Recently, I wrote a piece on my own blog discussing the question of whether today’s electonics products would look as clunky in 20 years as those of 20 years ago look today. My thought was that they probably wouldn’t, due to the superior quality of the design of many of today’s products. This spurred a rather lively follow up discussion, which particularly focused on mobile phone design. As it happened, after this discussion I discovered that I had lots more to say on both this subject and the question of just what features are and are not important to mobile phone users, and how the devices are evolving. People who read on can discover what exactly is I have to say, and will also get a bonus explanation of the meaning of this photograph of Perry.
The company that figured this out was of course Nokia, who produced phones with beautiful design and a simple user interface, which were for their times very small. However, their most successful phones were not packed with features. One of their most successful phones was the Nokia 8210. When first released, it was the smallest phone anyone had ever seen. It was simple, elegant, and had essentially no features other than voice calls and SMS messaging. More features would make the phone bigger, and Nokia knew that was bad. As it turned out, the fashion conscious portion of the customer base, which was most of the customer base, was not interested in any features beyond this so the phone was a huge success. Although at that time there was a huge fuss made about multimedia services (generally named WAP) this was all hype and no substance, because in its early forms WAP quite simply didn’t work).
Of Nokia’s two main competitors up until a couple of years ago, Motorola made technically advanced phones which were packed with features and hard to use, and Ericsson made technologically advanced phones which were ugly. Neither of these strategies were especially successful, although Motorola was dominant in the market segment of “business users wanting feature rich phones that worked throughout the world”. Unfortunately for them, this didn’t turn out to be the most lucrative segment of the market for the phone manufacturers. Although these users spent a lot of money on their phone bills, this money went to the network operators, and although they bought phones with an expensive retail price, these phones were also expensive to make. Plus, these users didn’t upgrade quite as often as the more fashion conscious.
Which was where we were a couple of years ago. If you had the smallest Nokia, it meant you were with it. If you had some other clunkier looking phone, or you had a phone of any kind that was physically larger, it meant you were not with it. Nokia built an amazing brand. The brand was so successful that people actually believed that Nokia were a company delivering great engineering as well. (I have a female friend who until recently had a Nokia phone because Nokia were obviously the best. However, this phone would not work in the US. Therefore, when she went to the US she had to borrow a Motorola from someone else. Despite this, when I told her that Nokia phones were generally less technically advanced than their competitors, she flat refused to believe me, despite the fact that her Nokia was not adequate for her own personal use).
However, two things have since changed. Firstly, as Eldan pointed out in the comments to my previous post, phones had reached the point where they were about as small as they could get. The Nokia 8210 and its ilk were fine if you had small fingers (which in practice often meant if you were a teenager or were female, or both), but for many people a slightly larger phone was better. Even for people with small fingers, an even smaller phone would not be useful. Secondly, Nokia’s competitors started to catch up with respect to size and design and the number of competitors in the market increased.
Last year I twice needed to get a new mobile phone, in the first case because my phone stopped working, and in the second because my phone was stolen. In both cases I was not able to get a handset subsidy on a new phone, and as I am presently in a situation where I have to be careful with my money I went looking for phones that were cheap and functional. In the middle of the year, the cheapest possible phone looked like this: it weighed 149 grams, and was 132 mm long: sufficiently large and clunky that it was slightly embarassing to pull it out of my pocket amongst a group of young people with cooler phones. At the end of the year, the cheapest possible phone looked like this: small, and quite elegant (105mm and 95 grams). When I pull it out of my pocket, people sometimes ask what it is, but there is no embarassment factor at all. (In fact, Sendo are a British company, and this is an interesting talking point about the phone). The user interface is quite nice, too. The phone can be bought outright without a subsidy for 40 pounds. If Sendo can sell me a phone for this price, it becomes clear just how large Nokia’s margins must occasionally be, given that they can sell phones for six or seven times the price.
Clearly, other manufacturers (even low end manufacturers) are catching up with Nokia in terms of design, which is potentially very bad for Nokia. Commoditisation appears imminent. So what happens?
