Not Convinced By .mobi

Russ has a post about why .mobi is a good thing:

Sometimes I’d wonder if we really needed Yet Another Domain Name or if a new domain name would cause confusion, or if separating out addresses for the mobile web was really a good idea. But I have finally decided once and for all it’s a Good Thing™. Here’s why in a sentence:

We’ll finally have a standard format for initial mobile web navigation: companyname.mobi

This is the thing that’s really missing from the mobile web. There’s been no way to know exactly what the URL was for the mobile URL. There’s a bunch of redirect or handset detection magic that happens, but the end URLs are all over the place.

I’m still not really convinced. Why not just do the redirects from your standard www.example.com URL? That’s the one everyone guesses or knows already, after all, and it’s not so hard to make it work seamlessly. Also, I don’t need to type in a search engine address, since the default page on the browser comes with a Google search box already.

I do think the guidelines as to what a mobile friendly site should be are worthwhile though. I suppose we’ll all have to buy the domains anyway though, but I’ll only do so grudgingly.

Tags: .mobi, mobile

PSP Media Manager

A couple of months back, Sony Media Software got in touch with me to ask if I would support RSS enclosures on some of the Tapestry comic feeds. They couldn’t tell me why, though.

Well, as you may have seen, the reason was the PSP Media Manager software. This software does a whole raft of groovy things such as transferring photos and movies to your PSP and allowing you to back up your game saves, but from the Tapestry point of view, you can subscribe to the feeds and get the comics downloaded straight to the PSP for reading on the move. Nifty. If you have a PSP, you should check it out. I know I would if I had a PSP (hint hint).

SMS your kettle

Via Martin, SMS your kettle:

Fear not, for PG Tips have invented the SMS kettle. You can now put the water on to boil at the flick of a switch.,..oh, sorry, touch of a button. Just text “switch on” to your kettle’s phone number (there are so many things wrong with this…do you want your kettle to have a phone number?) and it will oblige.

Surely Bluetooth would have been a better option? Although I’m sure it wouldn’t take long until there were horror stories of kettles being bluejacked (for want of a better term) and turned on repeatedly when there was no water in them…

Nokia 770: What It's Really For

I’m a little late in posting about this fancy new device, but there seems to be some confusion as to why Nokia have introduced it. Well, I can exclusively reveal that it is designed to overcome the final hurdle of ebook acceptance: reading in the bathroom.

As per usual, Om Malik has a good analysis, as does Simon Woodside.

Remodelling In Progress

As you may have noticed (unless you exclusively read this via the feed), I’ve been doing some remodelling here, finally adding proper mobile versions to the site (both XHTML Mobile Profile and bog standard WML).

Mobile Profile support was straightforward to add, since it’s just another stylesheet linked from the template (media type handheld), but supporting those phones which only handle WML 1.1 is a bit more tricky.  I wanted to have seamless mobile access as well, none of this “for the mobile edition visit http://example.com/mob/” nonsense. So some mod_rewrite voodoo has been invoked.

In addition to the main full-screen XHTML index page for the site, I also created separate index pages for the Mobile Profile and for WML. I didn’t need to create an extra index for the MP version of course, but I wanted a slightly different layout for it. Using mod_rewrite to send browsers to the appropriate pages was the work of a few moments:

RewriteCond %{HTTP_ACCEPT}  application/vnd\.wap\.xhtml\+xml
RewriteRule ^index\.html$   index.mobile.html          [L]
RewriteCond %{HTTP_ACCEPT}  text/vnd\.wap\.wml
RewriteRule ^index\.html$   index.wml          [L]

The individual entry pages for MP work fine just using the handheld stylesheet, but those pesky WML phones will require a different solution since you can’t just load (X)HTML straight on to them. It’s not ready yet, but basically I will be dynamically generating the WML entry pages via an XSL transform from the source XHTML. An additional rule in the .htaccess file will specify that any file in the archives directory with the extension .wml will be sent to a PHP script which will apply the transform to the XHTML original and send the resultant WML back to the handset. Et voila! What better excuse for learning XSLT could there be? I’ll provide an update once I’ve got my head round it all :-)

Still to come:

• More regular posting
• A Sparkline — to keep track of how often I’m posting :-)
• Some stuff I’ve thought of, and some stuff I’ve not thought of yet

Missed Calls As Communications

From an article over The Economic Times called “>Cellcos miss moolah on missed calls>

The Indian mobile user has mastered the art of missed calls communicating without answering the calls and causing a revenue loss to the service provider …

Very often, a couple or a group of friends who are supposed to meet at a particular place, choose to signal that they have reached by giving a missed call, rather than talk or send an SMS (text message) saving on the Rs 2 or Rs 1 that it could cost.

