Creating ecosystems: Standardization vs. Innovation and lessons from the console gaming industry

cambrian explosion.jpg

This post is based on an insightful blog from Mike Masnick , CEO of Techdirt who I had the pleasure of knowing for a while and meeting finally in Washington DC when we both spoke the State of the Net conference . Like me, Mike often has a libertarian viewpoint , and his post Which Is More Important For Innovation: A Standard Platform Or Competition? is an insightful and balanced view of standardization for the video games industry.

The question being addressed is a well known one: Within the console gaming industry, is it better to standardise on one platform or many platforms (as we have today – ex Sony PSP, Nintendo Wii and Microsoft Xbox).

I explore the wider implications of this question to the Web and to standardization.

The article says:

It’s just a question of shifting the competition from being between platforms to being on top of a single platform. For example, it’s good to have competition in who can sell you lamps, but it wouldn’t be good to have competition among different types of electric systems with different outlets. So, we standardize on a single electric system, and it allows all the competition on electric devices on top of it.

The above sentence encapsulates the key issue of standardization i.e. the balance between ‘platforms’ vs. ‘innovation’. In other words, standardization and interoperability should concern itself with platforms and should not concern itself with innovation (which is supposed to be ABOVE the platform layer)

Which leads us to the logical question: What is a platform?

Now, here are some of my thoughts on the above discussion as I extrapolate these findings to standardization in general:

• Most people would argue that physical networks like ‘trains lines’ and ‘telecom networks’ are platforms i.e. should easily be interoperable. Does the logic of ‘interoperable physical networks’ apply to games platforms? Clearly value can be availed from the separate games platforms even when they are distinct because they are individually viable ecosystems in themselves. Both the platform owners and also the third party ecosystem (developers) can avail value without the need to interoperate between the platforms.

Art and artistic abilities also play a part in the platform argument. Games have an artistic component. Hence, an IPR component. Art cannot be standardised.

• Now, Does the logic of interoperable physical networks’ apply to the Web? By the ‘web’ we mean any nodes connect by the http and IP protocols. At that level, the standardization DOES apply since the same train lines / connectivity argument applies (value is created from connectivity). However, the interconnect and interoperability argument breaks down beyond the simple connectivity protocols like http and IP. This accounts for the importance of non-standard innovations in the Web – ex Flash (used by YouTube), the chrome browser(which has a different architecture from any Web browser and the woes of standardising HTML5 )

• Thus, the web when viewed as http + IP is a platform. But beyond these simple connectivity protocols, the Web encroaches on the domain of innovation. Chrome, HTML5, Flash do not show the failure of standardization process as they show the ‘non applicability’ of the standardization process to more complex domains (which are the realm of innovation).

If we take a step back and ask ourselves: What is the goal of standardization? It is to create an viable ecosystem.

‘Ecosystem’ includes third parties including (developers, entrepreneurs) and also society at large. ‘Viable’ involves money i.e. commercial viable.

In the IT and mobile domains, the ecosystem will include many elements including open source and microformats

Coming back to the gaming console arguments: The acid test of a viable ecosystem is: Can radical innovation still emerge from within the ecosystem? Including proprietary innovation?

The Nintendo Wii is an example of radical innovation and it DID emerge in the console gaming ecosystem i.e. there was nothing to prevent a Nintendo Wii (or indeed another new platform) from emerging.

Clay Shirky recently said “Communications tools don’t get socially interesting until they get technologically boring.”

Another way to look at this is: Things get technologically boring when they are standardised.

To conclude, being a fan of history, pre history and watching waay too much National Geographic: Standardization is all about creating a viable ecosystem and the diversity in ecosystem is more complex today than ever before.

There was a period in prehistory called the Cambrian explosion around 530 million years ago which saw the rapid appearance of most major groups of complex animals.

One of the theories for the Cambrian explosion is the rapid increase in the concentration of Oxygen in the atmosphere. We are seeing a similar ‘cambrian explosion’ in content, service and connectivity at the current time. It means we will probably see a much more diverse ecosystem which standardization will struggle with. This is not a failure of standardization but rather a healthy sign of a vibrant and viable ecosystem.

The question to ask is: How to preserve an economically vibrant ecosystem and at what level to standardise so as to preserve innovation

There is something to learn from both gaming consoles and the Cambrian explosion

Image source: http://www.biologyreference.com/images/biol_01_img0060.jpg

CNN’s new website

cnn1%2Cjpg.jpg

Since the launch of their new mobile site recently , CNN in London have been liaising with me and keeping me updated about their vision. They requested me to comment on their new Beta site – so here goes!

As a frequent traveller, I am very well versed with CNN and have been following it for more than ten years.

The service itself has gone through many changes – but this change is very significant and I can see why CNN is so serious about this site.

In my book Mobile Web 2.0, we went through a very long section outlining the change in traditional media due to the emergence of user generated content.

It represents a fundamental change in strategy – an evolution into a new world of user generated content complementing traditional media. In the UK, the BBC has been at the forefront of change in media. And to be fair, the BBC is excellent. However, it is funded by license fees! i.e. it does not strictly ‘earn’ its money – so to speak. Also, it has no advertising.

In that sense, the transformation of CNN in a Web (2.0) and a Mobile Web 2.0 world is a major step – and indeed one to be watched.

The changes are subtle but significant. The site itself is clean and easy to navigate. This is in keeping with the Web 2.0 philosophy.

There is an emphasis on Video in a big way – which I like

Traditional viewpoints are complemented by views from Blogosphere through Sphere

I found the advertising non intrusive. The clips play one after the other (just like in a news bulletin) and are sometimes (but not always) interspaced by advertising.

The launch itself included a beta launch blog which got many comments (I can’t find that link anymore – maybe it was removed post launch)

The new features include (adapted from a CNN presentation)

- Switch to horizontal navigation(as seen above i.e. showing read, video, photos and map as navigation tabs against a single story)

- Embedded video (and a heavy emphasis on video in general)

- Regionally synchronized CNN Int’l TV schedule

- Localised weather and news

- Permanent, stronger presence of blogs, podcasts, and I-Report links

- Re-launched site, redefines how a news story is told.

- Leads with the strongest storytelling element

• Article

• Video

• Photos

- Free On demand CNN Video and Free Live video with CNN Live

- In page video replaces traditional pop-up player

- Multiple navigation formats: Navigation by Category, Most Popular, Top Stories or by following editorial picks.

- Offers customisation with a My Playlist ‘self choice’ section

The two things I could not get to work are

a) Playing videos on my phone. The service said that I had an older version of Flash – and I could not see how that could be updated

b) I found that some videos were not available in the UK i.e. there is some geographic selection of the sites

Thus, I see this site as a broader, strategic shift by major media and it was nice to be able to work with CNN on this and contribute to this evolution.

A beta tour is HERE

Conclusions

a) A major shift

b) A greater emphasis on video

c) More Web 2.0 features – Blogs, user generated content etc

d) The design of the site lends well to mobile

e) Revenue model is advertising – since videos are free

f) Conflict with existing channels to be seen since videos on the Web are free. My take is: The two will complement

g) Some small teething troubles

Overall, it’s a good experience and one to watch because it is a pioneering change in my view

Experience migrating to wordpress from movable type?

Does anyone have any experience (or can recommend someone) who has migrated

to wordpress from movable type?

I think movable type is really not listening to its customers.

Simple things like comment management lead to a lot of work for example:

wordpress has features like entering a keyword v.s. movable type which forces you

to use a typekey service – not in my interest or the interest of my

readers!

A new version of Movable type is out but don’t see anything great in it – so now I am

seriously considering migrating to wordpress!

Please email me at ajit.jaokar at futuretext.com

kind rgds

Ajit

future of web apps event ..

Hello,

I was going to attend/speak at this event but cant make it. However, I think its worth a visit considering the star speaker list. Details are as below. If you are interested, please contact them directly through link below

The Future of Web Apps

Sep 13-14, San Francisco $295 (for two days)

Speakers: Kevin Rose (Digg), Mike Arrington (TechCrunch), Carl Sjogreen

(Google Calendar), Dick Hardt (Sxip), Evan Williams (Odeo), Jeff Veen

(Google/Measure Map), Mike Davidson (Newsvine), Steve Olechowski

(FeedBurner), Cal Henderson (Flickr), Tom Coates (Yahoo!), Matt Mullenweg

(WordPress), Tantek Celik (Technorati) and Ted Rheingold (Catster/Dogster)

My interview in Javamagazin(German)

javamagazin.JPG

The German language version of my interview at Software development magazine was published in Javamagazin(German). Thanks to Sven Haiges for both of these!

more on mobile VOIP

Hello ..

Firstly ..

Apologies for the delay. Have had a bit of a PC crash and still recovering. Meanwhile, the positing on mobile VOIP seems to have generated a great deal of discussion on the mobile applications club and has made it to the best of mobile apps club category.

Here is the original post and all the responses.

Contributors and their links include:

Original post: Ajit Jaokar CEO futuretext

Rupert Baines : VP Marketing picochip

Gerry winsor : Senior consultant HP

Tom Byrd : Mako analysis

Vladimir Dimitroff Director Prism consulting UK

Julian Bond voidstar

Alex de Carvalho tapio.com

Tomi Ahonen Tomi Ahonen consulting

William Volk Bonusmobile

Steve Kennedy eurotechnews

The disruptive potential of mobile wifi and dual mode handsets .

With the success of companies such as skype, VOIP is seen to be a ‘hot’ technology. Following the hype surrounding VOIP, ‘mobile VOIP’ is also caught up in the hype cycle.

The idea is simple – cellular calls could be made over an IP network -thus saving cost.

Currently, mobile VOIP is synonymous with ‘voice over WiFi’ i.e. voice calls made over a WiFi network. Although the WiFi network is growing fast, the world is far from being a 100% WiFi enabled space. This means, mobile VOIP suffers from the physical limitations of being near a WiFi hotspot.

The requirement of being near a WiFi hotspot plus the high cost of Mobile WiFi handsets, means that mobile WiFi is currently a niche technology. It’s initial deployment is expected to be in the enterprise or within hotspots.

The real potential of mobile VOIP lies in the use of dual mode handsets. Dual mode handsets support the seamless handover between a cellular(in practise 3G and beyond) network and a WiFi network. The technologies used in this space are currently being defined for example – Unlicensed Mobile Access (UMA) and the Mobile Integrated Go-to-Market Network IP Telephony Experience (MobileIGNITE) alliance.

Predictably, the incumbents such as mobile operators, are reluctant to support mobile VOIP because it’s a threat to their existing business(mobile voice calls). However, companies from outside the

existing value chain are keen to promote mobile VOIP. Most notably fixed line operators and handset manufacturers

However, the biggest barriers to the uptake of mobile VOIP is the pricing for IP traffic. I believe that the technology will really take off only when cheap, ‘unlimited use’ IP traffic becomes possible.

Thus, as with so many services in the mobile data industry, the barriers are not technological but commercial. There is no doubt that mobile VOIP will have a part to play in the evolution of mobility in general. Its eventual success and role will depend on a range of technical and commercial factors some of which are outlined above. However, its real significance lies in the fact that it will put a downward pressure on voice call prices (which is still the mainstay of income for mobile operators).

If cheap unlimited use bandwidth becomes a possibility then the market may well take off in other directions. For example – it could be possible to make voice calls from a 3G network through an IP client on the phone without going through a dual mode handset etc. The success of such schemes depends on low costs for IP traffic. However, note that the technology exists even today to make this possible.

Ironically, the mobile network itself is shifting to an IP core with technologies like IMS. When that happens, it should be possible to make end to end VOIP calls!

What’s your view? Is this balanced enough or have I missed something?

Who is paying? Rupert Baines

Ajit

I don’t mean to be rude, but this is quite simplistic: The thing you are missing is “Who Pays?”

There is no such thing as free lunch – nor a free phone call, nor a free byte over the airwaves.

Someone has to pay somewhere. The question is who (the user, some other user, the tax-payer – or some generous VC who pours money in…)

IP sessions (both wired & wireless) have historically been cheap for two reasons

- As best effort sessions they “filled gaps” between voice calls, increasing utilisation & efficiency.

- The voice business was the primary revenue source; covered all the costs. Data was “for free” and any money was a contribution to profit.

(the analogy is selling tickets on a plane to gap year students; better to sell them cheaop and get some money, rather than leaving it empty).

But just as BA would go bust if everyone on the plane paid the price of the gap year student, so a carrier will go bust if everyone switched to flat-fare IP (unless they change the business model to compensate)

(marginal cost vs average cost)

Most VoIP stories rely on cross-subsidy: it is easy to be cheap (free) if someone else pays the bill.

VoIP over fixed network makes sense (sort of…maybe… if you include some complex factors…)

But over the airwaves it looks really silly: the scarcest resource is air-interface capacity.

There are a finite number of Erlangs, of TXUs, or bits-per-second-per-Hz-per cell to use up. Using them more efficiently is the fundamental goal: ten years ago we switched from analog to GSM because it was three times better.

(we certainly didn’t switch for quality, or coverage or battery life – all were worse. We switched because it was cheaper for carriers)

The best reason for 3G is more voice capacity – look at 3.

Again, better efficiency leads to lower costs – which help pay for the license.

But VoIP is far *worse* for efficiency: There was an article on Unstrung which suggested it was four times worse (25% of efficiency).

In other words, it costs four times more to provide.

Now cost isn’t price: something could cost more and be sold for less. For a while.

If you ran a restaurant you might look at all-you-can-eat as a way to fill empty capacity (that quiet Monday night), or if, on average, people paid their way. But if you were to offer all-you-can-eat at peak time, and found that people consumed four times more food than usual… how long will you stay in business?

If BA were to fill all the seats on all their planes with gap year students paying below cost, and then said “fly as much as you want, for free” – how viable is that?

VoIP relies on someone else paying the bill.

If it uses four times more capacity than the equivalent voice call, you need quite a few poeople to subsidise it…

If you are the carrier who invested in spectrum, and infrastructure why should you accept that?

the barriers are not technological but commercial.

Well, yes. So are the barriers to everyone being given a free Ferrari….

If cheap unlimited use bandwidth becomes a possibility

… then lots of things become possible, yes.

If cheap unlimitted petrol becomes a possibility, if cheap unlimitted beer becomes a possibility…

But back in the real world, where things cost money and those costs needed to be covered if you are to stay in business…

I don’t mean to be sarcastic or rude, but too much discussion on this subject seems to be based on very naive thinking, or ignoring basic economics (marginal cost vs average cost). Some of it based on fallacious arguments (“VoIP is more efficient in bandwidth”)

There are interesting discussions, and some complex arguments, but lets get to the level of deatil to have them.

(For example, BT’s 21CN – but the economics on fixed are different, especially where you don’t need to pay spectrum license fees)

Mobile VoIP = power to subscribers? Gerry Winsor

Rupert makes some very valid points, i.e. “Who pays?” – however I view mobile VoIP, i.e. a “mobile Skype” like service as a potentially disruptive techology that will provide subscribers with a tool to apply pricing power to the current Operator managed voice-tariff structures. UMA and MobileIGNITE are Operator managed integration technologies and as such obviously allow the Operator to maintain tariffing control. And as Ajit points out, as IMS is rolled out an all IP overlay becomes available; and the 3GPP Rel’6 Specifications certainly include WiFi (UMA) integration connectivity, but full IMS roll-out/availability is still several months (years?) away.

Certainly it is easy to argue that the current mobile-IP pricing model is an inhibitor to mobile VoIP — without an “All you can eat” pricing structure mobile VoIP will not gain as much traction as it has done in the “fixed” domain, where a $29 monthly IP subscription makes VoIP an interesting alternative; but with Operators trying to boost ARPU by encouraging subscribers to consume more data-services at some stage the IP-traffic pricing models will have to come down to further stimulate uptake thereby providing an ‘in’ for mobile VoIP? I understand the technical inefficiencies that Rupert describes, but I would suggest that it is the IP/data-service pricing model that will have a bigger impact.

These are definitely interesting times – that’s what I love about telecoms today!

thanks rupert/Jerry Ajit Jaokar

good thoughts as usual. shall respond more after reading them rgds Ajit

from tom byrd – mako analysis

as per link

Hello!? I’m on the WLAN – 03 July 2005

With every major mobile handset manufacturer producing several WLAN enabled devices in 2006, will mobile voice calls over WLAN become a reality for the end user and a represent a new threat to carrier’s revenues?