Well, the terms of the battle change. Rather than competition being principally about design, we suddenly start competing on features. Colour screens become a big deal. Rather than just a small number of GSM phones being tri-band and working in both Europe and North America, this feature is becoming standard, and even Nokia feel they need to offer it. Most importantly, for the first time since widespread use of SMS, it seems that there appear to be new features that mobile phone users want to use. Rather than competing on who can offer the best form factor, the form factor is settled and we are competing on who can offer the best feature set in that form factor.
As for what features are being added to phones, it seems to be a matter of what other gadgets can be merged into a mobile phone without affecting the form factor. Many of us would find a digital camera a useful thing to carry around with us, but not to such an extent that we would go to the trouble of placing it in our pocket every morning. However, merge it into our mobile phone that we are going to carry around anyway, and then suddenly we have it with us all the time and it is worth paying a little extra for the phone. One application of such cameras is for photographing what we are seeing and sending it to your friends as a way of describing how you are feeling – that is as an extension of the SMS experience. Another is simply whatever you would use a non-phone digital camera for. See something that is worth photographing, and you can take a picture. This is particularly useful if you are a blogger with a tendency to write about what you see from day to day. For instance, if you are sitting in a bar with Gabriel Syme and Perry de Havilland, you can get Gabriel to take a silly photo of Perry for later use. (Sadly, Gabriel had the camera and thus I do not have any photos of the marvel of humanity that is Gabriel. There were also some photos which show that I am much better looking than Perry, but alas Gabriel has not forwarded them to me).
Similarly, although the number of people who regularly carry a GameBoy around with them is relatively small, the number of people who would play games if they were bored and found themselves with one is quite large. At least some mobile phones have had a few simple games built into them for some years now – Nokia were again the innovator – but the types of games that can be played has been improved dramatically with the addition of higher resolution colour screens, as this New York Times article from last week discusses. (Link via slashdot). Plus, more sophisticated data capabilities on the phone network (plus more commonality in phone operating systems) make the download of new games possible. Ultimately we will also be playing networked games.
The impact of these trends has become quite dramatic in the design of new phones that have been released this year. Most notably, we have the latest state of the art Nokia phone that the company is selling to its best customers: the Nokia 7250. This phone is something that 18 months ago would have been almost oxymoronic: a feature packed Nokia phone. Tri-band GPRS. Colour screen. Built in camera. Downloadable games. And a partridge in a pear tree.
While still small, the phone is significantly larger and heavier than the previous couple of generations of flagship Nokia phones. (105mm, 92g – very close to the same as my 40 pound Sendo, and compared to 101mm and 79 g for the Nokia 8210). The mobile phone companies are no longer really competing on size, although there is some competition on form factor (some companies seem to prefer flip phones). What we instead have is a single pocket size electronic device that most of us will carry around with us at all times. This will be configurable with respect to colour and shape, but its size and form factor are now essentially settled (until some really compelling new application that is worth paying for with some inconvenience in terms of size comes along, anyway). What this device will be doing for the next few years is slowly sucking in new features.
(It is also interesting to observe that Nokia really does appear to realise that the race is now about features. It introduced its new form factor with the 7210 late last year, and has since introduced new phones with the same form factor but more features twice: firstly the 7250 and now the 7250i. It wasn’t long ago that upgrades like this were purely cosmetic. Now they are very feature oriented).
And just as a final aside, does this mean that the much vaunted “3G” mobile phones are now with us? Well, on the whole, no. The additional data speeds that 3G makes possible are only really necessary for real time interactive applications: video streaming and the like. Where the data being sent is either low bandwidth or non-realtime, the data speeds possible from existing 2G networks are more than adequate. The four companies with 2G networks in the UK are offering all these new services without the expensive business of building their 3G networks, although they continue to pay large sums of money to the government for the right to do so. A fifth company, Hutchison Telecom (trading under the stupid name ‘3’) has rolled out a 3G network. At the moment, it is selling one service that the other networks are not offering, which is video telephone calls. However, demand for video telephone calls appears to be small, as AT&T discovered in the 1970s when it spent immense amounts of money developing the technology for fixed line phones. Hutchison have very few customers. (They also have the disadvantage of not being able to offer the most popular mobile handsets, which are still being made for the much larger 2G market. The compelling application for 3G is not here yet. My best guess is that it may arrive with networked multi-user games, but this is still some time off).
