Tuesday, November 27, 2007
The World’s best phone?
The spectacular title suggest I have found the definite phone. How can I make a claim that there is a best phone, when we very much know that the phone is a mass market object of desire, many flowers can bloom. Here is my logic. Phones are voice communication devices, they are primarily used for voice, most of the cost is incurred from talking, so my argument is the phone that provides cheapest talk is the worlds best phone.
According to my analysis, the cheapest calls can be made using the Skypephone from 3. It is a small 3G phone with a crisp colour screen and simple user interface, very Nokia style, priced at GBP49.90. It offers free Skype calls anywhere, also when roaming in a 3 sister networks, so if you take holidays in Italy or Sweden for example this is perfect. Furthermore for GBP25 a month you can add International call saver, a sort of flatfee to call to 35 countries. I checked the fine print and realized that many mobile area codes are excluded, one seems to be able to call US broadly also mobile. Nevertheless a stunning deal. If you then top up more than the required GBP10/month to keep the service running you get call cost locally down to 3p a minute or 3p per text.
If you have family, friends and business spread over the world this is a stellar deal, and hence I claim it is the best phone.
I have used Fring, Gizmo, Truphone, Jajah on my mobiles, but the problem with them is that they either require data or wifi, which consume more energy than cellular, and the N95 for example cannot sustain that for long enough. There is no major magic here, the stuff they have done on this phone could be done on any platform, but it is the total package that makes it interesting and revolutionary.
The Skypephone makes a phonecall into 3 infra, where the call is routed to the Skype infra, where it leverages their VOIP technology. This to me is a best of both worlds. One gets the bandwith and energy efficient 3G call and the cost efficient VOIP call. A solution requires deep integration and dedication on both sides, one could even call it a partnership ;-)
The business logic is as follows. The user need to top up atleast 10GBP per month to keep the service running, for this the user gets some minutes that can be used to make calls 12p per minute/txt. Hence 3 will make atleast GBP120 from the user. Skype could get a small piece of this maybe 30%. With the International Flat fee, for GBP25 per month That adds up to GBP470 per year, a lot of money, but far less than the typical GBP100/Month I pay to have service from T-Mobile, where I am very very careful not to make international calls. The low price point will attract a lot of new customers for 3, they will get a taster for the service and if they are happy they will start to top up more and once you have your international deal you are hooked, the difference to the other providers are so stark. When then the Skype network effect kicks in we will really see the power of this partnership. This means one needs to wait about 6M to judge the success of this handset. The service that came closest are Vodafone with lots of minutes, passport and a business line from T-Mobile where one can use the bundle to make international calls. Vodafone is better when roaming, but it is really not an option.
What 3 does not offer is an ability to use the phone as a modem for data, which I have from my T-Mobile deal.
What would make this even better? First of all, I really would like 3 to offer 6M to 12M contracts. I would like to have it on a high end Nokia, which would not be tied to a contract. I like to keep handset and service separate, I switch phones all the time. OK, that is me, not the consumers en masse....
Congrats to the Hutch and the Skype teams. Really good job guys!
Tuesday, November 13, 2007
Android – not so Alien among mobile platforms
The first impression was positive, the user experience was a Yahoo Go 2.0 meets the Blackberry Pearl UI. Nicer notifications than iPhone, cleaner core navigation than iPhone. No softkeys, more fluid application switching. Clearly less alien, than I had hoped.
What I did not see was two things I have been hoping that would be innovated:
1. A new menu system, essentially removing the concept of applications. Toward task centricity and content centricity. See my post on the Transformer OS.
2. Powerful uber search as a navigation paradigm (think super QIX, just to point you in the right direction).
The keypad was very predictable, 5-way, back, Options, Home. (There is a funny cross on the image, don't know what it does.) They could have done with a more stylish keypad by integrated back and home, now they will not get balanced keypad designs.
That said, they are like any other mobile UI, just marginally better, but still same paradigm, one of menus.
Where they had innovated was in backstepping, notifications, and device2device / network2device communication.
They have a fluid backstepping model that allows the user to backstep accross applications. This means it is true backstepping rather than application silo backstepping we have been used to in phones. It is a good improvement as the goal is to create seamless web and native apps. I wonder does the system go to the network in the backstepping, for example if the backstepping involves a map app with GPS, if the device location has changed does it try to update and redraw the new tiles it downloaded from the network or not, or backstep directly. I assume there is some timer. This is a critical things to be wary of as most networks will charge at the attempt to make a connection. I have personally spent 120 Euro on 2.667 due to these kind of little data packets, was some error. This places in experience I call seams and in mobile experiences it is crucial to get the seams designed right.
The notification framework is interesting and its transparency is nice. Rich notification is crucial in the recommendation culture we are living in with services like Twitter. Again how to prioritize them, toggle them. Who decides the order, last served is not always right?
What sounded very interesting was the device2device communication through their XMPP Service which allows any applications to send device to device data messages to other android users. This is very potential enabling interesting P2P experiences.