I don’t think this phenomenon is unique to India, or even to mobile operators. I remember my grandparents always asking my parents to “give two rings” when we’d made it home safely if we were travelling in bad weather for example. And in fact, many taxi firms do exactly the same thing to let you know they are about to arrive. To be honest, the article doesn’t discuss hard figures, so it’s hard to say if this is really having an impact or not.

What was of more interest was this paragraph:

For teeny boppers, missed calls are a way of life. After all, they have to balance pre-paid cards on limited budgets yet at the same time send loads of messages, play games, download ringtones and talk for hours on the phone with their various friends, provided its an incoming call. Missed calls help to save on talktime a precious commodity for teenagers.

This is what operators the world over should be worrying about: if their future customers are already against spending money for voice communications, what’s it going to be like when they all have some sort of VoIP technology on their handset? Let’s face it, “teeny boppers” are about as early adopter as it gets for this kind of thing, and in a few short years they’ll be amongst the most affluent consumers on the planet — which might sound great, but they’ll also be the smartest consumers on the planet.

Mobile Product Manager at Six Apart

(Via Mobitopia) About Six Apart – Jobs:

Six Apart (http://www.sixapart.com/), the company behind the Movable Type and TypePad weblogging systems/services, is growing and we’re looking for a product manager for mobile applications and integration projects.

If I hadn’t just taken a job (more on that soon, possibly), I’d have applied for this. Looks like an interesting role and is right up my street (so to speak).

Self-Service Visions

Almost a year ago, I wrote an article at Mobitopia about how mobile network operators should offer customer self-service in order to reduce their costs and, frankly, make my life easier.

Now, I don’t want to gloat, but it seems that those laggards fine folks over at McKinsey & Co have cottoned on to this as well (free reg. reqd.). In fact, they make the case that operator margins are so low now, that they can’t afford not to (sigh) 'e-enable' themselves (emphasis mine):

To reduce costs and simultaneously improve service, these companies must persuade their customers to move to far more cost-efficient automated channels—the Web, automated telephone systems, or wireless transactions—and streamline their internal sales and support operations. If they don’t, they will not be profitable in the years ahead.

Even better than that, they’ve actually prepared some numbers to back up my intuition about how it would save money:

A typical call-center transaction costs $8 to $10; the same transaction online costs $0.15 to $0.80. Printed bills cost four times as much as e-bills…

J2ME Content Handlers Are Indeed Wicked Cool

Via Erik’s Linkblog, comes this article on mapping J2ME applications to content cypes with JSR-211 (CHAPI):

In July, the Java Community Process released JSR 211, the Content Handler API (also known as CHAPI), for public review. The release counts as one of the most useful and innovative new APIs for J2ME development. In short, the API defines a communication model between applications (either Java/J2ME or native applications), by letting developers specify MIDlets as the content handlers for one or more specific file types.

This seems to me like they’ve finally tidied up the Muglet technology that has been available on Sprint’s network for the last couple of years, which did exactly this.

We used this for a game which sadly still hasn’t seen the light of day, and I can tell you that it works a treat, and it even seemed to make for a pleasant user experience. It also opens up different business models for the application providers, so it’s a very exciting addition to the J2ME family.

What bothers me about all these extensions for J2ME is that who knows if we’ll ever see it in handsets on a large enough scale to make it worthwhile doing anything with? Extension packages are fine for desktop or server-side Java applications, since you can easily just drop in the jar to add support, but it’s not that simple on the phones. Perhaps what’s really needed is a self-updating VM for mobile phones.

Think I’ll leave that as a project for someone else, though…

Searching On The Move

Via moco.news, comes this story from the International Herald Tribune about searching from your mobile phone.

The piece ends with a quote from an Ovum’s research director for wireless software, Jessica Figueras, who says, "It’s going to be very structured searching – images, ring tones." I can’t argue with that assertion, because it’s kind of obvious—mobile services have to be structured, otherwise they won’t be of any use to anyone.

Mobile search is something that has been talked about on #mobitopia from time to time, and the route that most of the operators have taken to providing search (keep users within the walled garden) will eventually break. Why? Well, as the phones become more open, the operators have less control. As an example, take a look at SMSbug which completely transfers SMS fees from the operators to the application provider. I digress, however.