In any event, as someone who is much more feature oriented than appearance oriented, I find that for the first time in several years I am looking at mobile phones and getting a serious case of geek envy. I want one of these.
One of the huge factors shifting the market now is a move away from the old “one company makes all” approach and towards OEMing of devices.
This means a lot more players with unbadged product and a lot more interesting designs coming like Nokia Series 60 platforms and Microsoft Smart Phones.
Lots of changes ahead.
Opps… pressed POST too soon.
What this means is the Operators will be bringing out more operator branded phones in the GSM world.
This gives them much more power and aims to drive up ARPU (Average Revenue per User) by cunningly adding data services right onto the platform so you don’t notice spending the money.
Be warned – capitalism in action!
Once the market commoditizes and the platform becomes more or less standard I’m expecting the open source world to move in.
Linux on your mobile anyone? Roll your own strong crypto over wireless perhaps?
It’s going to happen eventually.
I am running a 2-3 yr old Motorola Timeport that I love. The display is clear and easy to read. The buttons are big for my large fingers. It sends & receives email and SMS. It works everywhere (Verizon where I live has the best coverage). All the doodads that are coming out on phones are of zero interest to me. I don’t care about taking pictures. I don’t care about games or what other scheisse they can deliver to my phone.
I want a phone that is easy to use and works all the time. Period. Some of the newer Nokia crap, like the 3650 appear to be nothing more than some designer’s masturbation than a practical phone. And I’ll be damned if I let a Palm or Micro$oft OS attempt to run my phone (it’s just more stuff to break…the Kyocera 7135 “Smartphone” comes to mind).
And no, I’m no luddite, either. I’m a sr. Unix systems engineer for a VERY large telco here in the US.
Because of the telco’s business model history they are very pay per transaction oriented, and as they control delivery of mobile services, these services are pay-per-transaction or some bundled variant. I am not aware of any technical reason for this, with always-on networks (at least with 3G) and always-on mobile devices (phones).
In contrast, the buiness model for the internet, was originated by ISPs, who were rarely telcos is mostly un-metred access excepting the telco dialup connection part if you use one.
In my view, the next big thing in the mobile phone space will be always-on applications, particularly in the personal security space. Tracking where your child, or rather senile grandfather, is. Other applications are personal panic buttons and always- on security devices in your home talking to your mobile phone.
Some may be concerned about loss of privacy. I worry more about government restrictions on my right to use these technologies as and when I see fit.
I would want a total of two features out of a cellphone: good sound quality and long battery life. I’ve seen phones with good battery life, but the sound quality seems universally poor. Of course all this is based upon occasionally borrowing others’ phones–I don’t talk on the phone enough to buy one…
Dave: Operator branded phones are something the GSM operators have tried to establish pretty much since they launched a decade ago, but they have never really been able to make the idea stick. Instead we have had people become very brand conscious about their phone but caring little if at all about the network they use. Along with handset subsidies and a rapid upgrade cycle (driven by the fact that most people don’t pay upfront for their phone), this has meant that a large portion of many people’s monthly bills have in fact ended up going to Nokia or Motorola rather than the operator. The operators have always wanted to break this cycle, and network branded phones have been part of this. Nokia in particular have gone to great trouble to keep the present state of affairs (through a lot of brand building advertising and loyalty schemes and the like). It remains to be seen whether this will change now. As you say, operators now want to sell phones with network specific services built into them, but this is a hard thing to do if customers walk into a store and ask specifically for a Nokia 7250i.
There is something of a shift now, as the more intensive and thus revenue generating services which impress the kiddies, are no condusive to non-operator branded phones.
Hence all the O2Activ and Vodafone Live stuff.