How does it work?

For the end user this will principally allow them route calls in a similar fashion to BT’s Fusion proposition (previously known as BT Bluephone). The theory is that an end user with a WLAN enabled mobile phone would be able to conduct calls free of charge via their mobile phones with the call being transported via VoIP through the internet. When the user moves out of WLAN coverage the call would seamlessly transfer to the wide area mobility network (GSM/CDMA etc) where it would be charged for on a per minute basis in the usual fashion.

So how realistic is this scenario, who are the players and what do they have to gain or lose? Mako Analysis has taken a look at the player’s involved and how priorities for this forthcoming capability will compliment and conflict.

Operator beware

Given that between 20-50% of mobile carriers minutes originate in the home or office environment any calls that could be rerouted over WLAN would represent a significant threat to their primary revenue source of transporting mobile voice minutes. Given the growing success story of fixed VoIP players such as Skype and Vonage and the minutes that these companies are detracting from traditional fixed line voice carriers, mobile operators are well aware of the interest mobile VoIP may generate.

Demonstrating the possible

The companies that are making this technically possible, most notably Canadian wireless equipment vendor Nortel of late, are certainly keen to push the development of this technology forward. Nortel were recently the first to demonstrate the live handover of a call from UMTS to WLAN, opening the public’s eyes to what may be widely achievable with their humble mobile phone in a few short years.

Device players keen

Handset manufacturers are also keen to drive this type of functionality into the market in an effort to continually add fuel to the end consumer’s desire to upgrade their handsets. Device replacement cycles are a significant metric of relative success for the mobile device market around the world, hence making WLAN integration a key strategy for the future. Mako Analysis has learned of at least two major handset manufacturers that are due to release at lease five WLAN enabled devices in 2006.

Of the three main parts of the mobile voice over WLAN value chain it seems only the mobile operator is approaching the situation with (well justified) hesitation. However we feel the true measure of its success will not sit with any of these three but with the end customer. As such it would be prudent to understand what the customer draw of this type of proposition is and will it adequately fit that need?

Customer dictates the demand

Its clear that the primary draw of mobile voice over WLAN is its cost advantages. If cost is the main driver, we have to consider that the home user will need to invest in broadband to the home, a personal WLAN set up and a mobile device with integrated WLAN which will be for the near future an item of particular high dollar value particularly in countries where handsets are unsubsidised. Many likely users will have a home broadband connection already and a portion of these may have a home WLAN set-up making the addition of a WLAN mobile less of an investment cost decision. The user will have to make some kind of fairly significant set up investment however and given the question marks over mobile voice over WLAN quality of service, the decision to plough ahead with this set up is less of a “no brainier” than originally planned.

For the business user set up it is clear that the vast majority of organisations will already have broadband to the office and a significant proportion of these will have WLAN already installed on site. The decision in the business market therefore rests mainly on the cost benefits of equipping its workforce with high end WLAN phones in order to gain access to the cheaper call promise of mobile voice over WLAN. In the majority of cases business source cheap low end devices for the majority of their workforce and even shy away from devices that have additional features such as a camera for fear of fuelling an increase in their average bill spend. A WLAN mobile will indeed be an expensive piece of hardware for some time to come and will feature many other multimedia features such as cameras, music and video players etc. When this is coupled with mobile voice over WLAN QOS question marks, its commercial application is far from keenly anticipated at present.

When we couple these customer obstacles with a high degree of uncertainty and hesitation from the operator community, we feel that, voice over WLAN is something that can be put on the back burner for a few years yet.

Someone must pay – really?! Vladimir Dimitroff

What a shocking discovery, indeed! And we all thought lunch was free

Intelligent and insightful, Rupert’s posting goes too far in challenging VoIP ‘wisdom’, so far that it represents the other extreme.

I don’t think this forum’s readers (even less – Ajit) are confused or dumb to be taught economics fundamentals. In fact, Ajit never said free bandwidth. He said cheap which, I think, is possible within the relativity of the concept (what is cheap) – and shouldnt’ be laughed at. Witty comparisons with petrol and beer only distract from the debate. Both beer and petrol, BTW, are available for unlimited use (anyone can buy as much as they like and any variety they prefer) as opposed to restrictive operator practices in respect of data.

While the efficiency of VoIP is debatable, the repeated statement of ’4 times’ (based on a single article and taken out of context) exaggerates just as unprofessionally in the opposite direction of VoIP blind fans.

Finally, the airline examples, I am afraid, only prove the opposite of Rupert’s case: yes, someone does pay – but for what? Why should the economy flyer pay for the wasteful (and ineffective) marketing costs of persuading (read: bull$#!tting) ‘full fare’ passengers that a breakfast costs £100? Or that 2inches extra space and a marginally better meal cost 300% (!) more, often the difference to first class price? Aren’t the EasyJets and RyanAirs of this world exactly a proof that cheap IS possible? (remember: they are NOT undercutting prices at a loss, they have a profit margin in their £10 ticket).

Back to our mobile industry: the equivalent of EasyJet is already happening (and it’s not EasyMobile, unfortunately – they got it wrong. But there are several low-cost operators inEurope and a bigger number in Asia that compete with bottom prices AND are profitable). Why should the low-usage (or even average) subscriber pay for the crazy acquisition subsidies and exorbitant marketing budgets, even more wasteful and inefficient than those of airlines? Who pays, really??!!! For the acquisition costs, for the self-induced churn (by encouraging promiscuity with subsidies). Why does a porminent 3G operator give me a free handset AND 500 free minutes of voice, and then hopes to charge me their egregious prices for data and multimedia content? Could it be that I actually pay for my voice minutes, and even handset, but my data traffic is cheap and not limited to their walled gardens?

The real world is not black and white, there are not only many shades of grey in-between, but also millions of colours. I would say reasonably priced bandwidth is both conductive to, and makes the most from the disruptive potential of VoWiFi (lest we forgot our topic). The invisible hand of Mr Market will eventually bring reason to the game, as clear signs already show. Dual handsets are not accidental or miscalculated adventures of their makers, they reflect strategic thinking. Similarly, BT’s 21CN shouldn’t necessarily be limited to fixed line.

Watch this space: a certain major operator (I can only disclose their name starts with “T” and ends with “Mobile”) have just declared that “they are proud to be a bitpipe”, and are offering unrestricted Internet access and a flat rate of €10 for data over UMTS and GPRS (so far in Germany, coming to the UK later this year – did oyu know this, Ajit?). Perhaps they have asked themselves the billion-dollar question “Who Pays?” – for marketing, billing, CRM, subsidies, loyalty programmes, portals… and by focusing, as good German engineers, on the core competence of Network Operations, they can save costs and pass the saving to MVNOs and end users – and make more money in the process… If they go that way, my take is, they will have followers.

Took too many words to say what I wanted, and it was: this is not about free lunch, but not about expensive lunch, either – it’s all about optimisation.

Cheerz,

V.

Knowledge: the only product which, once sold, remains with the seller. Give it away!

VOWifi Julian Bond

http://metrofreefi.com/ US Only (boo!). Lists free wifi hotspots. A PDA with Wifi and Skype would be free in those locations. Which is partly why the various Municipal Free Wifi initiatives are attracting so much bad feeling from the Telcos.

A long way off Alex de Carvalho

Further to Julian’s post, Skype for pocket PC has existed for a while and Skype has promised a version for Symbian and embedded Linux mobiles. So the VOIP client can reside on a dual mode smartphone as well as on a pda (see the upcoming Qtek 9100). Assuming the software can handle a handover from wi-fi to 3G (and I don’t see why not), the only barrier for mass adoption and usage is indeed commercial.

For instance, my operator (one of the top French telcos) is offering the first 5,000 Laptop subscribers 8h of 3G per month for 27 Euros per month (+47 Euros for the laptop 3G connect card). This offer has no call minutes. They’re also offering the first 30,000 mobile subscribers 4 hours per month 3G access (and 4h daytime calling time) for 49 Euros per month.

So prices are high and we’ll have to wait for increased consumer demand and lower smartphone handset prices before we see operators (or MVNOs) offering cheaper and unlimited IP traffic.

And with cheap IP, you can get presence and IM …

Alex

tapio.com

Vacation musings of a wireless nature Tomi ahonen

Hi gang

What a nice little debate we have here. Good postings all around. I am enjoying a nice cup of cappuccino with my 3G modem-equipped laptop and enjoying the sunset, decided to wade into this discussion…

I would take it that all who have commented would agree that for the next 5-10 years both will co-exist. We will have WiFi (and evolution) based VoIP phone devices, AND we will have cellular telecoms devices. In all countries, both will co-exist. Just like today you can read books and magazines totally for free – at a library – yet the majority of books and magazines are bought for, and read in the home. Both business models have existed for over a century and still happily co-exist all around the world.

So there is no point in arguing WiFi/VoIP on mobiles/portables will come – of course it will come – but equally no point in claiming Cellular is in any way dead bacause of it.

Lets put the big picture in context.

WiFi phones and VoIP systems (Skype, Vonage etc) have already well in excess of 45 million users worldwide. That is a very significant user base and remarkable growth for the past two years. And already about 3 million of those users pay for their VoIP service (eg Skype with its numbering services or receiving calls under certain situations etc). So its no longer all “free”.

Who and where will we find obvious major user groups for VoIP? The biggest groups to take WiFi type phones in use and put a lot of traffic on them are the various campus systems, eg corporations replacing their PBXs at their offices etc. Also homes with (fixed/wireline) broadband internet are obvious early targets and those are rapidly getting their WiFi handsets already.

On a very rough level, if there are 1.3 billion fixed line phones in the world, and roughly 30% of all fixed line phones are with business customers. If we assume, for the sake of argument, that all of those business lines are transferred to WiFi or wireless VoIP phones during the next decade, we’d see about 390 million business WiFi phone users. Quite an attractive number, when contrasted for example with 20 million iPod users (the current darling of the IT media). I can see the business case to be appealing to any corporate CFO considering a telecoms equipment upgrade and that essentially all PBXs will be replaced by WiFi/VoIP systems over the next decade. Probably also the various smaller business telecoms systems like Centrex etc. What is the replacement cycle of business telecoms systems, probably something like 7 years or so.

At the home its a bit of a different story. If you have (fixed wireline) broadband, then WiFi is an easy logical upgrade, especially for those with a tech-nerdy bent. But will all current fixed interent users migrate to broadband. I would say no, there are many who have internet for some given app that is not brandwidth-hungry – I am thinking of my own parents who use the internet only for their banking. Then the question starts to be more one of comparing alternatives. I would argue that the total populations of WiFi phones of all kinds, will be of several orders of magnitude less than sales of mobile phones (remember we’re selling 700 million mobile phones this year alone) and these are replaced every 24 months worldwide, and much more rapidly in many heated markets.

The more alarming trend is those homes where there is no fixed line to begin with. Already 40% of all homes in Finland have abandoned the fixed line totally. If you want these to use broadband, it will be 3G broadband. Now, there are limits of “fair use” to any 3G usage, and suddenly the “free” nature of VoIP is totally demolished. No sense in me using “free” Skype on this 3G connected laptop, because I have a limit to how much Vodafone lets me use 3G capacity in any given month.

So while there are obvious markets where WiFi phones will dominate (eg business), there are others where a VoIP based mobile phone is pretty well useless – imagine a taxi driver whose home phone and internet connections are all on the cellular network.

So that leaves the middle ground. In very many cases there is then the issue of costs and benefits. Here VoIP will have a hard time to win over the addicted “Generation-C” mobile phone addicts. Yes, an adult father, bread-winner, head of the household, will try to cut family costs by asking the wife and kids to move their telecoms traffic to the family WiFi phone. I can hear it. Yet you know that inspite of this, as each of the family members has their own mobile phones, there will be significant traffic on the cellular network.

There are two interesting stats that any VoIP-believer should keep in mind. People on the whole are dumb, lazy, and irrational. Look at e-mail vs SMS text messaging. In every market – yes now even emerging in the USA – when you give young people access to both. Where e-mail is always free and SMS text messaging costs per message every time – still the clear preference in every market is that young people prefer SMS text messaging as was first shown by the widely-quoted Siemens study of 2003.

And then there is expensive vs low-cost calls. TIM in Italy was widely quoted two years ago saying 70% of all mobile phone calls on their network originated from indoors. Where the mobile phone user was near a fixed-line phone with every time cheaper calls than those on the mobile phone. Yet we are lazy. If you call me, I see you on my mobile phone. I can hit one button and return the call. I don’t think is it 12 cents or 8 cents or 4 cents on what network. I just want to return the call. I’m not going to bother to dig out the number and start to dial digit-by-digit on a fixed line phone, even if I am right next to one.

As the “threat” from VoIP will grow, so too will the mobile operators’ responses in their pricing etc. The price differential will diminish over time, not increase. Remember already that 8% of all VoIP users are PAYING to use their “free” service, so the free gets to be less free, while the expensive is getting to be more reasonably priced. Witness price war effects in Hong Kong, Denmark and Finland to see how reasonable voice minutes can become in more mature telecoms markets.

There are about 750 million PCs connected to the internet via a fixed wireline (narrowband or broadband) connection. Will everyone of these add a WiFi phone to the PC connection. I severely doubt it, but many, perhaps hundreds of millions will. Already over 150 million people connect to the internet only via a cellular network (increasingly 3G). None of these are candidates for WiFi VoIP solutions, they will not go “backwards” on their evolution.

There are 1,300 million fixed line phones. Like I said, we can assume most business users will migrate (I doubt it all, but for the sake of argument I’ll give you over 90% of this population, meaning well over 350 million). Of the homes, remember we have the broaband internet people already in this group, and those internet users listed above, so if we get a total of half of these, I’d be very impressed. Maybe we get to 600-700 million users.

Note that is almost as many as there are internet connected PCs, and over ten times the number of PDA users worldwide. To achieve this level of success, the WiFi phone sales have to DOUBLE every year for the next six years, year-on-year, to achieve this (a most dramatic growth rate, something that has never been done in human history at anything approaching these scales). If WiFi phone makers achieve this kind of sales/success/growth, its a huge success for the IT and telecoms industry, making many many millionaires in the process.

Yet keep it all in context. By September of this year we will have 2 billion cellular phones in the world. Over 30% of the human population, and practically every economically viable person on the planet carries a mobile phone. Every one of those 45 million WiFi/VoIP/Skype/Vonage phone users also has a mobile phone – many carry two or more.

The global love affair with the mobile phone is not about to disappear with this latest tech toy. Yes, if I was a business development manager or competition analysis expert with an operator, I’d keep an eye on VoIP and WiFi, but I wouldn’t lose any sleep over it.. But if I was in the fixed wireline telco business, or was an ISP, then this was the big opportunity right now.

But hey, I’m on vacation so maybe I’m on a special mood ha-ha…

Dominate !!

Tomi Ahonen / HatRat :-)

NOT Dumb at all…. William volk

Regarding:

There are two interesting stats that any VoIP-believer should keep in mind. People on the whole are dumb, lazy, and irrational. Look at e-mail vs SMS text messaging. In every market – yes now even emerging in the USA – when you give young people access to both. Where e-mail is always free and SMS text messaging costs per message every time – still the clear preference in every market is that young people prefer SMS text messaging as was first shown by the widely-quoted Siemens study of 2003.

It’s the PUSH nature of SMS vs. “standard POP/SMTP EMAIL” that makes it preferable. No need to ‘check’ your messages. You know when you got one. This preference isn’t ‘dumb’ at all.

Of course this is EXACTLY how mobile email works in Japan, hence the popularity of Mobile-Email in Japan.

William Volk

CEO, Bonus Mobile Entertainment

Rudetones(tm), Guaranteed to Offend(tm)

@ Tomi Alex De Carvalho

This is a great discussion! Just to continue on a couple of Tomi’s points:

• We will have WiFi (and evolution) based VoIP phone devices, AND we will have cellular telecoms devices.

Indeed, a wifi device (PDA, laptop, game console) can become a VoIP handset as long as you can install the software and as long as there’s audio I/O. By WiFi phone I assume you mean a dual mode handset.