There are a couple of things which really annoy me about searching on the move. Let’s take them in turn.

Most operators provide a "Where’s my nearest?" service, where they will identify ATMs or petrol stations which are near your current location. Orange, my operator of choice, also allow me to search for takeaways, restaurants and a bunch of other things. They even now provide directions, which they never used to (thus rendering the search pointless). So I now have a list of Chinese restaurants on screen, and I can certainly look at their address, call them, and get directions. But how do I know if the restaurant is any good? Where are the user ratings? Isn’t this a communications tool I hold in my hand? Yes, quite sure about that. So what happened to the community? The user could text in a rating for the restaurant after dining there, or make a note via the same page that they found the restaurant details on. Suddenly that becomes a really useful tool, as opposed to a handy gimmick.

The second thing that annoys me (and this may only apply to Orange) is that the search system they provide is so very weak (something which is detailed in the IHT article). What use are search results returned in alphabetical order? Sure, I can access Google if I want to, but that isn’t the default search system. Not that the results are of astounding use anyway, but never mind.

The structured approach to searching will definitely be the style that wins out over the broad search approach (he says, sitting on the fence as ever), so Google’s specialist tools such as Froogle, image search and the allegedly forthcoming audio search will be more useful on mobile. Rather than just having the word “Search”, an input box, and a “Go” link, users will choose “Search for X”, where X is the name of whatever it is they want to look for. Or perhaps even better, people will use the word “Find” in some situations, just as Orange already do with their location based search (“Find restaurants”, for example).

I think that independent channels will also benefit from having strong search systems on them, and one of the things that we are working on at The Games Kitchen is having the most complete range of mobile games that we possibly can. Once we get to a certain point, one of the problems we’ll have is enabling users to battle their way through potentially several pages of valid search results – and users shouldn’t be battling to get anywhere, especially on a mobile device.

I’m not sure that there will be a dominant search tool for mobile, but perhaps there will be a dominant provider of mobile search tools. Of course, it already looks like it might be Google, but I feel sure that there will be a few opportunities for innovative new search tools for a variety of purposes. Yahoo should probably get a wireless version of Kelkoo to market pretty quickly, if they haven’t already done so. And if they have, they might want to publicise it.

Update:: Cross-posted at Mobitopia, at Jim’s behest.

Nokia Finally SNAPs

In Nokia’s Smart Move: Scalable Network Application Package, Matt picks up on a story I only kind of glanced at last week, and then didn’t have time to post a response to his thoughts before now on account of being busy. Here’s the gist of the story:

Los Angeles, California. May 12, 2004 – Nokia today premiered the first multiplayer Java games based on its SNAP Mobile solution at this year’s Electronic Entertainment Expo (E3).

At the slight risk of sounding like I’m showing off, Sega Mobile demonstrated some multiplayer Java games at E3 in 2003. That was before Nokia bought the SNAP technology, so we’ll let them off with that, I think ;-) However, although it’s cool that they’re finally doing something with this, I’m not as impressed as Matt is:

• Why don’t developers (maybe) get the API until Q3 2004? Surely that’s already fairly settled?
• Why is it available on MIDP2.0 on Series 60 only? They’ve been working on it long enough to have tried tunelling over HTTP, which seems to work fairly well on most of the handsets I’ve tried it on.
• It doesn’t really do anything new, and the other systems that do the same thing all work on MIDP1.0 handsets.
• It won’t work on N-Gage, which as I’ve previously reported, is reasonably popular for playing games on amongst the people who own one.

These points aside, Matt seems most excited by the business model Nokia have proposed:

It’s a perfect strategy too: a small mobile gaming startup creates a killer game. It includes some multiplayer aspects that require the Nokia server components. All of a sudden they’ve got to license the server side stuff directly from Nokia or make a deal with a carrier. Everyone is free to create a killer game, but if you need the server side stuff, Nokia’s going to be making a buck one way or another.

The first point is that even if a small mobile gaming startup comes up with a killer game, an operator won’t touch it unless it has a big massive brand attached to it (unlikely), it would most likely have to be done via an independent channel (of which there are none… so far). If said startup decided to do everything on its own, it would probably need some serious cash behind it to get the server side stuff (plus kit to run it properly), never mind actually promote it. Even if they go for the hosted setup, I doubt that is a cheap way out.

Nokia don’t note anywhere in the press release if they are actually reading my mind. Maybe there’ll be some announcements soon…

As an aside, Matt closes with:

Does anyone else smell an N-Gage 2 featuring S60 v2.0 and MIDP 2.0 coming in the next few months? I think I do. Jim at All About N-Gage thinks it might be a possibility too.