Actually, most of the reasons why the service providers want to switch to “3G” don’t have anything to do directly with the ability to offer new features to customers. It’s actually because the 3G systems all utilize spectrum far more efficiently, which means the service provider can support more customers for a given allocation of bandwidth.
All the 3G systems use some variant of CDMA, which is much better at this than the TDMA system used by GSM. But the way that a CDMA system is designed affects this, and there’s a gain even for the upgrade from IS-95 CDMA to CDMA2K.
But it’s really dramatic for the upgrade from GSM-TDMA to UMTS (which is CDMA), because it’s going to approximately triple the effective number of calls that can be carried per unit spectrum per cell. It’s difficult to quantify the gain; it depends on circumstances. It isn’t all that easy to quantify the actual capacity of a CDMA system, for one thing since it has what’s known as “soft capacity”. The actual gain can be anywhere from 2:1 to as much as 5:1, but 3:1 is pretty typical. CDMA2K and UMTS are about equally good at utilization of spectrum.
As a side effect of the change to 3G, though, it will give them the ability to dynamically allocate a huge amount of bandwidth to a single user on an as-needed basis (with the allocation potentially changing on the order of 50 times per second). That’s something which was available to a limited extent with IS-95, and which is pretty much fully supported by both CDMA2K and UMTS, but which was pretty much impossible with GSM-TDMA. But that’s really a secondary feature from the point of view of the service providers. Relieving the problem of saturation of spectrum is the big gain for them, but about the only way most users will notice this is a better success rate at originating calls and a lower miss rate on receiving calls, and possibly a slight decrease in service rates.
This upgrade wasn’t really motivated by the lure of being able to offer new features to customers.
This is true, (and as I have mentioned before, I am actually a much bigger fan of IS-95/CDMA2000 than GSM/UMTS) but it isn’t especially relevant in the UK at the moment. There are actually commercial reasons why the one UMTS operators that has launched is emphasising new features, and network capacity issues aren’t really a problem at the moment. The situation is that there are four GSM operators (Vodafone, O2, T-Mobile and Orange).
In a very cleverly designed auction intended to maximise the price paid, the British government auctioned five UMTS licences. Each of the four existing operators felt obliged to buy one (otherwise they might find themself at a competitive disadvantage compared to the other operators – in justifying the huge outlay of money, the operators did a lot of ranting at the time about the great multimedia future in front of this, although they may or may not have actually believed this) and the various other people that wanted to enter the UK mobile market ended up fighting for the fifth (but the auction was designed in such a way that the incumbents had to pay essentially the same amount as the new entrant). Hutchison (‘3’) ended up with the fifth licence.
Having won the 3G licences, the incumbents found themselves with mountains of debt and an environment where raising additional capital was difficult, so none of them felt any great urge to actually build a UMTS network unless their competitors did first. On the other hand, Hutchison had no existing business with which to generate income to finance the debt, it had to build a UMTS network to have a business at all. So it did. It wasn’t a matter of an upgrade so much as simply building a network.
Having built its UMTS network, Hutchison then found itself at a competitive disadvantage due to the fact that all the most popular handsets are made for GSM. Therefore, it launched an advertising campaign based on features that its competition couldn’t offer, in particular video phone services. This advertising was very prominent and must have cost a lot of money, but Hutchison have only signed up a tiny number of customers. I am told that they are about to launch a new campaign based on relatively simple price competition. However, given that I am still hearing reports that the UMTS handsets do not work especially well (or at all), I am not expecting to see much success soon.
And as for the need to alleviate network congestion, I don’t see much at the moment, at least not in this market. I use a GSM mobile in London all the time, and I seldom find myself unable to complete calls or having my calls drop out. (When I used a GSM mobile in Sydney three or four years ago, it was much worse). The British mobile market has reached saturation – pretty much anybody who could conceivably want a mobile phone has one. Voice traffic is still rising, but relatively slowly, as it is due to people making more calls rather than the number of phones on the network increasing. The present GSM network seems adequate for voice calls. It may be that to get this level of reliability the operators have had to build a great number of base stations with lots of tiny cells, but having reached this level of reliability, I doubt the capital costs of maintining it are enormous. And the capital costs of rolling out UMTS are large.