So mobile VoIP can be used just through a 1) WiFi enabled device, 2) through a dual mode handset or 3) just through cellular broadband. Number 2 is an issue of dual mode handset sales as well as the availability of cheap unlimited cellular broadband. Number 3 is an issue of smartphone sales and cheap unlimited cellular broadband.

Unless the change is imposed (by the operator), some people may never adopt a WiFi device, a dual mode handset or a smartphone.

• Witness price war effects in Hong Kong, Denmark and Finland to see how reasonable voice minutes can become in more mature telecoms markets.

Lower priced voice minutes are a great benefit to consumers.

“Presence” and IM can be demand drivers in themselves. With unlimited cellular broadband you might leave your mobile IM on all day (real IM, not text messaging). Rather than calling someone, you could IM them first, as is the convention on Skype. With presence, you don’t have to guess whether people are available to communicate or not. IM is superior to SMS texting and VoIP is superior regular phone calls when there’s a presence server.

• Already over 150 million people connect to the internet only via a cellular network (increasingly 3G). None of these are candidates for WiFi VoIP solutions, they will not go “backwards” on their evolution.

I believe the tradeoff between WiFi and cellular broadband is a question of bandwith, cost and availability.

• Note that is almost as many as there are internet connected PCs, and over ten times the number of PDA users worldwide. To achieve this level of success, the WiFi phone sales have to DOUBLE every year for the next six years, year-on-year, to achieve this (a most dramatic growth rate, something that has never been done in human history at anything approaching these scales). If WiFi phone makers achieve this kind of sales/success/growth, its a huge success for the IT and telecoms industry, making many many millionaires in the process.

Indeed growth is not linear and will follow more of a J-curve as economies of scale kick in. Linear growth for dual mode handsets would occur if the change was imposed by operators during handset replacement.

• Every one of those 45 million WiFi/VoIP/Skype/Vonage phone users also has a mobile phone – many carry two or more.

The 45 million users have mobile phones, but in many cases will prefer to make a call (particularly an international call) on IP rather than cellular networks.

• The global love affair with the mobile phone is not about to disappear with this latest tech toy.

I agree the love affair is not over / will never end… but I “love” my 6680 more than any “regular” mobile phone, for all of its features (the list is long). In fact, sometimes I switch it off the cellular network and just use it as an MP3 player, camera, game console, audio recorder, pda … My only regret is that it doesn’t have WiFi!

• Yes, if I was a business development manager or competition analysis expert with an operator, I’d keep an eye on VoIP and WiFi, but I wouldn’t lose any sleep over it.. But if I was in the fixed wireline telco business, or was an ISP, then this was the big opportunity right now.

Although ISPs have the ressources, marketing, billing, customer service, clients, technology etc. to be an MVNO, I believe dual mode (and smarphone) handset prices are too high for them to create an economically viable business plan for just mobile VoIP. And consumer demand for mobile VoIP is too low because you can defer your VoIP calls until when you’re at your connected PC. So growth for mobile VoIP will depend on consumer demand, which will depend on the availability of other great mobile IP services as well, on the whole.

• But hey, I’m on vacation so maybe I’m on a special mood ha-ha…

Enjoy your vacation!

Alex

tapio.com

VoIP and mobile Steve Kennedy

There’s a big difference between mobile and fixed networks. Fixed networks are moving to VoIP as it’s considerably cheaper to build and run packet networks than TDM networks (about 40% of the cost), for a company like BT that ammounts to lots of savings.

It’s also true that VoIP adds considerable overhead to codecs i.e. GSM is 13Kb/s on a TSM based system, but when encapsulating in IP there’s at least 20% overhead, probably more realistically 25% in terms of bandwidth requirements. TDM is also synchronous while packet networks tend to be asynchronous, meaning a 2Mb/s leased line will handle 30 64Kb/s voice calls, however translate that into packet terms and you need 2Mb/s each way (assuming no packet overhead) which isn’t so easy on ADSL etc or even over the public Internet. Then there’s QoS, yes things like Skype work well most of the time, but there are times when the call quality is terrible. Would you risk your business telephony on a service with no QoS.

You can run QoS on IP networks, but that adds costs, and suddenly VoIP isn’t so attractive. Single sites looking at VoIP are probably going to get a better deal from traditional telephony services using IDA or CPS etc. Multi-site organsiations should be able to make savings if they’re paying for intersite calls (or even by standardising on just a single packet infrastructure).

Once you move into the mobile space, the problems are the same. You also have to worry about data tarrifs, as all you’re doing is changing from using the operators call tarrifs to their data tarrifs. Currently (in the UK at least) they aren’t brilliant and it’s likely to cost more for a VoIP call (well after you’ve used your inclusive MB or whatever). Skype uses ilbc which is an 8Kb/s codec, which after encoding is probably 10Kb/s+, very quickly eats your monthly allowance.

Not all operators allow IP access outside their walled gardens (3 is a good example) so running VoIP is a no go in the first place, but then they are buying market share by offering silly voice tarrifs.

Steve Kennedy

Skype on PSP? William Volk

I wonder if anyone has ported Skype to the Sony Playstation Portable?

William – what are Rudetones? Post a new thread please Tomi ahonen

Hi William

Good posting and I agree yes, push vs pull etc. I’ll comment in a moment but William – whats this on your signature? Rudetones, guaranteed to offend? Tell us more!!!

Please post a new thread so it won’t get lost in this discussion, but please tell us. I’m intrigued. If its anything like I’m thinking, I believe I’m addicted already…

PS nice to see you’re also speaking at the 3G event by Informa in Brussels :-)

Dominate

Tomi Ahonen / HatRat :-)

Quick comments of essentially agreement (sorry…) Tomi ahonen

Ok, read thru the replies and a few comments

William – I agree about the push/pull nature of SMS, e-mail, mobile e-mail (like Blackberry and e-mail in Japan). An interesting note – what if both are enabled on your phone and the culture uses both. Japan is an exception in a way, that they got mobile e-mail so early, that an SMS/text messaging/short messaging culture didn’t have time to catch hold of the younger generations. Consider Korea and think of this to yourself, and especially younger people around you. In Korea the young professionals of today say that they use SMS text messaging for all messaging communication that it can practically be used for (obviously not in writing multiple-page documents) with all contacts business and personal, who are of the same age or younger. They say that only the older generation – like their bosses – use e-mail.

An interesting twist. Not so much that the speed is the same, and that everybody has the device. But a culture-preference shift. Suddenly if you use mobile e-mail you are “an old fogie” and to show you are hip with the young, you need to use their method. Sound awefully similar to the thinking of the teenagers here in Britain, and/or the youngest of your team members etc? (Sorry, William, you’ll need to imagine this a lot harder there across the oond than we do in Europe ha-ha…)

Regardless, don’t want to twist our discussion here from the juicy one on VoIP to another potentially never-ending one on SMS.

Alex/Tapio – I agree with just about all you wrote as well (no surprise ha-ha). A few tidibits of my thoughtlets now again with a vacation-mindset..

About presence and mobile phones. I’ve been thinking about this for many years and I am increasingly thinking it is another technology solution in search of a practical customer need. First, I totally agree that presence will greatly help those who are heavy into IM, or using Skype etc. So it is a good solution for those “network age” technologies like the accessing the (fixed/wireline/narrowband/broadband/traditional) internet, voicemail, e-mail etc. But if you are on a “connected age” system, ie mobile phone which is always with you (every addicted mobile phone user has his phone either upon his/her person or within arm’s reach at all hours when awake, and the famous BDDO study from this spring said 60% of us take the phone to bed with us) and which is always on – we dont’ even turn it off – ha-ha, Alex you are probably an “inbound-overloaded” exception who turns off his phone at times – but rather turn the ringing sounds off at times and still take in the messages and logs of attempted calls..

So presence is an excellent “crutch” for old-fashioned networks to try to replicate the connectedness of the mobile networks. But for us in mobile (cellular) telecoms, the addition of presence info is perhaps a brilliant bit of technology utterly without practical uses… We’ll see, am still chewing on that item.

But I think you’re right Alex, presence will greatly enhance IM and Skype

The 45 million users have mobile phones, but in many cases will prefer to make a call (particularly an international call) on IP rather than cellular networks.

Here I come back to my earlier argument about people being lazy and irrational. Yes, first let me agree, that for many – but definitely not all – of those 45 million who have Skype or Vonage or similar solutions in use, they will often prefer to use it especially for expensive calls. Many who were among the first to get this technology have a strong need for it – very many long-distance or international calls, or for example being extremely price-sensitive, say a struggling small business or a college student etc. So these people have crossed those hurdles of learning to use the system etc.

Still, remember the Italy TIM study. 70% of all mobile phone calls are placed from indoors, essentially almost all are thus within easy access to a fixed wireline phone. Remember that Italy is one of those rare countries (like Finland and Korea) where handset subsidies are illegal. So it is not a case of a huge bucket of call minutes on lengthy monthly pre-pay contracts to subsidise the costs of handsets. No, Italy is one of the leading pre-pay markets and users pay going rates for every minute.. Yet being next to a cheaper fixed wireline phone, they still place those more expensive calls on the mobile phones.

I do believe it is the majority, mass-market, pattern for behaviour. It is too easy, lazy, to call with the mobile. Again, the early users or Skype were either very heavily in need of the savings, or else are very “engineering-type” nerdy PC people. Engineers (and people with that kind of mind-set) are not like regular people. Engineers want to always optimise and maximise etc. “Normal” people ie the mass market couldn’t bother to do a spreadsheet analysis of the best options. They go for what looks nice or is the right colour or has the appropriate feature or happens to be on sale that month or whatever.

So I’d say for parts of the existing 45 million users we will continue to see the preference of shifting the expensive telecoms traffic to VoIP etc. But also, as the price differentials come down – due to competition – this “incentive” will diminish. If the price differential is not 5x more expensive or 2x more expensive, but only 20-50% more expensive, how many will bother.

On 6680 – yes me too. I find taking it out of my pocket so much more than any recent phone or device. I got myself the 512 MB memory chip and now have a huge collection of music, music videos, film/TV clips etc.

Oh, that reminds me – in Finland they had a cool ad for the public transportation mobile payment for the Helsinki underground etc. They said the reason you want to pay with your mobile is to give you another excuse to show off your latest cool phone… I can kinda relate to that (wait, Tomi Ahonen an egomaniac show-off? what a revellation!!)

Anyway – for the Wifi you of course then have to carry the 9500 as the other phone… (oh, totally annoying, now with the high power of the 6680, even though the 9500 has the wider screen, any film clips will actually play better – with less dropped frames, less jitter and better sound synch – on the 6680 than on the 9500. I had thought I’d use the 6680 as my music player and the 9500 (with the 256 MB memory chip) as my videoplayer, but the videos are visibly better on the 6680… But I guess that is the price of rapid development isn’t it..

Steve – again I totally agree. This brings me back to what was earlier in the thread about airlines and extra capacity. Yes, fixed line telcos tend to have huge excess capacity in the basic network, and any ways to fill up that at the least additional revenues will make sense. But mobile operators are working on a congested space with a scarce resource (the spectrum) and since demand seems to be exceeding supply at every generation of mobile from 1G to what we now project for 3.G and 4G, there is a strong reason to charge a premium for that imbalance in supply/demand.

Sorry guys I couldn’t stir up more of a controversy this time ha-ha… :-)

Dominate

Tomi Ahonen / HatRat :-)

One more thing I hope you’ll agree with – Vladimir Dimitroff

Tomi, you wrote:

To achieve this level of success, the WiFi phone sales have to DOUBLE every year for the next six years, year-on-year, to achieve this (a most dramatic growth rate, something that has never been done in human history at anything approaching these scales). If WiFi phone makers achieve this kind of sales/success/growth, its a huge success for the IT and telecoms industry, making many many millionaires in the process.

Not a prblem at all 3G phones are doing it, as did GPRS ones before, Bluetooth ones, colour screen and camera phones (in a most most spectacular way). The VoWiFi or VoAnything phone is not a competing new device, it’s still a mobile phone – with one added feature, optional (!) VoIP capability.

It is as easy for the handset makers as adding the previous ‘novelty’ features – perhaps easier now, with SDR (software defined radio). Watch this space! (You’re into trends, Tomi) With the kind of modern DSP chips already in current handsets it becomes only a matter of software to make a ‘chameleon’ device that can switch instantly to a different frequency band, modulation algorithm, transmission standard, data protocol etc.etc.

I won’t be surprised if some of the hacker communities (you’re into communities too?) are already producing downloadable Java applets that can do this with some of the latest handset models.

Just like the walled gardents of 3G and limited traffic your Vodfone allows you, operators will want to control this, makers will gladly oblige and there will be many a handset model with this feature deliberately disabled – but I still believe in the ‘hidden hand’ and you cannot keep this under a pot for long.

Ajit’s use of the word disruptive was somewhat misleading in the thread title – we are talking convergence here, not substitution – and multi-mode, SDR-based devices are likely to play a major role in such convergence. Although less qualified to extend this to the infrastructure side, I can imagine similar developments will enable multi-mode base stations and network operations – largely removing bandwidth challenges and leading to optimisation of capacity, QoS, and most of all, cost.

Just my ignorant opinion, gurus know better -

Cheerz,

Vlado

Knowledge: the only product which, once sold, remains with the seller. Give it away!

Dual-mode chips Alex de Carvalho

Talking about chips, Quorum Systems has developed the Sereno QS2000, which combines cellular (GSM/GPRS/Full EDGE) and WLAN (802.11b/g), on a single chip (Check out their powerpoint presentation):

“This offers smaller footprint, lower manufacturing costs and unprecedented flexibility compared to today’s multiple chip approach.”

Inexpensive, entry-level dual-mode handhelds + voice and multimedia over WLAN = mass consumer adoption. Will carriers play this game?

Alex

tapio.com

Good viewpoint Vlado.. Tomi Ahonen

Good point Vlado. Ok, gotta admit that made sense (ha-ha)…

So it does come down to the handset manufacturers. If they start to include this ability, yes then it goes in the natural replacement cycle. But then it very significantly comes down to the (major) mobile operators/carriers wanting to sell it. Remember what happened with Nokia’s N-Gage. Launched at Nokia’s absolute top market share prowess, at almost 40%, Nokia if any could have rammed this “obvious” youth-gadget down the throats of the mobile operators. Didn’t happen. They would not support it, not subsidise it, and – together with many other Nokia marketing blunders and N-Gage design faults – the original concept flopped.

The critical point is, that all mobile operators know this story. Today if they see a disruptive technology being proposed by their handset vendors, they will just say “N-Gage” and the manufacturer will shy away from that project.

The commitment and support of the mobile operator/wireless carrier is vital for this adoption. And I’d add, that those most likely to consider this avenue would be the smallest players in any given market, and definitely not the major global brands and their alliances. So it will be a tough sell.

In the short run at least the operators will strongly resist this… :-)

Dominate – or sunbathe – whichever is your preference :-)

Tomi Ahonen / HatRat :-)

Such an interesting topic we just can’t stop – Vladimir Dimitroff

Absolutely agree about the crucial role of operators – and I expect many to want this feature disabled in their branded handsets. (Just like the pretty advanced LG phones provided by 3 (Hutchinson) have no Bluetooth and ‘cannot’ transmit pictures via IR because the operator doesn’t profit from direct transfers.

It also makes sense that smaller, niche operators may break this trend. But some ‘early bird’ developments suggest bigger players may have their own reasons:

- If T-Mobile put their money where their mouth is (with the flat €10 unlimited data tariff and the ‘we want to be a bitpipe’ mantra), they will welcome multi-mode handsets

- O2 with their Fusion have done it, albeit not (yet) on WiFi – but just watch this space

- if Motorola are better than Nokia (or Sony) in anything, it’s not stylish design or game animation – but they have an unbeatable pedigree in radio and deep vaults full of patents and yet-to-be-disclosed technology. Moto have been supplying SDR to the military long before it came as a buzzword to the mobile telcos. Coming with the earliest dual-mode handsets they are not targeting the small niche operators, believe me – those are not Motorola’s raison d’etre.