I believe I already told them to do that, specifically by focusing on the J2ME environment ;-) The MMC cards are a losing strategy.

Carriers Clinging To Outdated Business Model Ideas Redux

As a follow up to my recent piece about carriers being more open, The Feature have a similar piece about why iMode has been so successful.

The answer? They created an ecosystem that allowed officially approved sites to make money, while they took a very small cut of the proceeds to cover billing costs, and then they kept all the packet charges as normal. DoCoMo correctly identified that they were not the best people to understand content, how to create it, and how to present it, but that they did know how to help other people make a lot of money by staying the hell out of the way.

Carriers Clinging To Outdated Business Model Ideas

Techdirt has a good piece on why mobile network operators should just focus on being what they are: network operators.

Bringing the content and channel into one organisation will not create greater value, and just because your network is wireless rather than a fixed line will not make a blind bit of difference to that fact.

As a carrier, all you should care about is utilising all the capacity and infrastructure you’ve invested in, and if you look at the investment required for rolling out a nationwide 3G service, there is no way on earth that one company (even if they are Vodafone) can invent and maintain all the services which will make full use of all that capacity.

People Play Games on N-Gage Shock

Russell Beattie, in a post primarily about the Nokia 7610, points to a Nokia document where we learn:

... the N-Gage may be a failure from a sales point of view, [but] the people who did buy an N-Gage have a massive ARPU. Very interesting – it shows that Nokia isn’t going to give up on the category. There’s an N-Gage 2 in the works already and Nokia realizes that even though the gaming market is a hard nut to crack, if they can find the sweet spot, the carriers are going to love them for it. In the presentaiton Nokia says, for example, that on average about 2-5% of consumers buy Java games, but 50% of N-Gage users buy them. Wow. If you’re a carrier, those are the sorts of numbers that make you re-think.

These ARPU figures are no great surprise to me, and it is basically what I as alluding to in this post about device divergence. It stands to reason that specialised devices such as N-Gage will have a lot more success with the types of activities they are designed for. The N-Gage is a phone that is great at playing games, so it’s no great shock to learn that the people who own an N-Gage buy and play games on it.

Anyone who’s listened to me talk about the N-Gage will know I’m no great fan of the device because of the strategy Nokia have taken. I said above that the N-Gage is a phone which is great at playing games. However, that does not make it a games console. Someone at Nokia obviously had delusions of grandeur when they started planning N-Gage, and figured they’d pitch it as a games console, thus setting themselves up for the massive amounts of criticism which has come their way since then.

Nokia have previously introduced new categories very carefully: the first music-oriented phone was very much a niche product, which paved the way for all the other music phones we’ve seen since then; and the first time I heard Nokia talking about N-Gage (prior to public announcement), they alluded to this fact and hinted that they would follow that approach again in future. As we know, N-Gage didn’t follow that approach, and has been slammed for it. If we imagine for a moment that Nokia had indeed followed the niche product path, then the sales figures aren’t anything like as bad; in fact, I’d declare N-Gage a success on that basis.

The problem comes with the other element of the strategy: pushing MMC based native games for the device. Big mistake. If they had instead concentrated their energies on having a kick-ass MIDP 2.0 plus all relevant JSRs implementation, then people would still have been able to make great looking, Bluetooth enabled games. Nokia, however, decided to “persuade” game publishers to bring their games to N-Gage. I’m not going to go into much detail, but basically, these games are not a reason to own an N-Gage, and they have cost Nokia the chance to claim that their games business unit is breaking even in 2004.

And of course, the marketing sucked and indeed, was recently banned in the UK (I would have thought for the fact that it sucked, but apparently some of it was vaguely rude).

So, Nokia, listen up – here’s a brief to-do list for N-Gage 2 (in the order I write them down):

1. Don’t call it N-Gage 2 – I don’t think there’s too much brand equity to lose, and having device sequel makes me think of it in the same as a film sequel: the same thing all over again, only a bit shinier;
2. Focus on the MIDP environment – that’s what is making money for the carriers and the game developers;
3. If you really want to be a game publisher, spin off a company, and hire some people who know about games;
4. Fix all the stuff people already told you was broken.

Historically, Nokia has excelled at producing platforms, and letting other people get on with the business of using those market dominating platforms to make new businesses. On N-Gage, it completely dropped the ball and got carried away in its own hype. Nokia is smart, though, so I’ve no doubt that their next games phone will be a whole lot better.