As I said in the post, additional services at the moment seem to be mostly non-realtime, multimedia message service (MMS) related. These are not terribly demanding on the network, as finding some spare network capacity to send the message some time in the next half hour is much easier than finding capacity for some real time interaction right now.
The incumbents are all obliged by the terms of their liecences to built UMTS networks eventually (although the government is ignoring this requirement, as the last thing anyone wants is to send any of them into bankruptcy), and they are watching the Hutchison situation with interest. However, I tend to think that they rather wish the whole 3G situation would just go away. It’s not so much that capacity constraints have forced the upgrade as that nothing is forcing the upgrade, but they have spent a lot of money committing themselves to doing it anyway. As it is, they are sitting around hoping for a compelling new application.
(Certainly there are some markets where capacity constraints on 2G networks did force the move towards 3G. A lack of capacity on NTT DoCoMo’s PDC network in Japan is the classic example, and I am sure there are more in Asia. However, I don’t think this is so much the case in Europe).
There’s a lot more issues here than just about regulation.
These are not terribly demanding on the network, as finding some spare network capacity to send the message some time in the next half hour is much easier than finding capacity for some real time interaction right now.
Actually, MMS is quite demanding on network resources. Its not like SMS, on a 2.5G network, the phone must establish a data connection over GPRS to the MMS-C and then download the message data to the MMS-C. At the other end, when somebody receives the notification of an MMS, they have to then initialise a GPRS session for the message download. All relatively time consuming and bandwidth consuming – especially for little video messages.
What the operators want from any network is dramatically increased ARPU (Average Revenue Per User) – the only way in the current market is through more cunning use of data and other applications, reverse billing SMS and the like. 3G doesn’t necessarily help them do this anymore than on a 2.5G network.
The real fight is going to be for the roaming boradband data only customers and I have a feeling that 3G might loose that one.
True WCDMA doesn’t work as advertised at the moment. Neither did CDMA at first. Biggest problems are the cell handovers from 2G to 3G and data traffic handovers. This was always going to be a complete nightmare, but its a requirement that the GSM and WCDMA networks interoperate – it would be a problem regardless of how the upgrade was done and one which will go away over time.
Reading Michael’s piece on CDMA vs GSM many things occur to me, but this is the wrong forum to discuss them. GSM isn’t just Europe. I travel all over the world on business. With the exception of Japan and Korea, my phone works when I get off the plane, even in the US now I have a choice of GSM networks. That’s where GSM wins hands down – the roaming options are excellent.
Ultimately WCDMA will take over – but I don’t expect to see much of a rush to adopt CDMA2K over GPRS, particuarly when the P Channel stuff starts working.
Actually, MMS is quite demanding on network resources. Its not like SMS, on a 2.5G network, the phone must establish a data connection over GPRS to
the MMS-C and then download the message data to the MMS-C. At the other end, when somebody receives the notification of an MMS, they have to then initialise a GPRS session for the message download. All relatively time consuming and bandwidth consuming – especially for little video messages.
Okay, thanks. Video messages are going to use a lot of bandwith however you do them. (I am not sure they were what I was talking about though. I am not sure there is much demand for them, and the GSM operators are not emphasising them at the moment. Is this for network capacity reasons, lack of capable handsets, or what?)
Still, though, latency must make a difference. It must be possible to wait for times of less network congestion before sending the message, and it must also be possible to send the message at a lower data rate than is necessary if you are sending it real time, which would mean the message could use fewer timeslots (although obviously it would tie them for longer). If you reserve (say) one timeslot for MMS, then everybody’s MMS message is going to get there eventually (although it may take a while). On the other hand, with voice if you run out of timeslots then that is that. This is a pretty fundamental difference.
I would expect that people are working on clever software for this kind of optimisation (although I don’t know this).
The real fight is going to be for the roaming broadband data only customers and I have a feeling that 3G might loose that one.
For the operators’ sake, one can hope. It remains to be seen how big that market is, and I suspect this depends (a lot) on pricing. (It would also be interested to see how big this market is in places that do have CDMA2K 1X data services going (places like Japan, Korea, and New Zealand).