I expect some, if not all, of the major carriers to embrace dual- and multi-mode, and to profit from it.

Any bets?

Dominating in a 4 x 100 race (grams of vodka, that is) and moon-bathing -

Cheerz,

Vlado

Knowledge: the only product which, once sold, remains with the seller. Give it away!

BT Fusion Gerry winsor

As Vladimir indicates Moto have deep knowledge in this space and hence it wasn’t surprising when BT released Fusion — Phase 1: GSM + Bluetooth using a Moto phone; Phase 2: GSM + Bluetooth + WiFi due Nov. ’05? by then BT should have 2 or 3 phones available – and maybe even a Nokia variant!

My feeling is that the Fixed Operators (i.e. Fusion-like services) that will, through an MVNO relationship drive this … it will be interesting to see the BT Fusion sign-up rates sometime soon.

Ha ha, more mobile than mobile.. Tomi Ahonen

Hmmm, thinking “outloud” – this means more mobile than mobile?

I can see it coming. Through MVNO arrangements a fixed/wireline operator/carrier can give “full” mobile network services but then bundling on their WiFi and VoIP, they can be even “more mobile” than those services offered by the mobile operators. For example airplanes, we’ll get WiFi deployed on airplanes much faster than the picocells for cellular phones. A fixed operator to offer more mobility than a mobile operator? I’d like to see a player really push this, but most of those who have the means – ie BT, AT&T etc – seem to me to be way too burdened with bureaucracy to do it.. Maybe a player like Tele2?

Dominate

Tomi Ahonen / HatRat :-)

International roaming Julian Bond

Perhaps the killer app is International roaming. I’ve just been stung by the extortionate rates on mobiles in Europe as I try and call my daughter in France on her birthday. Both of us end up paying something completely absurd like £1.49 a minute because they charge her for incoming calls as well as me for outgoing. Now imagine that she’s got a dual mode VOIP device and is on a cheap or free wifi hotspot. The charge, if there is one at all, would be more like £1 for 10 minutes and only at one end.

International roaming around Europe is a huge cash cow for the operators. That makes mobile WiFi VOIP extremely disruptive technology.

Roaming? WiFi?? William volk

The WiFi hotspots in the USA have done such an ABYSMAL job of implimenting roaming, that I don’t know if using the two words together makes any sense.

I have a T-Mobile WiFi account (Starbucks) and have yet to find A SINGLE WIFI ROAMING SYSTEM that works with my account.

Inexcusable.

I now use my GPRS phone to connect my laptop to the phone when I travel (unless there’s a Starbucks nearby).

thanks all ..Ajit Jaokar

apologies for not participating. had a pc crash and still sort of recovering. thanks for the contribution. kind regards Ajit

The full thread can be found HERE

mcommerce in the UK

mcommerce.bmp

Another in the series of ‘Best of Mobile apps club’

Discussion started by Justin Pearse , Features Editor, New Media Age (NMA)justin.pearse@centaur.co.uk

Other contributors include:

Mr Ewan J. MacLeod CEO and co founder neoone

Paul Buckley : Managing consultant IBM

Steve Procter : CEO itagg

Nick Hancock : CEO callwise

Walter Adamson : CEO Digital investor

Ms Mary Anne Tolentino Asst. Vice President, President, Director

Bank of the Philippine Islands, Philippine

Mark Len: Alpheus solution

Jack Stevenson

Pauli Visuri : CEO Addwit

Louis Sequeira : 3D e-imaging

Tomi Ahonen : Tomi Ahonen consulting

Dr. Christian Mayaud : The verticomgroup

William Volk CEO Bonusmobile

Mobile commerce? [ justin pearse ] [ 15-Mar-05 5:01pm ] [ edit ]

Hi everyone,

I’m writing a piece looking at mobile commerce in the UK. With the EC having finally overturned their e-money ruling (that banned pre-pay users from using premium SMS to purchase anything that wasn’t consumed on the phone) premium SMS could be a perfect payment method from everything for online content to goods like CDs.

For instance, why couldn’t a newspaper, with every album review, add “text ‘U2′ and your address to have the album delivered, billed to your phone.”

Is mobile commerce like this going to take off?

I’d be very interested in hearing from anyone doing interesting work in this area.

Cheers,

Justin

very interesting .. [ Ajit Jaokar ] [ 15-Mar-05 9:29pm ] [ edit ]I hope to spark a debate on this topic. I think the industry would truly open up with we got the mcommerce right. Like Jerry Mc Guire says ..’Show me the money’!

Its sad that legialation is holding back what is otherwise a perfect cross industry mechanism.

Challenge [ Ewan J. MacLeod ] [ 15-Mar-05 10:24pm ] [ edit ]

The operators will need to reduce their cut first Justin.

More trust needed [ Paul Buckley ] [ 15-Mar-05 11:07pm ] [ edit ]there is an additional level of trust that needs to be earnt by the mobile operators before we will all switch to using our phones as a frequent payment vehicle for larger (£s vs p) transactions…if I bought that U2 album on my credit card, I trust that the correct amount will be billed to my account…I might try that on my phone today but I would check it each and every time right now.

Reminds me of the telcos becoming banks debate that raged a few years back…I trust my bank, they have never got a transaction wrong…telco billing errors are significantly more common (unfortunately).

I want to be able to trust my phone provider to the same extent I do my bank/credit card provider as a needed step before mass take-up of mobile payment solutions like this.

The Future is Greed [ Steve Procter ] [ 15-Mar-05 11:18pm ] [ edit ]Justin

At long last the EC have seen the light instead of acting like a bunch of bureaucratic idiots – of course it will take the FSA 23 years to ratify it, but we are close. However, the networks are still extremely greedy and as Ewan says, nowt will change until they re-work their greedy model.

Pay for newspapers!? – get real!!! Imagine if every time you went into the newsagent and handed over your 50p for a newspaper, a little oik from a bank jumped in between you and the shopkeeper, took your 50p, pocketed 30p and handed over the remaining 20p to the shopkeeper. Yeah that sounds like a great concept…not…

why is it that loan sharks are frowned upon for taking unreasonable percentages yet 6 of the biggest companies in this country get away with such out and out greed.

Why on earth do the networks think they can get away with this nonsense!? We are very very close now to having secure credit card payments on mobiles and the banks will dive in head first to support this. The mobile networks as payment takers will be crushed at that point and their premium sms will be something we all laugh about at parties in a few years time.

BTW, I find Virgin and 3, followed by Orange payout the lowest, what do other aggregators think? Lets compile a chart…

cheers

Steve

2-way sms for £99 per year

Steve Procter. Chief Executive. +44 (0)8712 777 111.

Greed II [ Steve Procter ] [ 15-Mar-05 11:22pm ] [ edit ]

Justin

Sorry, to answer your specific question, if the networks sort out their cut and get down to a few percent like the banks then yes, premium sms will become the killer app to kill all other apps dead…it will be majorly massive…in a very big way…..big…

cheers

Steve

2-way sms for £99 per year

Steve Procter. Chief Executive. +44 (0)8712 777 111.

A huge opportunity for the mobile operators but … [ Nick Hancock ] [ 15-Mar-05 11:24pm ] [ edit ]

My view is certainly that there is a huge opportunity for the mobile operators if they can get this right.If I could pay for goods, easily, securely, cheaply from my mobile why would I continue to carry around a wallet as well.

However … it does seem unlikely at the moment that the mobile operators are capable of achieving this because of the reasons Paul and Ewan have pointed out. My advice to the Operators would be

1) Sort out your billing systems or better, do a deal with a Bank to do it for you and

2) have the belief that a small percentage of a huge volume of transactions will be worth more than a huge percentage of a small number of transactions.

Nick

Check out the BT Callwise trial at www.bt.com/callwise for cheap international calls from your mobile

It’s easier to just swipe the phone [ Walter Adamson ] [ 16-Mar-05 3:16am ] [ edit ]

There’ll be opportunities for premium SMS of course and it will fill a niche but for items where you are physically present (e.g. buying a newspaper or at the convenience store) it will be far easier to just swipe the handset such as in Felica in Japan.

http://www.imodestrategy.com/2005/02/_050227_weekly_.html#50227-1

- Walter

www.imodestrategy.com

www.digitalinvestor.com.au

yes, the operators forget that there are other options .. [ Ajit Jaokar ] [ 16-Mar-05 6:02am ] [ edit ]

consider the now almost ubiquitous oyster card.

It could very easily double up as a payment system. In fact, I would like it to. Its great and convient. I believe toronto has a system as well.

Assuming other things are okay(such as premium rate %), I see this as a huge opportunity.

I believe that operators will drop their cut(without which there will be no market as such)

Mobille apps are all about volume .. so I think operators will recognise that ultimately ..

However, with the EC regulation, the first hurdle seems to have been crossed.

See my new book at : www.opengardens.net

Mobile Banks [ Steve Procter ] [ 16-Mar-05 7:16am ] [ edit ]To become major payment providers, the networks need to:

- gain trust that they can look after every single penny of your money. Many would argue that the networks’ billing systems will not stand up to the sorts of transaction volume increases that would be seen if mcommerce really took off.

- offer payouts of up to 99% of the revenue. As everyone on here knows, that’s what you can get from accepting credit cards in ecommerce.

- improve the delivery reliability. Until last year delivery receipts on premium sms where not even widely available. Even now, not a high enough percentage of premium sms is delivered and charged sucessfully to the phone. This makes the model very difficult for some content providers as they might have already delivered the content. Delivery receipts need to become realtime (just like card payments) – the business wants to know if the recipient has been charged, not necessarily if he has received the sms on his phone – these two things must be seperated out; a major undertaking for the networks.

- improve payout terms. The networks hold on to premium revenues for around 1 month before paying out. This has to be much improved on if more and more businesses are to be able to take advantage of building mcommerce into their business strategy.

- convince their shareholders that this is what they should be doing rather than making their phone networks better and improving on customer satisfaction, retention and experience.

- go head to head with the traditional banks and credit card companies who will soon be appearing on a mobile phone near you.

cheers

Steve

2-way sms for £99 per year

Steve Procter. Chief Executive. +44 (0)8712 777 111.

G-Cash Mobile Commerce [ Mary Anne Tolentino ] [ 16-Mar-05 7:18am ] [ edit ]

The mobile commerce model in the Philippines launched by a major telco, G-cash seemed to be a good model for m-commerce. Previous to this its major competitor Smart Telecom, launched a few years back its card based m-commerce service (Smart Money). I don’t think it really helped grow mobile commerce. There were many obstacles in usage.

This time, I believe Globe struck the gold mine with its G cash – mcommerce via sms. It recently won in the recent GSM Association Awards in Cannes.

One can use gCash for payments, international and domestic money remittance. One can also cash in their gcash load in various merchant partners nationwide. They had to pass through our central bank for approval.

Philippine Ecademy Club

thanks Steve .. [ Ajit Jaokar ] [ 16-Mar-05 7:20am ] [ edit ]

know your views on this one!

gcash .. [ Ajit Jaokar ] [ 16-Mar-05 7:21am ] [ edit ]this is great maan. Good to hear from you, hope you are well. rgds Ajit

Alternatives to SMS [ Mark Len ] [ 16-Mar-05 9:13am ] [ edit ]The operators have taken steps to address many of the concerns raised above. Trust, slice of the pie etc seem to have been addressed through the formation by 5 of the large operators of Simpay.

In essence the operators or shops can provide the content/goods and if they are connected to the Simpay scheme then the payment for the goods can be made using your phone with the settlement done by the credit card company (or the phone company if you wish). All parties get a small slice of the pie with the shop getting the most of the price of the goods.

It does address one of the major criticisms of the operators, namely trust. I agree with Paul that the revenue assurance in the

mobile companies is nowhere near the level of financial institutions and this scheme should address that area. I for one hope that when launched it is a success as I never really trust the premium SMS market. It will mean operators adopting the “sell more for less of a cut” rather than the current approach.

On another note, the possibility of paying for goods via SMS can already be done in the Czech Republic amongst other countries. There are quite a few soft drink vending machines that accept SMS payment in Prague. SMS a number on the machine and a Coke pops out!

Cheers

Mark

Why SMS when you can swipe [ Jack Stevenson ] [ 16-Mar-05 9:51am ] [ edit ]

As user not a provider, I think this would be great. If this was everywhere I would not need a wallet (where is your ID ?) I agree with Walter on the swipe idea. If I had a choice between swiping and having to type in an SMS message I know which I would prefer. I know it has been done and seems to be working in Japan, also the Oyster model seems to be working.

I have to agree with the comments about trusting the phone companies to handle separate financial transactions, but then I don’t think I would trust a bank to provide my mobile services. Perhaps some sort of agreement between them, maybe some sort of guarantee on the transactions from one of the ‘trusted’ (I use that word reservedly) banks.

Over all the sooner the better.

Jack

Note cost difference between small cash and bigger transactions [ Pauli Visuri ] [ 16-Mar-05 11:27am ] [ edit ]

Note that when considering the transaction cost (or the “cut” that operators take), one should keep in mind that small transactions are always much more expensive. While one may pay only a few % charge for card payments of £10 and up, cash handling is a completely differnt thing – there are costs with cashing, transporting, counting, bank charges etc – plus the cost of crime & fraud (and safety measures). Most proposed “electronic wallet” schemes have found that charges of up to 20% for transaction cost are well received for petty cash applications, especially where the whole sale is automatic such as in vending machines.

In Helsinki, Finland you can buy a tram ticket through SMS. The manual alternative, selling a £0.50 single paper ticket from a kiosk, must certainly have cost more than 50% of that ticket price!

it may not be as logical as that .. [ Ajit Jaokar ] [ 16-Mar-05 7:11pm ] [ edit ]i.e. in Korea .. I have seen SMS being used for payments. same with philipines and same with finland. Its like SMS itself .. probably not logical but still being used by people.

See my new book at : www.opengardens.net

what exactly was the EC ruling? [ Ajit Jaokar ] [ 16-Mar-05 7:12pm ] [ edit ]Is it all ‘go’ now for payment via SMS (from a legal standpoint)

ofcourse there are still the operators but .. from a legislative hurdle .. is the e-money bill as applicable for mobile efectively dead?

See my new book at : www.opengardens.net

Mobile phone companies as banks? [ Louis Sequeira ] [ 17-Mar-05 1:10am ] [ edit ]In the light of this discussion I think, an old article on Macalla’s site from 2003 should make interesting reading

Here is some info on the Czech operator Oskar’s set up on Axalto’s site

Louis

——————————————————————————–

Footprints on the sands of time are not made by sitting down.

What with all the doom and gloom again.. :-) [ Tomi Ahonen ] [ 17-Mar-05 2:32am ] [ edit ]Hi gang, sorry was away in Berlin speaking at a CRM conference, so dropping into this thread rather late.

But come on, you all at the Mobile Apps Club should know better. Of course we can have m-commerce, there are too many success stories all around the world to ignore. You can pay for almost anything by mobile phone already. Its no longer limited to “micropayment” such as the vending machines (first introduced in Finland back in 1998, about the same time when the very first mobile VAS – Value Add Service ie downloadable ringing tone was launched also of course in Finland), from buying train tickets in Austria to ski tickets in Norway to airplane tickets in Japan to buying movie tickets in the USA to paying your speeding tickets in Abu Dhabi, etc etc etc. When all the world was looking at the introduction of the i-Pod and i-Tunes, quietly the Korean market introduced direct MP3 music file downloads to 2.5G and 3G phones, and for example Ricky Martin sold 100,000 songs in just a week all the way back in 2003.

M-Commerce is not a trivial thing either, totally the opposite. Here in London in less than a year 20% of all congestion charges were already paid by mobile (don’t have latest numbers, know it is much higher by now). When Robbie Williams performed in Vienna Austria, 20% of his tickets were sold by mobile phone. In Helsinki Finland already 40% of all single tickets sold to public transporation are paid by mobile phone. In Croatia half of all parking tickets are paid by mobile phone. In Japan a massive 84% of all mobile phone users already subscribe for some paid news content on their mobile phone.

Yes, the global music industry? Already 13% of the global music industry revenues are paid for by mobile phone. Don’t pooh-pooh the humble ringing tone; already today our British pop stars earn more from the ringing tone versions of their Top-of-the-pops hits than they earn from the sales of Single CDs. And following right in the footprints of music is the second huge global shift, the videogaming industry, where 5% of the global gaming software revenues come from games downloaded to – and paid for by – the mobile phone.