Linktone IPO

Linktone, a Chinese mobile content provider (ringtones, icons, etc) have gone public by issuing American Depository Shares on NASDAQ, and raising a phenomenal $86 million in the process.

Compare this to the recent IPO of Britain’s own Monstermob, who raised GBP 4.8 million in November 2003.

Admittedly, the Chinese market is marginally larger than the UK’s, but that’s still a tremendous amount of money Linktone have managed to raise.

No Multifunctional Devices Please

Via Tom Hume: Consumers don’t want it all, and they don’t want it now:

Many consumers are not interested in handheld devices that offer multiple functions beyond making phone calls or holding data, according to a survey by Guideline Research

This just bears out my long-held belief that the only effective convergence we’ll see at device level is network integration (whether GSM or WiFi).

Pricing Mobile Content

Via moco.news, I came across this report of a talk which Charles Dunstone, CEO & Founder of Carphone Warehouse made at the recent 3GSM World Congress. The story is primarily about how sales of the Nokia 7600 3G phone have out-done their expectiations, but towards the end, Dunstone starts talking about the pricing of mobile content:

Speaking about channels such as CNN trying to charge for their news, he said: “Everyone is too greedy. News is too freely available for people to think they can charge €5 per month [for it].”

Yep, I totally agree with that. I suppose some people might pay for a 10-second video clip associated with a news story, but that should be charged at €0.10 or so. The actual stories themselves should be free, because they are free everywhere else.

Dunstone goes on:

He added ringtones and games shouldn’t be priced above €0.99 and MMS photo messages not more than twice that of SMS.

This is the bit I disagree with, specifically about pricing games. Games for mobile devices should not be priced at the same level as ringtones. Why? Infinitely more work goes into producing a mobile game than producing a ringtone, which can largely be done automagically. On the other hand, a game is created from scratch with only some reference artwork being re-used if it is an edition of a game which already appeared on another games platform, or is based on a cartoon or movie or whatever.

The promise of J2ME’s write once, run anywhere may be realised with MIDP3.0 (I just made that up, in case you’re wondering) but for now, developers need to cope with a huge range of different handsets, meaning KVM bugs, memory limitations, screen sizes, file sizes, OEM-specific APIs, all of which differ from manufacturer to manufacturer, handset to handset, and even between different firmware revisions of the same handset.

However, even beyond that, games should still be more expensive than ringtones. Consumers are used to this distinction already, based on the situation in the physical world, where (generally speaking) games are around 3 times as expensive as a CD. Further, ringtones are highly disposable, whereas the well-designed games are not. In fact, I can virtually guarantee that you’ll get more enjoyment from even a not-so-great game than you will from a ringtone.

And why on earth should the operators be setting the prices? Isn’t it the job of of the game publisher to set the price according to what their research tells them?

It seems to me that a lot of the current thinking in the mobile content industry is a bit bass-ackwards…

Device Divergence

A few people have mentioned the recent Nokia investor day, where Nokia laid out their plans for the future (for links, see Russell Beattie’s post).

A few years back, I presented at a couple of investment conferences here in Scotland, long before people got interested in doing anything other than speaking and texting with their mobiles ;-). One of the key obstacles at that point was convincing investors, bankers, lawyers, et al that people would be doing these things with their phones, and they’d be doing them a lot sooner than they thought.

However, one of the slides that I originally used (but in the end didn’t) was one which laid out my vision of what would happen to wirelessly connected devices: wireless connectivity in itself would cease to be a selling point; it would just be there.

My idea was that wireless connections (GSM, GPRS, or whatever) would become as ubiquitous as the electric motor, and would find itself into a future GameBoy device, digital audio players, cameras, and other portable devices; and also onto fixed devices such as CD players and TVs. That’s now starting to happen, and with it will come new growth for the mobile device market. There are already 1 billion (give or take) wireless devices in the world (probably way more than that), but with divergence, that will easily grow to at least 3 or 4 billion by 2010. Using different levels of connectivity (eg, GSM in a CD player so that it can query CDDB) means that operators don’t have to throw existing, perfectly functional infrastructure out the window, and they can sell capacity to device manufacturers. So in my CD player example, the device manufacturer would take out a contract to use the network and pay the operator a flat fee or a fee based on usage. This cost could obviously be included in the original RRP of the product, because I don’t want to have different SIM cards and contracts for every piece of consume electronics around the house. Everyone wins.