True WCDMA doesn’t work as advertised at the moment. Neither did CDMA at first. Biggest problems are the cell handovers from 2G to 3G and data traffic handovers. This was always going to be a complete nightmare, but its a requirement that the GSM and WCDMA networks interoperate – it would be
a problem regardless of how the upgrade was done and one which will go away over time.
I am sure it eventually will. The question is when. (Hutchison are no doubt doing the operators a favour by trying to run a network and get the bugs out, even if it is a disaster for themselves). The one proviso I do have is that I have been hearing that UMTS will work properly soon for a very long time now – several years. Getting the bugs out appears to have been a big deal.
Reading Michael’s piece on CDMA vs GSM many things occur to me, but this is the wrong forum to discuss them. GSM isn’t just Europe. I travel all
over the world on business. With the exception of Japan and Korea, my phone works when I get off the plane, even in the US now I have a choice of GSM
networks. That’s where GSM wins hands down – the roaming options are excellent.
That’s a very fair point. International roaming was one thing that GSM operators really did get right. And getting smooth roaming between those makets that do have CDMA is something the CDMA operators have failed to get right. (I would love to hear the story about why the uplink and downlink frequences are reversed in Japanese CDMA compared to the rest of the world some time). The issue is whether GSM’s dominance will be carried forward into UMTS. And I still do not believe this issue is resolved.
That piece is nine months old, and was a bit of a rant. I linked to it partly to remind Mr Den Beste that I was the same person who wrote it. I still stand by my basic points, but I would now be a little more moderate about them. Although we have drifted a bit from the topic of my original post, you are certainly welcome to contest what I said here if you want to. (I think we may be about to drop off the Samizdata from page, however).
The problem with GPRS and CDMA2K is at the end of the day they aren’t quite yet, true data solutions. You have to set up a connection and do the data transfer. This is a pain and something 3G solves. However, by the start of 2004 most GSM operators plan to have GPRS supporting the same. At the moment, there’s no way to have all this happening in the background, which is something of a huge Pain in the A***.
“3” are doing the whole industry a favour right at the moment by trying to solve the WCDMA problems, but many of these were blindingly obvious from day one and will eventually be got around. I do recall having a meeting with a well known Scandanavian manufacturer in 1999 where they told me that they’d have 3G services up and running in 2000. I thought it a bit insane at the time.
The issue is whether GSM’s dominance will be carried forward into UMTS. And I still do not believe this issue is resolved.
Its not.
However, the GSM operators proved pretty good at getting a global pick up for GSM and, more or less, getting the devices to interwork. I can’t see them letting that do now there is an upgrade path to true 3G which, at the end of the day, will eventually work.
For a regular traveller the ability to turn on the phone at an airport and start working is invaluable and its something I think the former pure CDMA operators failed to grasp.
Whether data services work is another question. Certainly they may fail against Wi-FI networks which is an interesting case study my wife is doing for her MBA project.
What is this? Metric in a Samizdata posting?
I come from a part of the Anglosphere (Australia) that uses metric units pretty much exclusively. I am only vaguely familiar with British units, and I am not especially comfortable with them. In this case, though, I was just copying from the manufacturers datasheets, and they game metric units – even Sendo, which is actually a British company.
Like it or not the UK is a metric country and has been since the 1970’s.
Sadly I belonged to the lost generation who didn’t get taught to use any system well until I started my A Levels.
Even in my engineering degree, the age of our equipment meant I still had to remember how many thousands of an inch were in a milimetre…
Its time to move on.
The weird thing about Britain is the way it is half one system and half the other. You buy petrol in litres, but road signs are in miles. To quote Buffy, “I’d like to think I speak for everyone here when I say ‘Huh’?”
If the government was going to try to convert the country to metric units, you would think the least they could have done is manage to get the whole government to go along with it (such as the people who erect road signs) but no.
It is odd, I’ll agree.
Ireland is still one of the most infuriating where some distances are km’s, others are miles, and speeds are in MPH.
?