The stats are overwhelming, overpowering, indisputable, irrefutable. Compare to ANY previous innovation in payment mechanisms, including paper money, cheques, wire transfers, credit cards, debit cards – never before has ANY industry seen such rapid cannibalisation as those top numbers I listed – public transportation 40% in two years, parking 50% in three years, and paid news content in five years. The internet, a dramatically disruptive technology, did not achieve these incredibly fast rates of cannibalisation in any industry that the Internet has changed.

Don’t kid yourselves, m-Commerce is going to happen globally, everywhere, of course. I do agree with many of the side comments here, such as that we need mobile operators to be less greedy, although clearly the market is perfectly happy to live with the 10% – 90% split as the extremely healthy and robust m-Commerce markets in Japan and Korea attest. (We don’t need to go to 1% – 99%). European operators tend to be offering deals ranging from keeping 40% to keeping 20%, depending on which operator (smaller give better deals) and which market (more mature markets have better deals).

I had over 100 m-Commerce ideas in my first book Services for UMTS and another about 150 more m-Commerce ideas in my second book m-Profits. Essentially all of those have been deployed in some variant by now. I track mobile services and apps and list my favourite each month at my website – and have discussed over 500 of them at public conferences over the past five years – so Justin if you need some specific types of examples, let me know and will be happy to give you more ha-ha…

PS Incidentially, m-Commerce succeeds well when it is one of the Early Eight classes of services (see my third book ha-ha, but I have talked about the E8 about a year ago or so here at the Mobile Apps Club) and especially when the transaction happens several times per week.

PSPS the current edition of Vodafone’s customer magazine, Receiver, has my article on killer apps for 3G. It is the evolution on the series of articles I’ve written on what for the most part is m-Commerce, such as in the Financial Times last year, Telecommunications two years ago and in 3G Mobile three years ago. The Receiver article can be read for free at this site

http://www.receiver.vodafone.com/12/articles/index04.html

Dominate !

Tomi Ahonen / HatRat :-)

welcome back Tomi! [ Ajit Jaokar ] [ 17-Mar-05 6:18am ] [ edit ]

Its time to Dominate!

In the US, We Just Don’t “Get it” At All … [ Christian Mayaud ] [ 17-Mar-05 6:41am ] [ edit ]I always respond with bewiderment when europeans (and chinese, koreans, and indians for that matter) talk about these huge opportunities with “cellphone based e-commerce” … the only example we have in the states of anything close is the whole “downloadable ring-tone” market … which looks completely insane to me … $2 for a 10 sec ringtone???

I think the “downloadable ring-tone market” is purely a phenomena of “someone else is footing the tab” (we have a similiar phenomena in US healthcare economics) … in this case, teenagers have no disincentive to not download the new ringtone since it is their parents who pay the phone bill anyway and may not even notice it … if the teenagers themselves had to pay the two dollars for the new tone, I suspect the market would be quite a bit smaller than it is …

yes, we do use alot of SMS (both teenagers, families, and yes, a growing amount of “intracompany” communications as well) but to think of SMS as an e-commerce platform?

You europeans are just “way out there” on the bleeding edge — we see no sign what-so-ever in the US market that “cellphone based ecommerce” is even a remote possibility …

Christian Mayaud | Managing Director | The Verticom Group

cmayaud@theverticomgroup.com | 914/239-3733 | Skype ID cmayaud

“Intelligence is like Four-Wheel Drive … It just gets you stuck in more remote places.” — Garrison Keilor

premium sms [ Steve Procter ] [ 17-Mar-05 6:42am ] [ edit ]

Tomi

You are talking about the wider remit of commerce being done via mobile phones in general, not specifically premium rate sms – which is where this debate started. Correct me if I am wrong but paying your London congestion charge by mobile does not use premium rate sms, it is simply a mechanism to trigger your already registered credit card into being debited. Now don’t get me wrong, this kind of use of a mobile to trigger payments is indeed fantastic. Maybe we need more of that. But ultimately this is a credit card payment not a mobile transaction.

And if we are looking at all sorts of ways of using a mobile for commerce then lets not forget that the latest versions of both wap and j2me are secure. So seeing wap sites or java apps take credit card details is going to happen more and more. Indeed Vodafone themselves use cards over wap to get your card during their 18+ verification checks. Well they did on mine anyway; I hope they do some intelligent checking of the phone type to ensure they aren’t presenting a card screen on wap phones that are still unsecure, or that would be another very interesting story for Justin…..

cheers

Steve

2-way sms for £99 per year

Steve Procter. Chief Executive. +44 (0)8712 777 111.

Oyster, Smart, and the US market [ Walter Adamson ] [ 17-Mar-05 8:02am ] [ edit ]The Oyster card is the Felica Networks/Sony consortium. In Japan they have just moved up a notch and gone through the pain of integrating the card into the phone, which is painstakingly complicated, and testing readers that are 100% reliable and super fast when there are 100 commuters jamming behind you at the station.

The Smart Communications applications in the Philippines are huge models of SMS m-commece success.

The integrated Felica card in Japan, as operated by DoCoMo, has no transaction fee. That is, in plain english, DoCoMo takes no transaction fee. I don’t know why not, but probably as an incentive to get the new handsets and the systems very widely adopted.

And in the US, Christian, the games market is huge and growing and this is much more than ringtones. And my predictions about business applications bursting forth this year driven by the US also will mean an explosion of mobile commerce applications, starting in B2C and then in B2B.

But Tomi’s point is right. What are we all discussing here? This is all a reality?!

- Walter

www.imodestrategy.com

www.digitalinvestor.com.au

Good points yet.. [ Tomi Ahonen ] [ 17-Mar-05 9:59am ] [ edit ]Hi Walter, Steve, Christian and Ajit

Good points Walter, Steve. Steve on the topic itself, you make a valid point, but I did understand that Justin was looking for such examples in a more general sense, he didn’t seem to limit it to premium SMS

On premium SMS, I think the perfect model – Christian you shuold especially look into this – is Habbo Hotel (also discussed here at the Mobile Apps Club) – where youngsters pay for small payment value content on the fixed Internet, through premium SMS on their personal cellphones. In Finland in the 12-16 age bracket, Habbo Hotel is about the most addictive legal thing you can peddle to kids this side of Crack Cocaine ha ha..

While on the subject, Christian, the CTIA just announced USA finally hit the 60% penetration level. This is the critical point when the “total” economically viable population in Western countries has a cellphone, meaning that true change in society can start to happen, and the transition goes from technology enthusiasts pushing the technology, to average people and businesses inventing hundredfold more uses all by themselves. Yes, a big step for you Americans, but remember Finland hit that milestone back in 1998. You are literally 7 years behind. So all what is now happening in USA, you can read from history books as having happened in Finland in 1998, in Sweden Norway and Italy in 1999, in Israel, Hong Kong, Singapore in 2000, etc. Yes, Ringing tones were literally the start of it all. Yet mobile interenet CONTENT revenues world wide are a 15 Billion dollar industry – many times that of the fixed Internet. And where 70% of the fixed internet incomes are from “questionable” services like adult entertainment and gambling, the 70% of cellphone content revenues are from music (ringtones), games and news.

Believe – ha-ha, and Dominate !

Tomi Ahonen / HatRat :-)

re: what exactly was the EC ruling [ justin pearse ] [ 17-Mar-05 10:01am ] [ edit ]

Re Ajit’s question on the EC ruling. In the UK this still needs to be interpreted by the FSA, which will publish a final guidance note on it, in the next few months hopefully. But Vodafone, for one, has already sent out letters to aggregaors giving the green light for services to launch. The rest are under pressure to follow

Habbo Hotel [ Steve Procter ] [ 18-Mar-05 7:23am ] [ edit ]Tomi

Please tell us how much a user in your Habbo Hotel example pays for an example service. Then please tell us how much of that money is received by Habbo Hotel and how much the operators keep?

thanks…

cheers

Steve

2-way sms for £99 per year

Steve Procter. Chief Executive. +44 (0)8712 777 111.

Interesting Stat from the Walled Gardens talk at CTIA [ William Volk ] [ 19-Mar-05 7:14pm ] [ edit ]Completion of purchase on a web site (mobile content):

Credit Card: 5%

P-SMS: 35%

stats [ Steve Procter ] [ 20-Mar-05 2:48pm ] [ edit ]

William

Many web based businesses pay less than 2% bank charges. And with good house-keeping, security, etc chargebacks are not as high as many people like to think.

Yes, psms charges can be as good as 35% but worse case take this example…..on a 25p psms (which is actually 21p +vat) the payouts many aggregators give on some of the networks is down to around 7p…..which equates to a “bank charge” of 66%!!!

cheers

Steve

2-way sms for £99 per year

Steve Procter. Chief Executive. +44 (0)8712 777 111.

Sorry don’t have Habbo Hotel actual numbers [ Tomi Ahonen ] [ 21-Mar-05 10:05am ] [ edit ]Hi Steve

I don’t have the revenue-sharing numbers for Habbo in the public domain, so whatever I have, I can’t share.

But about the time when Habbo launched, the second largest Finnish mobile operator, Radionlinja (since renamed Elisa), was publically quoted that for a typical 1.50 Euro (=1 UKP) payment (including tax), the tax people keep 18%, the mobile operator keeps 27%, and the content owner gets 55%. If we remove the taxman, the revenue-sharing split was therefore 33% to the mobile operator, 67% to the content owner. I would guess that TeliaSonera (the biggest player) matches Elisa’s levels, and the smaller players (DNA, Saunalahti etc) will offer slightly better deals for content owners. Also I would expect that this level of revenue-sharing is still shifting, with gradually more going to the content owners.

Sorry, thats the best I have. What have you other Ecademists seen in your markets?

Dominate

Tomi Ahonen / HatRat :-)

Image source: http://www.seri.org/frimg/forum/01/1987_f.jpg

which mobile dev platform for greatest coverage?

dev platform.bmp

Hello all

As many of you know, I run the Mobile applications club

I am starting a new series on my blog – The best of mobile apps club

This is the first post

Loads of useful information here

Started by Paul Buckley - Managing consultant – IBM global services

Others who have contributed to this thread include

William Volk – CEO Bonus Mobile Entertainment- Bonusmobile

Richard Spence – Director Bluetrail

Tom Weiss – VP – T-mobile International

Tom Hume – Director – futureplatforms

Alex Kerr – Director – It assist

walter adamson – director imode business strategy

Dr Richard Dallaway – director – spiralarm

Mr Aju kuriakose accelicim.com

which mobile dev platform for greatest coverage?

[ Paul Buckley ] [ 10-Jun-05 4:58pm ] [ edit ]

If you an apps developer, and you want to launch an app in Europe, which do you pick today, i.e. which is closest to critical mass or has the greatest potential to achieve that?

what do we have out there…Symbian Series 60/70/80/UIQ, MS Smartphone/Pocket PC, Palm, J2ME, Brew…others I’m sure.

I’ve noticed that when people show me apps they have developed, they are usually Symbian Series 60 or J2ME but I’ve got no basis for saying that they are the most popular or even which is the greater of the 2.

Presuming resources are limited, which do you do to get greatest coverage?

rgds,

Paul

good question .. [ Ajit Jaokar ] [ 10-Jun-05 8:43pm ] [ edit ]but depends on the application. First impressions – SMS, WAP and J2ME. followed by Symbian

But it all depends on which application.

Doesn’t … [ William Volk ] [ 11-Jun-05 4:29am ] [ edit ]Java allow you to write the app and run it on any platform? That’s what the kind man from Sun told me.

:) [ Ajit Jaokar ] [ 11-Jun-05 4:57am ] [ edit ]

write once run anywhere is for the textbooks

I always find it amusing but Sun dont make that into a claim nowadays(at least for mobile)

But then they also famoulsly said ‘We are the dot in the dot-com’ ..

Another claim I have not heard for a long time!

My point is … [ William Volk ] [ 11-Jun-05 5:16am ] [ edit ]

No easy solutions. If you can do it effectively with SMS, go for it.

Java’s advantage is a richer applications and more rapid response.

I think the incompatibility issues will help Flash gain share.

Sounds familiar doesn’t it?

flash .. interesting .. [ Ajit Jaokar ] [ 11-Jun-05 5:25am ] [ edit ]

Do you think something like bitflash ?

write once develop on mobile? [ Paul Buckley ] [ 11-Jun-05 8:15am ] [ edit ]so I’m no J2ME developer (can you tell ) but I had at least hoped that if you were a Java developer and had say experience in a server environment, that moving into the mobile world should be fairly straight forward.

Now there should be LOADS of Java developers around so in my mind that should make J2ME quite a desirable platform? There are IDEs out there, like Netbeans Mobility that make it really simple…dont they?

How about flash/bitflash – is this already loaded onto some handsets? Any applications already launched using this?

rgds,

Paul

j2me flash etc .. [ Ajit Jaokar ] [ 11-Jun-05 8:24am ] [ edit ]

the problem with J2ME is not the technology but the fragmentation at an implementation level. To be fair Sun does not control all of it(operators do). Thus, you could write a game once but porting it to many different operator platforms, gaming sites etc is a major hassle. In fact, porting alone is a BIG business(mostly outsourced to the likes of India)

flash .. I am not sure .. william suggested this .. so keen to hear from him

bitflash has been around for a while but has competitors

Test Test [ Richard Spence ] [ 11-Jun-05 8:58am ] [ edit ]

I would echo Ajits comments .. it is tough development environment but as Tom Hume may well add “if it was easy then everyone would be doing it” ;-)

It makes you chukkle really when you remember having to worry about four or five browser implementations back in the late 90′s now we have to have a WURFL.

See http://wurfl.sourceforge.net/backgroundinfo.php

Coverage [ Tom Weiss ] [ 11-Jun-05 9:30am ] [ edit ]

There are 100s of millions of Java and Brew handsets in the market but only 10s of millions of Symbian handsets.

However, native Symbian apps to me tend to have a better user experience so it’s a difficult call for me in terms of which way to go. There is a good chance that Flash will become the standardised platform for user interface development which would be a great step forward but of course there will still need to be lower level code to work closely with the network.

t.

Tom Weiss

It’s Deja-Vu All Over Again [ William Volk ] [ 11-Jun-05 12:52pm ] [ edit ]In 2000 I founded a ASP (Application Services Provider) now known as ZipProof. Java applet used for content proofing (markup) of simple ads and print jobs.

In 2001 we actually exhibited in Sun’s booth at the old Internet World show. The programmer had to spend the night before the show changing the applet so it would run on the Solaris version of Netscape.

Back to the present. I don’t see why there can’t be a certification process and a test suite for the J2ME implimentations.

The ‘sandbox’ model of Midlets make them an ideal mobile programming platform, but that advantage is lost when we find ourselves having to create different versions for even handsets in the Series 60 family. (For example the famous 6600 won’t support signed-midlet issue).

Since Sun now can say that Java represents the clear leader in installed platforms, why wouldn’t they want to make a start at tackling these issues.

As to Flash-lite … if only it was Flash 5 compliant.

Question: What is the scripting language for bitflash?

Flash vs Java vs SMS vs WAP vs Everything else [ Tom Hume ] [ 11-Jun-05 1:25pm ] [ edit ]I’m a big fan of Flash, Macromedia are a client of ours and I think it could be an excellent way of doing mobile apps… but there are problems with running mobile flash on restricted devices, just as there are with phones. It may be way better visually, but under the hood it’s still a virtual machine running in a very limited environment.

I suspect that once we start seeing a groundswell of mobile Flash apps (there have been a few commercial launches but mostly they’re proof-of-concepts), it’ll be apparent that going down the Flash route doesn’t let you magically avoid all the problems that on-phone development raises. Right now it feels like we’re comparing the gritty reality of Java development with what Flash *promises* to deliver… which isn’t a fair comparison.

The rule I can see with all this stuff is the better-distributed the platform, the less capable it is. So the relatively minor development platforms (e.g. Symbian – which is massively successful but still minor compared to say WAP or SMS) are best for doing apps targeting specific niche markets.

