identity to a genuine source of funds.

The "who I say I am" category is the most misleading - and unfortunately the one that's most common - when in fact auto generated "anonymous" accounts would be fine for many solutions - without the implied identity that misleads people.

The network as a hardware feature?

Amazon released the second version of the Kindle in the US yesterday. While it's frustrating that there isn't an equivalent solution in the UK so I'm not able to try out the solution for real (I don't believe the hype until I've tried using a device for one month). What Amazon have realised is that if you focus on what people want (to be able to read books) and use technology to help them achieve that goal in the simplest way possible you can get a hit - and it does seem to be a hit, according to Jeff Bezos 10% of the book sales in the US were ebook sales.

Now not withstanding that it's only going to be people who read a lot who will buy a Kindle in the first place (at $359 it's not a small expense - clearly a luxury purchase compared to a $10 book) it's still clear that Amazon have got something right.

What is really interesting is that they have eschewed the traditional technology solutions to the connections - no USB and PC tethering, no WiFi for free internet access, but a private, one network only wireless connection. No monthly rental either - it's a one off cost. Of course Amazon can keep you paying for that network connection, through three simple revenue generators

• sell more ebooks
• sell newspaper and magazine subscriptions
• sell new kindle models every 18 months

So the network here becomes a hardware feature - a cost that's built in to the device, rather than decoupled (buy a phone, then choose the network you want). This approach can be applied to more sorts of devices - portable navigation devices (for traffic updates), wireless photo frames (for picture updates) - I'm even sure it could be applied to remote controlled rabbits. So how many mobile operators offer simple development kits and business models to companies interested in creating these sorts of solutions? In the UK none that I'm aware of. The conservatism of the mobile operators is a limiter here - maybe part of the reason that Amazon have not, yet, launched the Kindle outside of the USA.