Now, I’m not for a minute suggesting that I want to use my iPod as a phone, because then I’d look stupid. However, it would be cool if I could download music to my iPod (as I’ve written before), and now O2 have launched a music service that kind of lets you do that. Kind of. It’s going to require a lot of changes in the marketplace for this kind of thing, but that’s coming, believe me.

Anyway, the point is that convergence has never been, is not now, and never will be the focal point of product development. Divergence is where it’s at…

Nokia Looking To Buy Psion, Take Over Symbian

Just last Friday we pointed to research from ARCchart saying that the majority of Symbian sales would come via Nokia. Now, it’s sounding like Nokia is realizing this and giving up the whole “owned by multiple companies” aspect of Symbian. A few months ago, we reported that Motorola, one of the founding Symbian members was selling out and selling their shares to Nokia and Psion. Now, the rumor is that Nokia is taking the whole pie and is looking to buy Psion. Nokia currently owns 32% of Symbian compared to Psion’s 31%. Combined, they would have total control over the operating system, and could better position themselves to take on Microsoft – though, it would cut off just about any hope of getting Symbian traction on other manufacturers’ handsets.

Strangely, I suggested to Martin a couple of weeks back that perhaps Nokia should just buy Symbian. I disagree with the point about getting Symbian on other handsets; most Symbian handsets license Series 60 from Nokia anyway, as a quick, cheap way to get all the basic functionality you need from a phone before building their own unique features on top. This story from Reuters has more information.

New Nokia Launches

Nokia announced a whole bunch of things at their Mobile Internet Conference today: Nokia – NMIC 2003 - Launches.

A lot will be written about the devices elsewhere (especially the 7700), but what is notable for me is that Nokia are starting to categorise devices by names other than just “phone”, giving us “Messaging Device” and “Media Device” today.

As I wrote before, niche operators and devices are one way to re-ignite this sector, and Nokia are continuing to show the way here.

Do Mobile Services

Martin’s latest piece over at Mobitopia, Think Mobile Services provides some commentary on a piece in The Economist: Beyond the bubble.

Here’s the quote I’m going to pick on:

[Dave Dorman, CEO of AT&T says] the best prospects are at the network’s edges, not at its core, and revolve around providing complex services, not merely dumb capacity.

All well and good, but this quote still shows that the telcos see their “shiny new networks” at the heart of the business. They all seem to be saying the right things, but I still think the folks at the head of most of these organisations are too tech focused (something that NTT DoCoMo realised and fixed very early on).

Looking particularly at the wireless segment, some operators are claiming to be innovating on data services. Let me ask you a question: do you consider purchasing ringtones from your handset to be innovative?

Vodafone seem to be getting it most right with their live service, but the key difference with that is the integration of the live branding into (some of) the handsets.

One of the problems at the user-side is the sheer cost of some of the tariffs available, which are not geared to high data usage (unless you’re the operator). This is one space that is wide open for the application of some imagination, when combined with services.

Let’s imagine that Apple launch a new iPod, with some form of packet-switched connectivity integrated along with a mobile iTunes store (N.B., this device is not a phone, but the ability to send text messages might be nice). As things stand, operators would probably put this on a £15 pcm fee, which includes 1Mb of data and maybe some cheap texting rate, an extra £5 per 1Mb of data and be done with it. As a user, what the hell is 1Mb of music? Now, you and I might realise that it is about 3 songs, but users should not have to do calculations when they are trying to use the device.

Obviously, this kind of device should not be on a monthly tariff at all. Network operators will probably have some form of mild shock on a business model with non-recurring fees such as this, but given that pre-pay has been such a success, why is this any different? The best payment model for a device like that described above would of course be a certain amount per song, with a standard flat-rate text message rate.

Yes, my example is a different kind of device, but I believe that this kind of segmentation for the new devices (such as N-Gage) is inevitable, and operators either need to start creating a new set of tariffs to match the services, or, to open up their networks to start-ups who can further grow the market with a new set of devices and services. Phones are not necessarily the best devices to take full advantage of what people might want to do. Nokia are probably doing more than anyone to come up with new devices (like their cellular security camera), but the lack of imagination at the operators means that something like the N-Gage needs to be hamstrung by squishing everything into it at once, making for a non-great experience in anything.

Basically, I’m proposing that niche operators are one way to spark new life into the sector (witness the MTV deal)—whether they are run by the operators or by others is immaterial. Not so different from the cable TV market, I guess, where channels such as HBO and others have done so well.

So yes, think mobile services, but at the end of the day, you need to do mobile services. And do them well.