IMHO Java is just at the point where it’s starting to be everywhere. There are lots of problems developing cross-handset Java apps, and you need to be pragmatic: targeting, say, the top 75% of the market and working with operators or other delivery partners to identify what handset the top 75% are using.

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

so what went wrong with J2ME certfication? [ Paul Buckley ] [ 11-Jun-05 1:51pm ] [ edit ]William,

There already is a certification process for all Java specification developed through the JCP.

This is defined by the Reference Implementation (RI) and Technology Compatibility Kit (TCK) which is required for each JSR. The RI is code that implements the spec, and the TCK is the test suite.

J2ME is no different.CLDC is JSR139 and MIDP is JSR 118 for example. The RIs ad TCKs are both there on the home pages.

So what went wrong. Did the handset manufacturers just sort of forget to certify and confirm to the spec, purposely ignored it or ??

rgds,

Paul

This must be … [ William Volk ] [ 11-Jun-05 2:16pm ] [ edit ]A definitiion of certification that I was previously unaware of.

I’ve posted on this before, there are major incompatabilities even within a family of handsets. What’s more some of the manufactuers won’t even support developers or provide documentation on the extensions … for example camera phone operation.

See: Fixing portability

Tricky tricky tricky tricky [ Tom Hume ] [ 11-Jun-05 2:23pm ] [ edit ]

Plus event if you can write-once run-anywhere, you often don’t want to.

For instance, a game might work radically differently on a portrait-sized screen to how it will work on a square screen. Levels for a platform game might have to be redesigned for smaller (or larger) screen sizes. You might want to stick to underlying conventions of the handset (e.g. on motorolas the left softkey tends to cancel, on nokias it confirms).

Differences in J2ME implementations are a big problem, that’s true – and developer sort is poor (but I suspect that like healthcare spending or network operator customer care, it’s impossible to ever spend enough and get it right).

But the underlying issue is that creating a unified platform for apps which runs on a huge range of mass-market, limited devices is *really hard to do*. And if you look at where we are now wrt the handset as a platform for applications, and where we were 5 years ago… there’s been a hell of a lot of progress.

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

All true [ William Volk ] [ 11-Jun-05 2:44pm ] [ edit ]

But if you could write a app to the spec, no extensions, and be assured it would run resonably well across MIDP platforms … it would be a good start.

Right now there’s a huge opportunity for Adobe (Macromedia) to do it right. It will be interesting.

mobile devices RADE.. [ Nick Meese ] [ 11-Jun-05 3:03pm ] [ edit ]

Diff tools for different apps – standalone apps on devices can be achieved with Java apps. When it comes to client server apps on devices – There are some interesting tools out there from www.Momote.com. Sat Forms (used to be Intellisync RADE but now belongs to Thacker technologies) is very good for Palm and PPC – build the app and target the device or OS (no Symbian or Java). www.appforge.com claim to support palm, ppc, smartphone, PC and Symbian devices and permutations of lack of local filtering hinders some palm developers for the time being.

Nick Meese ….Informed mobility

REBOL [ Alex Kerr ] [ 12-Jun-05 5:39pm ] [ edit ]

I don’t know if many in the mobile space are familiar with the REBOL (www.rebol.com) platform/language or have tried it, but I believe REBOL running on mobile would be an absolutely ideal marriage made in heaven, to put it mildly It has all the characteristics that offer the solution we all wish J2ME was on mobile, and more. Briefly – very small, powerful platform; very easy to program; makes everything from advanced network enabled apps to whizz-bang graphics and GUI’s child’s play; extremely small apps, truly write once, run anywhere.

It’s available on many desktop and server platforms, but not yet on mobile. I’m trying to convince them to change this

Yes… [ William Volk ] [ 12-Jun-05 7:10pm ] [ edit ]

I know the REBOL folks from my Amiga days (I was a co-founder of Aegis Development) tell Carl I said hello. Very clever stuff.

There are a lot of REALLY TINY languages that could have been better choices than J2ME. Heck J2ME should have been better than J2ME.

In the late 1980′s/early 1990′s we had a Scheme-like OOP scripting language who’s kernal was about 20kb in size (compiled C code). Totally “sandbox-ed” we used it on adventure games and multimedia titles. Started as a way of porting HyperCard titles to the PC.

Measure how many J2ME instructions are being executed per second on a typical phone. It’s a SOBERING number.

Well… [ Alex Kerr ] [ 13-Jun-05 1:03am ] [ edit ]

My thoughts were along the lines of we clearly need some sort of solution to the J2ME fragmentation problem, something that encompasses as many of the existing handsets out there as possible many of which won’t get upgraded for months or a year or two, something simple and elegant that sat on top of and/or beside J2ME on the handset (given that you can’t replace it) and smoothed over all the inconsistencies and offered a completely level playing field again.

REBOL’s done it on the PC in a technically very elegant way. I guess my question is why can’t it be done on mobile? Is it feasible? I’d like someone to pursue it until we get a definite no, because the alternative is to keep plugging away at all the parties over the coming months and years to get J2ME defragmented (as it were) and then wait for it to slowly permeate the existing user base out there via the standard phone upgrade cycle. I’d say this is a process that would take several years realistically? In the meantime developers and users suffer the consequences of the existing situation.

And yes, if not REBOL then something that fits the bill. Yes the whole stack would be technically inelegant (phone hardware, phone OS, J2ME, REBOL, user app), and yes it could be faster if you didn’t have all the layers, but a phone’s not a server or a workstation, and most have not-half-bad CPUs these days, and how fast does a phone have to be? I think the rewards more than compensate for the tradeoffs, until J2ME gets consistent across the market. Oh, and it would be optional and free for users, just as REBOL runtime is now – if the user wants the app they download and install once the REBOL runtime, then run the app. If they don’t want the app, no worries, no one’s forcing it on them. That way REBOL (or whatever) becomes your abstraction layer for your app to smooth over the inconsistencies (ignoring all it’s other advantages for a moment).

Alternatively, someone writes an abstraction layer as a J2ME app that offers the standard J2ME APIs as it’s APIs, thus effectively becoming invisible but removing all the J2ME inconsistencies across handsets. Now there’s a market opportunity!

Off topic – Amiga gets my vote as greatest digital device ever made. Those were the days

It’s not about the lingo… [ Tom Hume ] [ 13-Jun-05 3:07pm ] [ edit ]

I suspect that it’s not really an issue of which language you go for. If you have tens of different manufacturers implementing a standard, each on tens of devices over a 5-year period you’re going to get inconsistencies and problems, whatever.

William: yes, Macromedia have an opportunity here, but from what I’ve heard on various lists the same old memory-shortage etc. problems rear their heads. They’re doing what’s fundamentally the same thing: implementing a virtual machine on top of lots of different sorts of restrictive hardware. Now, if it’s just MM implementing this then there’s a better hope for consistency between implementations (and consistent QA of these implementations)… but I’m not sure how one company will manage to deploy across every handset out there.

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

My Vote: [ William Volk ] [ 13-Jun-05 5:21pm ] [ edit ]

Sun should build compatibility test suites so phones can be ‘certified’ as compliant. There could be a sticker etc to tell users that their phone could run these standard apps ..

The rule would be that an app written to these specs would run on these phones unmodifed. It might not take advantage of everything, but it would be a way of supporting many handsets quickly.

As I mentioned before, we were faced with the same problems with CD-ROM drivers in the late 1980′s, we (Activision) actually created a test suite that we shipped with product and sent to vendors to get everyone in line. Within a year, the issues were resolved.

I suspect that Macromedia will only focus on the high end of the market. I don’t expect a Series 40 version of Flash lite.

No hope for REBOL unless it gets into the native code of the phone. Running on top of Java would be a bad idea, the JVM’s are quite slow, and with some of the memory restrictions (i.e. 56kb for a JAR on Chinese Series 40 Nokias) you just can’t do it.

It would be easy to use the Symbian kit to build a Series 60 REBOL, but looking at the REBOL site I see that haven’t gotten to OS-X yet.

Certification [ Tom Hume ] [ 13-Jun-05 8:54pm ] [ edit ]

I presume there is some certification required before a handset can advertise itself as Java-compatible or use the Java logo… but yep, certification would be good – presuming that Sun had the bandwidth to do it (how many Java handsets get released nowadays anyway) etc.

If Macromedia only go after the high end of the market, then there’s still a good few years where Java is the only option for the low-end…. i.e. we have to learn to live with it.

Where’s your CD-ROM post? I’ve had a search but can’t find it

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

RE: Certification [ William Volk ] [ 14-Jun-05 12:53am ] [ edit ]

I presume there is some certification required before a handset can advertise itself as Java-compatible or use the Java logo,

Not much testing as far as I can tell. Stuff as basic as what sort of array boundary conditions will throw exceptions isn’t even reliable.

Aren’t we underestimating Flash Lite’s influence already? [ Walter Adamson ] [ 14-Jun-05 12:58am ] [ edit ]Under the Yankee Group Microscope:

NTT DoCoMo’s Experience with Flash Lite

Flash Lite has become a de-facto standard for most large Japanese

mobile internet content providers. Approximately 15 million NTT DoCoMo subscribers have Flash-enabled devices.

Sophisticated information services using Flash Lite have characterized the first wave of innovation using Flash Lite. But Flash Lite extends readily to the critical task of developing dynamic, highly customizable user interfaces. The animated graphical user interface simplifies complex search operations, eliminating the need to navigate through a series of hierarchical text menus.

• We estimate that at least 15 million DoCoMo subscribers (approximately 35% of total i-mode subscribers) will have Flash-Lite-enabled phones by year-end 2004.

• More than 60% of the 4,400 official i-mode sites are based on Flash Lite.

• Approximately 20% of the (88,000:wja) unofficial commercial i-mode sites are Flash-enabled.

- Walter

www.imodestrategy.com

www.digitalinvestor.com.au

Link to my post on Java issues [ William Volk ] [ 14-Jun-05 1:00am ] [ edit ]See:

http://wireless.ecademy.com/module.php?mod=club&op=forum&c=24&t=165746&m=187903

that’s sad [ Paul Buckley ] [ 14-Jun-05 8:53am ] [ edit ]hmmm, that’s very sad William, a) that the problem even exists at all and b), that in the 6 month when that was discussed, nothing apparently has changed.

rgds,

Paul

Open standards and monoliths. [ Tom Hume ] [ 14-Jun-05 9:01am ] [ edit ]Walter: is it coincidence that the markets which seem to enjoy fewer problems of compatibility amongst handsets are the ones which tend most towards being monolithic?

I-mode: specified and tightly controlled by DoCoMo, who specify and QA handsets and content services. BREW: a similar situation, run by Qualcomm.

Perhaps these problems are a natural consequence of our preference for open standards; that interoperability problems arise from this.

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

Open Standards? [ William Volk ] [ 14-Jun-05 6:04pm ] [ edit ]

No, I will lay the blame at the MIDP spec and the lack of testing suites. This is basic stuff.

The intial MIDP spec basically was an invitation to fragmentation. As an example the keypress issue alone was handled badly, with no standardization of game control inputs (software).

The BREW folks do complain about some fragmentation as well, but obviously having one vendor (as with Flash Lite) enforces better standards compliance.

I just am amazed that Sun doesn’t treat this as a major issue. To have the majority of a market and to be aware of this and not launch at least a series of test suites seems incredable to me.

so let’s raise the issue at JavaOne [ Paul Buckley ] [ 14-Jun-05 8:01pm ] [ edit ]JavaOne is coming up soon. Richard Dallaway is attending and has offered to raise any issues (demonstrable bugs) we could suggest with the J2ME team.

So let’s give him some ammunition! Minimal code test cases which can be independently reproduced would be ideal.

rgds,

Paul

Two examples for you… [ William Volk ] [ 15-Jun-05 12:22am ] [ edit ]1. Character array out of bounds exceptions behave differently on different phones. According to the J2ME standard, the String.substring(int beginIndex, int endIndex) method throws an IndexOutOfBoundsException “if the begin index is negative, or end index is larger than the length of this String object, or beginIndex is larger than endIndex.” This works in practice on the Nokia 3650. However, on the Motorola i95cl the IndexOutOfBoundsException is thrown when beginIndex is EQUAL TO OR LARGER than endIndex. This does not conform the standard and causes code to run differently on the two phones.

2. HTTP I/O issues. Standard Java tricks such as closing an http connection to force a timeout don’t work on most devices. You have to monitor I/O processes to kill threads when they freeze.

How to report a bug [ Richard Dallaway ] [ 15-Jun-05 8:23am ] [ edit ]

To demonstrate a problem I’d need:

1. The minimal code that exhibits the problem. Ideally as a JAR file and build script.

2. A statement of how to reproduce the problem.

3. A statement of what happens (screen shots, debugging output)

4. A statement of what you think should happen.

5. The reference (document, page or section number) in the appropriate specification that tells you that the behaviour you expect is mandated.

6. Description of the platform(s) where this occurs.

7. A note of where you have raised the issue already (if at all) e.g., with the handset vendor.

8. Depending on the issue or the device, I may need to borrow the handset.

And it’s probably best to cc: it from this thread to my email address, because I don’t always check here as often as I’d like to.

Sorry, I can’t take just text statements of an issue: it needs to be something I can reproduce with your code. William’s first example is pretty close to a good example, I think.

My feeling is that if something is broken we should be reporting it and getting it fixed in the conformance tests.

Cheers

Richard

I suggest [ William Volk ] [ 15-Jun-05 2:51pm ] [ edit ]That Sun or someone go to this fourm:

http://www.microjava.com/

And ask for the above.

I am also pretty disappointed that my array example isn’t ‘good’ enough for them. That’s a ‘winning’ attitude for sure.

Do they WANT to repeat their browser experence on the handset?

AMAZING.

Tom, you’re right [ Walter Adamson ] [ 17-Jun-05 5:40am ] [ edit ]That’s one of the benefits that DoCoMo sells other carriers on when promoting i-mode. And it makes a developer’s life more simple.

The control of the handset standards and the publication of what each handset and its series or level supports is a key to reducing development and testing costs. However even within this i-mode partnership it is not entirely straightforward and different levels of compatibility and features can be time consuming for developers trying to work out their greatest spread of deployment across the i-mode alliance.

It’s a good system, but not necessarily the only one, for example a more rigorous “compatibility lab” approach might be a good alternative, so that the standards are set by a carrier but without forcing or influencing or choosing handset makers. Each maker is free to compete and offer their handsets as long as they meet the compatibility standard, and the carrier would provide resources in the “compatibility lab” to help the handset makers.

The benefit of this approach is that the costs should be less, for everyone, both the carrier and the handset provider. For example the handset provider can make a globally optimal commercial and technical decision about handsets in relation to the compatibility requirement, and build that into their global handset strategies. This is in contrast to the DoCoMo model where the handset makers have to make locally optimal decisions, that is on a local node (i-mode) rather than a global optimal.

The sub-optimal nature of the local decision-making is evident in the fact that Nokia and Motorola have no DoCoMo i-mode presence, or rather have had none up to now but have been enticed to enter the fray by DoCoMo investing in their handset development in order to compensate for the sub-optimal decision to supply i-mode. DoCoMo has decided that the cost of enticing them in is going to pay off by forcing down the price of handsets from their other i-mode handset providers in the local market, whose prices in turn have been pushed up by having to conform to the i-mode requirements. It’s an interesting circle, or more than interesting it is somewhat deadly to the local manufacturers like Sanyo who are losing money hand over fist.

In ths case above DoCoMo is indeed biting the hand that fed it, locally.

Also to blame for this intense pressure on DoCoMo over pricing is the absurd and destructive practice of handset subsidies. These all came about because carriers are hooked and incapable of selling anything except technology and features and iron. When life became just plain too hard for the so-called marketing and the sales groups, when they had to think about NON-priced based competition, and no longer selling iron, they closed ranks and came up with masterpieces like handset subsidies and bundling. These are even more deadly sins, which just paper over the fundamental issues in the short-term. They look great for the first 2 or 3 years, and by then the carriers have dug themselves into an even bigger hole which requires even more radical solutions.