What I was waiting for was some word on security and extensibility of the user experience.
Listening to the videos on YouTube (about 30min), the word security was not mentioned a single time. I guess that is good. I am not too fond of app signing as I see that being a control point rather than insurance for the user I hope developers will not go after this openness, just because it is Google that did this platform.
What still makes me most skeptical about Android is the user experience control model. I am worried if Google controls it and I am worried if they let the operators control it. Both are suboptimal, the latter is terrible.
We have to remember that it is one of the biggest if not the only neutralizer in addition to brand the handset vendors and Msoft have to fight against total control of the carriers is the UI. The battle for the UI in the handsets has been one of the more serious technology battles in the past 10 years and due to intense competition we (consumers) are better off, only thing we have had to do is to learn a bit, which is not bad for the brain (I know some mobile UI's are worse than others, but that is just what competition is all about.). If Google plays their cards right things are good, if they play it wrong, an even stronger hand lands in the hands of the operators, and then, what can I say, the system won...the androids helped the system win.
If I look at the mobile business from Google's point of view, I really think android is a good bet. Creating GMail and Maps apps alone is not going to make them relevant in mobile. Search is still up for grabs, but one really needs to re-engineer the back-end for mobile the web results have little relevance in my mind, but even when that is done, I am not sure search alone would make much of a dent in the mobile universe.
So Googlers, next you have to change the game in wireless access! Bid, don't bend. This is in my opinion your social responsibility as the elected representatives and bureaucrats converged towards greed. Remember those 3G auctions.
Tuesday, November 06, 2007
Do Androids dream of Electric Sheep...
Do Androids dream of Electric Sheep was the Title of Philip K. Dick's book which was the basis for the cultfilm Blade Runner. The plot, where Harrison Ford playing Android hunter Dackart, is trying to spot Androids among people. Faxes from real. I do not know why I woke up with this thought and thinking of Google's play in the mobile phone operating system market. It this is serious effort or a fake one?
I am generally pleased to see that more players are moving into this space as competition stimulates innovation, which benefits the end users. Mobile phone operating systems are one the one hand extremly sophisticated and on the other hand very primitive. With primitive I means as they are such gigant SW efforts of hundreds of people they evolve slowly, looking like they are stuck in the past. Remember that Nokia makes 10000 permutations of their phones. See this interesting piece at Business Week.
Linux will change the game, in the way that it will replace the core part of the SW, and shift the innovation to a higher level. When it in the past took 4-5 years to build an mobile operating system it now could take much less, maybe 2 years or so. The effort is probably also reduced from several hundreds of people to maybe a hundred to make a phone, less if you do not want the differentiation. Again in that time one does not make a customisable platform only a phone. If one wants to sell it to lots of players it needs enless amount of customisation and I wonder is Google going to do that? Maybe they can hire some Androids to code these boring customisation features making them run in the 800+ mobile networks, all wanting to have differentiated handsets one way or another.
The big question for me is: who will control the user experience? It seems to me that the operators now get SW assets that will allow them to control it further, hence possible everyone looses, as they will promote their own device and having even more control than when these SW assets were not controlled by them. Quoting the Open Handset Alliance Q&A"Why is an open platform good for mobile operators? The overall cost of handsets will be lower and mobile operators will have complete flexibility to customize and differentiate their product lines. Furthermore, they will see much more rapid innovation in handsets and in services." This is a dangerous thing for the end users, it gives more power to the operators. If the UI is controlled by Google, and they dictate the reference UI then things get interesting. They can do that, by controlling the tools and by controlling the licensing agreement under which this is open sourced. If Qualcomm would take this asset, add it on top of their chipsets, add UI One, you suddenly can get the whole thing from Qualcomm, hmmm. pretty interesting.
It will be interesting to see how the open source community reacts, one of the lessons from the MobileCamp London where the creators of the OpenMoko phone platform was present, was that Linux will enable increased fragmentation in the mobile application market. This is one place where Google can add some value, they can provide the reference and maybe have the momentum to see it prevail. But then it is the GPhone. My qestion is who will raise to become the champion of the open mobile UI, as that is the real next battle ground. Google just made that a battle of more players not less. In democracy a leader is still needed.
Another thing which is really interesting is monetisation. It is not free to have several hundred of SW engineers working on something. The Google guys are not dumb, once there is 'Google' software in thousands of phone models and millions of phones, it could in a magical way turn itself very Google friendly and 'whopsee' very Yahoo unfriendly, it could auto update with new Google apps, it could feed Google profile information, now polling at system level rather than from an app silo in Java or Symbian as present.
What Google gets is a client server architecture deployed, which is crucial when dealing with mobility, as good data coverage is never as good as good voice coverage. For example the phonebook needs to be native on the device or the sim card rather than in the sky. Yes, it also needs to be in the sky for auto updating etc.
So in the end I ask myself, can one trust an Android, atleast I need to consult my open source friends and some lawyers to help me read their TOS, before I resort to retina scans, like Dackard in Blade Runner...back to sleep, back to dreaming...