But of course none of these people are stupid. They know. They just can’t give up – these pseudo-cost-based methods are drugs, and the top-line that they bring is also tied to the annual bonus. Because they know is why carriers like DoCoMo are in grave fear of new entrants in Japan like eAccess/eMobile. eMobile is going to follow the compatabilty lab approach – an open market with best global optimum prices from handset makers. No subsidies. On paper eMobile is a feeble small competitor, with very modest subscriber goals, not enough to threaten DoCoMo – on paper. But in fact this attack on the structure of the industry and the previous lazy habits and the hole that the sales people have dug for their companies is hugely significant, and threatening for the telcos hooked on the old models.

And don’t forget, the people behind eMobile are the same people who founded KDDI, who is eating DoCoMo’s lunch every day in Japan. KDDI has taken control of the innovation branding and the youth market, and DoCoMo is in severe catch-up mode, especially while having to convert subscribers from their 2G MOVA network to the 3G FOMA network without them switching to KDDI-au. So guess what – more handset subsidies!

A key outcome of eMobile’s strategy is that it is going to have very competitive (maybe even low!) handset prices and a wide range of handsets to offer – on day one. And subsidies, if any, will be an internal choice of the handset providers. Such decisions will hit the handset makers bottom line, not eMobile’s. This is a flying start to a mobile business, in contrast to some of the slow launches around the world due to high handset prices and limited choice. The compatibility lab will ensure some degree of acceptable user experience, as well as network compatibility.

So coming right back to your question, the i-mode alliance model is the most successful of its kind to date in providing developers with a group of assured handset standards. But it is not necessarily the only model that will work effectively, and watch out for eMobile in 2006 as a test of an alternative.

- Walter

www.imodestrategy.com

www.digitalinvestor.com.au

Another approach [ William Volk ] [ 17-Jun-05 4:46pm ] [ edit ]

Would be to have a 3rd party act as a certification lab for handset companies that wanted a “seal of approval.” These handsets would have access to a larger application pool as apps written to a standard would run on them unmodified.

From the trenches… Maybe UK operators are “gettng it at last” [ Simon Cavill ] [ 17-Jun-05 10:57pm ] [ edit ]As someone who has done a lot of mobile handset development work, I have found the biggest handicap is actually testing applications across a wide range of different handsets. In the past, most operators were extremely unhelpful in this area although we have found Orange UK to be very proactive in this space. We joined their developer scheme sometime ago and attended their developer training event in France last year which was extremely helpful in getting in from of the right people at Orange.

More importantaly they provide access to their labs were you can turn up with an application and test it on every handset they ship at one go. This means you can build and test apps on most of the current handsets on sale across the Uk and most of Europe as many operators share the same devices. I think they deserve a lot of kudos for doing this and will hopefully reap the rewards from increased WAP/GPRS traffic as well as download revenues. I wish other operators would take a note out of their book….

Simon

DoJa enables games developers to create… [ Walter Adamson ] [ 21-Jun-05 8:30am ] [ edit ](O2’s i-mode chief architect, Jag Minhas)… claims that building an i-mode version of an existing HTML site can take as little as 10 minutes. Crucially he also praised the i-mode version of Java for handsets which is known as DoJa.

He reckons DoJa enables games developers to create one version of a game than runs on all i-mode handsets. By contrast some games available on O2’s existing WAP portal [Active] have no fewer than 60 different variants.

The Enquirer 17 June

- Walter

www.imodestrategy.com

www.digitalinvestor.com.au

? [ Tom Hume ] [ 21-Jun-05 11:26pm ] [ edit ]How many handsets is “all I-mode handsets”?

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

Re: which mobile dev platform for greatest coverage? [ Aju kuriakose ] [ 22-Jun-05 5:30am ] [ edit ]hi,

I know lot of messages are being exchanged and lot of experts speaking on the topic. Although I am not an expert but as a person who has been marketing for a company which has a focus on mobile devlopment applications also here are my comments.

The key to choosing your platform is what you want to develop and the deployment strategy on the wireless platfrom. There are various options avaliable

(a) SMS – for low commplexty low data entry applications. SMS development does not need much of technical knowledge from the developer communicaty as there are lot of ready msde toold which will easily plug in to an existing solution. But deployment becomes achallenge some times as you would have to run behnd the operators to get the shot code to run the application. However if you are talking about Enterprise based application it could be done using simple GSM Modems.

(b) J2ME- Although most of the mobilephone and PDAs can run j2me Application you would need to do minor rework on the application to suit it to a particular devices like Palm, symbian, pcoket pc etc . becasue of multiple reasons like VMs being used, Screen size etc.. But the rework involved is less. limitation al the native APIs and functionality of the devices might not be readily avalaible to j2me . Easy to deploy as there is no operator dependency

(c) Device specific/platfrom specific tools like metrowerks for symbain or appforge etc. This tools are good if you plan to reap full advantage of a specific device or platfrom. for example if you plan to make a application which take advantage of the mutli language capability of symbian or Camera functionality etc. it is alway better to do the programming on specific platfroms rather than using the generic platfroms.

I can throw some more light and also open to tlak to you more if you need more inputs from my end

Regards

Aju

Image source: HERE

which mobile dev platform for greatest coverage?

Hello all

As many of you know, I run the Mobile applications club

I am starting a new series on my blog – The best of mobile apps clun

This is the first post

Loads of useful information here

Started by Paul Buckley - Managing consultant – IBM global services

Others who have contributed to this thread include

William Volk – CTO – Bonusmobile

Richard Spence – Director Bluetrail

Tom Weiss – VP – T-mobile International

Tom Hume – Director – futureplatforms

Alex Kerr – Director – It assist

walter adamson – director imode business strategy

Dr Richard Dallaway – director – spiralarm

Mr Aju kuriakose accelicim.com

which mobile dev platform for greatest coverage?

[ Paul Buckley ] [ 10-Jun-05 4:58pm ] [ edit ]

If you an apps developer, and you want to launch an app in Europe, which do you pick today, i.e. which is closest to critical mass or has the greatest potential to achieve that?

what do we have out there…Symbian Series 60/70/80/UIQ, MS Smartphone/Pocket PC, Palm, J2ME, Brew…others I’m sure.

I’ve noticed that when people show me apps they have developed, they are usually Symbian Series 60 or J2ME but I’ve got no basis for saying that they are the most popular or even which is the greater of the 2.

Presuming resources are limited, which do you do to get greatest coverage?

rgds,

Paul

good question .. [ Ajit Jaokar ] [ 10-Jun-05 8:43pm ] [ edit ]but depends on the application. First impressions – SMS, WAP and J2ME. followed by Symbian

But it all depends on which application.

Doesn’t … [ William Volk ] [ 11-Jun-05 4:29am ] [ edit ]Java allow you to write the app and run it on any platform? That’s what the kind man from Sun told me.

:) [ Ajit Jaokar ] [ 11-Jun-05 4:57am ] [ edit ]

write once run anywhere is for the textbooks

I always find it amusing but Sun dont make that into a claim nowadays(at least for mobile)

But then they also famoulsly said ‘We are the dot in the dot-com’ ..

Another claim I have not heard for a long time!

My point is … [ William Volk ] [ 11-Jun-05 5:16am ] [ edit ]

No easy solutions. If you can do it effectively with SMS, go for it.

Java’s advantage is a richer applications and more rapid response.

I think the incompatibility issues will help Flash gain share.

Sounds familiar doesn’t it?

flash .. interesting .. [ Ajit Jaokar ] [ 11-Jun-05 5:25am ] [ edit ]

Do you think something like bitflash ?

write once develop on mobile? [ Paul Buckley ] [ 11-Jun-05 8:15am ] [ edit ]so I’m no J2ME developer (can you tell ) but I had at least hoped that if you were a Java developer and had say experience in a server environment, that moving into the mobile world should be fairly straight forward.

Now there should be LOADS of Java developers around so in my mind that should make J2ME quite a desirable platform? There are IDEs out there, like Netbeans Mobility that make it really simple…dont they?

How about flash/bitflash – is this already loaded onto some handsets? Any applications already launched using this?

rgds,

Paul

j2me flash etc .. [ Ajit Jaokar ] [ 11-Jun-05 8:24am ] [ edit ]

the problem with J2ME is not the technology but the fragmentation at an implementation level. To be fair Sun does not control all of it(operators do). Thus, you could write a game once but porting it to many different operator platforms, gaming sites etc is a major hassle. In fact, porting alone is a BIG business(mostly outsourced to the likes of India)

flash .. I am not sure .. william suggested this .. so keen to hear from him

bitflash has been around for a while but has competitors

Test Test [ Richard Spence ] [ 11-Jun-05 8:58am ] [ edit ]

I would echo Ajits comments .. it is tough development environment but as Tom Hume may well add “if it was easy then everyone would be doing it” ;-)

It makes you chukkle really when you remember having to worry about four or five browser implementations back in the late 90′s now we have to have a WURFL.

See http://wurfl.sourceforge.net/backgroundinfo.php

Coverage [ Tom Weiss ] [ 11-Jun-05 9:30am ] [ edit ]

There are 100s of millions of Java and Brew handsets in the market but only 10s of millions of Symbian handsets.

However, native Symbian apps to me tend to have a better user experience so it’s a difficult call for me in terms of which way to go. There is a good chance that Flash will become the standardised platform for user interface development which would be a great step forward but of course there will still need to be lower level code to work closely with the network.

t.

Tom Weiss

It’s Deja-Vu All Over Again [ William Volk ] [ 11-Jun-05 12:52pm ] [ edit ]In 2000 I founded a ASP (Application Services Provider) now known as ZipProof. Java applet used for content proofing (markup) of simple ads and print jobs.

In 2001 we actually exhibited in Sun’s booth at the old Internet World show. The programmer had to spend the night before the show changing the applet so it would run on the Solaris version of Netscape.

Back to the present. I don’t see why there can’t be a certification process and a test suite for the J2ME implimentations.

The ‘sandbox’ model of Midlets make them an ideal mobile programming platform, but that advantage is lost when we find ourselves having to create different versions for even handsets in the Series 60 family. (For example the famous 6600 won’t support signed-midlet issue).

Since Sun now can say that Java represents the clear leader in installed platforms, why wouldn’t they want to make a start at tackling these issues.

As to Flash-lite … if only it was Flash 5 compliant.

Question: What is the scripting language for bitflash?

Flash vs Java vs SMS vs WAP vs Everything else [ Tom Hume ] [ 11-Jun-05 1:25pm ] [ edit ]I’m a big fan of Flash, Macromedia are a client of ours and I think it could be an excellent way of doing mobile apps… but there are problems with running mobile flash on restricted devices, just as there are with phones. It may be way better visually, but under the hood it’s still a virtual machine running in a very limited environment.

I suspect that once we start seeing a groundswell of mobile Flash apps (there have been a few commercial launches but mostly they’re proof-of-concepts), it’ll be apparent that going down the Flash route doesn’t let you magically avoid all the problems that on-phone development raises. Right now it feels like we’re comparing the gritty reality of Java development with what Flash *promises* to deliver… which isn’t a fair comparison.

The rule I can see with all this stuff is the better-distributed the platform, the less capable it is. So the relatively minor development platforms (e.g. Symbian – which is massively successful but still minor compared to say WAP or SMS) are best for doing apps targeting specific niche markets.

IMHO Java is just at the point where it’s starting to be everywhere. There are lots of problems developing cross-handset Java apps, and you need to be pragmatic: targeting, say, the top 75% of the market and working with operators or other delivery partners to identify what handset the top 75% are using.

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

so what went wrong with J2ME certfication? [ Paul Buckley ] [ 11-Jun-05 1:51pm ] [ edit ]William,

There already is a certification process for all Java specification developed through the JCP.

This is defined by the Reference Implementation (RI) and Technology Compatibility Kit (TCK) which is required for each JSR. The RI is code that implements the spec, and the TCK is the test suite.

J2ME is no different.CLDC is JSR139 and MIDP is JSR 118 for example. The RIs ad TCKs are both there on the home pages.

So what went wrong. Did the handset manufacturers just sort of forget to certify and confirm to the spec, purposely ignored it or ??

rgds,

Paul

This must be … [ William Volk ] [ 11-Jun-05 2:16pm ] [ edit ]A definitiion of certification that I was previously unaware of.

I’ve posted on this before, there are major incompatabilities even within a family of handsets. What’s more some of the manufactuers won’t even support developers or provide documentation on the extensions … for example camera phone operation.

See: Fixing portability

Tricky tricky tricky tricky [ Tom Hume ] [ 11-Jun-05 2:23pm ] [ edit ]

Plus event if you can write-once run-anywhere, you often don’t want to.

For instance, a game might work radically differently on a portrait-sized screen to how it will work on a square screen. Levels for a platform game might have to be redesigned for smaller (or larger) screen sizes. You might want to stick to underlying conventions of the handset (e.g. on motorolas the left softkey tends to cancel, on nokias it confirms).

Differences in J2ME implementations are a big problem, that’s true – and developer sort is poor (but I suspect that like healthcare spending or network operator customer care, it’s impossible to ever spend enough and get it right).

But the underlying issue is that creating a unified platform for apps which runs on a huge range of mass-market, limited devices is *really hard to do*. And if you look at where we are now wrt the handset as a platform for applications, and where we were 5 years ago… there’s been a hell of a lot of progress.

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

All true [ William Volk ] [ 11-Jun-05 2:44pm ] [ edit ]

But if you could write a app to the spec, no extensions, and be assured it would run resonably well across MIDP platforms … it would be a good start.

Right now there’s a huge opportunity for Adobe (Macromedia) to do it right. It will be interesting.

mobile devices RADE.. [ Nick Meese ] [ 11-Jun-05 3:03pm ] [ edit ]

Diff tools for different apps – standalone apps on devices can be achieved with Java apps. When it comes to client server apps on devices – There are some interesting tools out there from www.Momote.com. Sat Forms (used to be Intellisync RADE but now belongs to Thacker technologies) is very good for Palm and PPC – build the app and target the device or OS (no Symbian or Java). www.appforge.com claim to support palm, ppc, smartphone, PC and Symbian devices and permutations of lack of local filtering hinders some palm developers for the time being.

Nick Meese ….Informed mobility

REBOL [ Alex Kerr ] [ 12-Jun-05 5:39pm ] [ edit ]

I don’t know if many in the mobile space are familiar with the REBOL (www.rebol.com) platform/language or have tried it, but I believe REBOL running on mobile would be an absolutely ideal marriage made in heaven, to put it mildly It has all the characteristics that offer the solution we all wish J2ME was on mobile, and more. Briefly – very small, powerful platform; very easy to program; makes everything from advanced network enabled apps to whizz-bang graphics and GUI’s child’s play; extremely small apps, truly write once, run anywhere.

It’s available on many desktop and server platforms, but not yet on mobile. I’m trying to convince them to change this

Yes… [ William Volk ] [ 12-Jun-05 7:10pm ] [ edit ]

I know the REBOL folks from my Amiga days (I was a co-founder of Aegis Development) tell Carl I said hello. Very clever stuff.

There are a lot of REALLY TINY languages that could have been better choices than J2ME. Heck J2ME should have been better than J2ME.

In the late 1980′s/early 1990′s we had a Scheme-like OOP scripting language who’s kernal was about 20kb in size (compiled C code). Totally “sandbox-ed” we used it on adventure games and multimedia titles. Started as a way of porting HyperCard titles to the PC.

Measure how many J2ME instructions are being executed per second on a typical phone. It’s a SOBERING number.

Well… [ Alex Kerr ] [ 13-Jun-05 1:03am ] [ edit ]

My thoughts were along the lines of we clearly need some sort of solution to the J2ME fragmentation problem, something that encompasses as many of the existing handsets out there as possible many of which won’t get upgraded for months or a year or two, something simple and elegant that sat on top of and/or beside J2ME on the handset (given that you can’t replace it) and smoothed over all the inconsistencies and offered a completely level playing field again.

REBOL’s done it on the PC in a technically very elegant way. I guess my question is why can’t it be done on mobile? Is it feasible? I’d like someone to pursue it until we get a definite no, because the alternative is to keep plugging away at all the parties over the coming months and years to get J2ME defragmented (as it were) and then wait for it to slowly permeate the existing user base out there via the standard phone upgrade cycle. I’d say this is a process that would take several years realistically? In the meantime developers and users suffer the consequences of the existing situation.

And yes, if not REBOL then something that fits the bill. Yes the whole stack would be technically inelegant (phone hardware, phone OS, J2ME, REBOL, user app), and yes it could be faster if you didn’t have all the layers, but a phone’s not a server or a workstation, and most have not-half-bad CPUs these days, and how fast does a phone have to be? I think the rewards more than compensate for the tradeoffs, until J2ME gets consistent across the market. Oh, and it would be optional and free for users, just as REBOL runtime is now – if the user wants the app they download and install once the REBOL runtime, then run the app. If they don’t want the app, no worries, no one’s forcing it on them. That way REBOL (or whatever) becomes your abstraction layer for your app to smooth over the inconsistencies (ignoring all it’s other advantages for a moment).

Alternatively, someone writes an abstraction layer as a J2ME app that offers the standard J2ME APIs as it’s APIs, thus effectively becoming invisible but removing all the J2ME inconsistencies across handsets. Now there’s a market opportunity!

Off topic – Amiga gets my vote as greatest digital device ever made. Those were the days

It’s not about the lingo… [ Tom Hume ] [ 13-Jun-05 3:07pm ] [ edit ]

I suspect that it’s not really an issue of which language you go for. If you have tens of different manufacturers implementing a standard, each on tens of devices over a 5-year period you’re going to get inconsistencies and problems, whatever.

William: yes, Macromedia have an opportunity here, but from what I’ve heard on various lists the same old memory-shortage etc. problems rear their heads. They’re doing what’s fundamentally the same thing: implementing a virtual machine on top of lots of different sorts of restrictive hardware. Now, if it’s just MM implementing this then there’s a better hope for consistency between implementations (and consistent QA of these implementations)… but I’m not sure how one company will manage to deploy across every handset out there.

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

My Vote: [ William Volk ] [ 13-Jun-05 5:21pm ] [ edit ]

Sun should build compatibility test suites so phones can be ‘certified’ as compliant. There could be a sticker etc to tell users that their phone could run these standard apps ..

The rule would be that an app written to these specs would run on these phones unmodifed. It might not take advantage of everything, but it would be a way of supporting many handsets quickly.

As I mentioned before, we were faced with the same problems with CD-ROM drivers in the late 1980′s, we (Activision) actually created a test suite that we shipped with product and sent to vendors to get everyone in line. Within a year, the issues were resolved.

I suspect that Macromedia will only focus on the high end of the market. I don’t expect a Series 40 version of Flash lite.

No hope for REBOL unless it gets into the native code of the phone. Running on top of Java would be a bad idea, the JVM’s are quite slow, and with some of the memory restrictions (i.e. 56kb for a JAR on Chinese Series 40 Nokias) you just can’t do it.

It would be easy to use the Symbian kit to build a Series 60 REBOL, but looking at the REBOL site I see that haven’t gotten to OS-X yet.

Certification [ Tom Hume ] [ 13-Jun-05 8:54pm ] [ edit ]

I presume there is some certification required before a handset can advertise itself as Java-compatible or use the Java logo… but yep, certification would be good – presuming that Sun had the bandwidth to do it (how many Java handsets get released nowadays anyway) etc.

If Macromedia only go after the high end of the market, then there’s still a good few years where Java is the only option for the low-end…. i.e. we have to learn to live with it.

Where’s your CD-ROM post? I’ve had a search but can’t find it

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

RE: Certification [ William Volk ] [ 14-Jun-05 12:53am ] [ edit ]

I presume there is some certification required before a handset can advertise itself as Java-compatible or use the Java logo,

Not much testing as far as I can tell. Stuff as basic as what sort of array boundary conditions will throw exceptions isn’t even reliable.

Aren’t we underestimating Flash Lite’s influence already? [ Walter Adamson ] [ 14-Jun-05 12:58am ] [ edit ]Under the Yankee Group Microscope:

NTT DoCoMo’s Experience with Flash Lite

Flash Lite has become a de-facto standard for most large Japanese

mobile internet content providers. Approximately 15 million NTT DoCoMo subscribers have Flash-enabled devices.

Sophisticated information services using Flash Lite have characterized the first wave of innovation using Flash Lite. But Flash Lite extends readily to the critical task of developing dynamic, highly customizable user interfaces. The animated graphical user interface simplifies complex search operations, eliminating the need to navigate through a series of hierarchical text menus.

• We estimate that at least 15 million DoCoMo subscribers (approximately 35% of total i-mode subscribers) will have Flash-Lite-enabled phones by year-end 2004.

• More than 60% of the 4,400 official i-mode sites are based on Flash Lite.

• Approximately 20% of the (88,000:wja) unofficial commercial i-mode sites are Flash-enabled.

- Walter

www.imodestrategy.com

www.digitalinvestor.com.au

Link to my post on Java issues [ William Volk ] [ 14-Jun-05 1:00am ] [ edit ]See:

http://wireless.ecademy.com/module.php?mod=club&op=forum&c=24&t=165746&m=187903

that’s sad [ Paul Buckley ] [ 14-Jun-05 8:53am ] [ edit ]hmmm, that’s very sad William, a) that the problem even exists at all and b), that in the 6 month when that was discussed, nothing apparently has changed.

rgds,

Paul

Open standards and monoliths. [ Tom Hume ] [ 14-Jun-05 9:01am ] [ edit ]Walter: is it coincidence that the markets which seem to enjoy fewer problems of compatibility amongst handsets are the ones which tend most towards being monolithic?

I-mode: specified and tightly controlled by DoCoMo, who specify and QA handsets and content services. BREW: a similar situation, run by Qualcomm.

Perhaps these problems are a natural consequence of our preference for open standards; that interoperability problems arise from this.

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

Open Standards? [ William Volk ] [ 14-Jun-05 6:04pm ] [ edit ]

No, I will lay the blame at the MIDP spec and the lack of testing suites. This is basic stuff.

The intial MIDP spec basically was an invitation to fragmentation. As an example the keypress issue alone was handled badly, with no standardization of game control inputs (software).

The BREW folks do complain about some fragmentation as well, but obviously having one vendor (as with Flash Lite) enforces better standards compliance.

I just am amazed that Sun doesn’t treat this as a major issue. To have the majority of a market and to be aware of this and not launch at least a series of test suites seems incredable to me.

so let’s raise the issue at JavaOne [ Paul Buckley ] [ 14-Jun-05 8:01pm ] [ edit ]JavaOne is coming up soon. Richard Dallaway is attending and has offered to raise any issues (demonstrable bugs) we could suggest with the J2ME team.

So let’s give him some ammunition! Minimal code test cases which can be independently reproduced would be ideal.

rgds,

Paul

Two examples for you… [ William Volk ] [ 15-Jun-05 12:22am ] [ edit ]1. Character array out of bounds exceptions behave differently on different phones. According to the J2ME standard, the String.substring(int beginIndex, int endIndex) method throws an IndexOutOfBoundsException “if the begin index is negative, or end index is larger than the length of this String object, or beginIndex is larger than endIndex.” This works in practice on the Nokia 3650. However, on the Motorola i95cl the IndexOutOfBoundsException is thrown when beginIndex is EQUAL TO OR LARGER than endIndex. This does not conform the standard and causes code to run differently on the two phones.

2. HTTP I/O issues. Standard Java tricks such as closing an http connection to force a timeout don’t work on most devices. You have to monitor I/O processes to kill threads when they freeze.

How to report a bug [ Richard Dallaway ] [ 15-Jun-05 8:23am ] [ edit ]

To demonstrate a problem I’d need:

1. The minimal code that exhibits the problem. Ideally as a JAR file and build script.

2. A statement of how to reproduce the problem.

3. A statement of what happens (screen shots, debugging output)

4. A statement of what you think should happen.

5. The reference (document, page or section number) in the appropriate specification that tells you that the behaviour you expect is mandated.

6. Description of the platform(s) where this occurs.

7. A note of where you have raised the issue already (if at all) e.g., with the handset vendor.

8. Depending on the issue or the device, I may need to borrow the handset.

And it’s probably best to cc: it from this thread to my email address, because I don’t always check here as often as I’d like to.

Sorry, I can’t take just text statements of an issue: it needs to be something I can reproduce with your code. William’s first example is pretty close to a good example, I think.

My feeling is that if something is broken we should be reporting it and getting it fixed in the conformance tests.

Cheers

Richard

I suggest [ William Volk ] [ 15-Jun-05 2:51pm ] [ edit ]That Sun or someone go to this fourm:

http://www.microjava.com/

And ask for the above.

I am also pretty disappointed that my array example isn’t ‘good’ enough for them. That’s a ‘winning’ attitude for sure.

Do they WANT to repeat their browser experence on the handset?

AMAZING.

Tom, you’re right [ Walter Adamson ] [ 17-Jun-05 5:40am ] [ edit ]That’s one of the benefits that DoCoMo sells other carriers on when promoting i-mode. And it makes a developer’s life more simple.

The control of the handset standards and the publication of what each handset and its series or level supports is a key to reducing development and testing costs. However even within this i-mode partnership it is not entirely straightforward and different levels of compatibility and features can be time consuming for developers trying to work out their greatest spread of deployment across the i-mode alliance.

It’s a good system, but not necessarily the only one, for example a more rigorous “compatibility lab” approach might be a good alternative, so that the standards are set by a carrier but without forcing or influencing or choosing handset makers. Each maker is free to compete and offer their handsets as long as they meet the compatibility standard, and the carrier would provide resources in the “compatibility lab” to help the handset makers.

The benefit of this approach is that the costs should be less, for everyone, both the carrier and the handset provider. For example the handset provider can make a globally optimal commercial and technical decision about handsets in relation to the compatibility requirement, and build that into their global handset strategies. This is in contrast to the DoCoMo model where the handset makers have to make locally optimal decisions, that is on a local node (i-mode) rather than a global optimal.

The sub-optimal nature of the local decision-making is evident in the fact that Nokia and Motorola have no DoCoMo i-mode presence, or rather have had none up to now but have been enticed to enter the fray by DoCoMo investing in their handset development in order to compensate for the sub-optimal decision to supply i-mode. DoCoMo has decided that the cost of enticing them in is going to pay off by forcing down the price of handsets from their other i-mode handset providers in the local market, whose prices in turn have been pushed up by having to conform to the i-mode requirements. It’s an interesting circle, or more than interesting it is somewhat deadly to the local manufacturers like Sanyo who are losing money hand over fist.

In ths case above DoCoMo is indeed biting the hand that fed it, locally.

Also to blame for this intense pressure on DoCoMo over pricing is the absurd and destructive practice of handset subsidies. These all came about because carriers are hooked and incapable of selling anything except technology and features and iron. When life became just plain too hard for the so-called marketing and the sales groups, when they had to think about NON-priced based competition, and no longer selling iron, they closed ranks and came up with masterpieces like handset subsidies and bundling. These are even more deadly sins, which just paper over the fundamental issues in the short-term. They look great for the first 2 or 3 years, and by then the carriers have dug themselves into an even bigger hole which requires even more radical solutions.

But of course none of these people are stupid. They know. They just can’t give up – these pseudo-cost-based methods are drugs, and the top-line that they bring is also tied to the annual bonus. Because they know is why carriers like DoCoMo are in grave fear of new entrants in Japan like eAccess/eMobile. eMobile is going to follow the compatabilty lab approach – an open market with best global optimum prices from handset makers. No subsidies. On paper eMobile is a feeble small competitor, with very modest subscriber goals, not enough to threaten DoCoMo – on paper. But in fact this attack on the structure of the industry and the previous lazy habits and the hole that the sales people have dug for their companies is hugely significant, and threatening for the telcos hooked on the old models.

And don’t forget, the people behind eMobile are the same people who founded KDDI, who is eating DoCoMo’s lunch every day in Japan. KDDI has taken control of the innovation branding and the youth market, and DoCoMo is in severe catch-up mode, especially while having to convert subscribers from their 2G MOVA network to the 3G FOMA network without them switching to KDDI-au. So guess what – more handset subsidies!

A key outcome of eMobile’s strategy is that it is going to have very competitive (maybe even low!) handset prices and a wide range of handsets to offer – on day one. And subsidies, if any, will be an internal choice of the handset providers. Such decisions will hit the handset makers bottom line, not eMobile’s. This is a flying start to a mobile business, in contrast to some of the slow launches around the world due to high handset prices and limited choice. The compatibility lab will ensure some degree of acceptable user experience, as well as network compatibility.

So coming right back to your question, the i-mode alliance model is the most successful of its kind to date in providing developers with a group of assured handset standards. But it is not necessarily the only model that will work effectively, and watch out for eMobile in 2006 as a test of an alternative.

- Walter

www.imodestrategy.com

www.digitalinvestor.com.au

Another approach [ William Volk ] [ 17-Jun-05 4:46pm ] [ edit ]

Would be to have a 3rd party act as a certification lab for handset companies that wanted a “seal of approval.” These handsets would have access to a larger application pool as apps written to a standard would run on them unmodified.

From the trenches… Maybe UK operators are “gettng it at last” [ Simon Cavill ] [ 17-Jun-05 10:57pm ] [ edit ]As someone who has done a lot of mobile handset development work, I have found the biggest handicap is actually testing applications across a wide range of different handsets. In the past, most operators were extremely unhelpful in this area although we have found Orange UK to be very proactive in this space. We joined their developer scheme sometime ago and attended their developer training event in France last year which was extremely helpful in getting in from of the right people at Orange.

More importantaly they provide access to their labs were you can turn up with an application and test it on every handset they ship at one go. This means you can build and test apps on most of the current handsets on sale across the Uk and most of Europe as many operators share the same devices. I think they deserve a lot of kudos for doing this and will hopefully reap the rewards from increased WAP/GPRS traffic as well as download revenues. I wish other operators would take a note out of their book….

Simon

DoJa enables games developers to create… [ Walter Adamson ] [ 21-Jun-05 8:30am ] [ edit ](O2’s i-mode chief architect, Jag Minhas)… claims that building an i-mode version of an existing HTML site can take as little as 10 minutes. Crucially he also praised the i-mode version of Java for handsets which is known as DoJa.

He reckons DoJa enables games developers to create one version of a game than runs on all i-mode handsets. By contrast some games available on O2’s existing WAP portal [Active] have no fewer than 60 different variants.

The Enquirer 17 June

- Walter

www.imodestrategy.com

www.digitalinvestor.com.au

? [ Tom Hume ] [ 21-Jun-05 11:26pm ] [ edit ]How many handsets is “all I-mode handsets”?

Future Platforms Ltd

www.futureplatforms.com

e: Tom.Hume@futureplatforms.com

t: +44 (0) 870 0055924

m: +44 (0) 7971 781422

Re: which mobile dev platform for greatest coverage? [ Aju kuriakose ] [ 22-Jun-05 5:30am ] [ edit ]hi,

I know lot of messages are being exchanged and lot of experts speaking on the topic. Although I am not an expert but as a person who has been marketing for a company which has a focus on mobile devlopment applications also here are my comments.

The key to choosing your platform is what you want to develop and the deployment strategy on the wireless platfrom. There are various options avaliable

(a) SMS – for low commplexty low data entry applications. SMS development does not need much of technical knowledge from the developer communicaty as there are lot of ready msde toold which will easily plug in to an existing solution. But deployment becomes achallenge some times as you would have to run behnd the operators to get the shot code to run the application. However if you are talking about Enterprise based application it could be done using simple GSM Modems.

(b) J2ME- Although most of the mobilephone and PDAs can run j2me Application you would need to do minor rework on the application to suit it to a particular devices like Palm, symbian, pcoket pc etc . becasue of multiple reasons like VMs being used, Screen size etc.. But the rework involved is less. limitation al the native APIs and functionality of the devices might not be readily avalaible to j2me . Easy to deploy as there is no operator dependency

(c) Device specific/platfrom specific tools like metrowerks for symbain or appforge etc. This tools are good if you plan to reap full advantage of a specific device or platfrom. for example if you plan to make a application which take advantage of the mutli language capability of symbian or Camera functionality etc. it is alway better to do the programming on specific platfroms rather than using the generic platfroms.

I can throw some more light and also open to tlak to you more if you need more inputs from my end

Regards

Aju