Everything and the Mobile Software Universe…

Great read from TomSoft about Mobile Ajax:
TomSoft » Myths of mobile Web2.0 (and mobile Ajax).

Really in line with what Thomas said, especially this point :

…Ajax applications will run the same on mobiles than on PC, and this will save us some porting costs. Wrong! Seems that the Write Once Run Anywhere myth is back!! It was actually already not achievable through technology designed for this, so I did not see how Ajax app…

Same for me … the more I work in this industry with european, chineese,Korean partners the more I see:

• How skilled they are
• NO UNIFORMITY : for each client, and even for each phone for each clients, many things are different and can’t be abstracted generalizing a concept.
• Requirements are from Operators (Carriers), Graphic Designers etc… and no one has the same approach for UI and data representation…

A good technology should help this diversity and free the creativity of those actors… wrong idea to lock them in one scheme, one way to do … how will they differentiate?
Customization is key, and is not at all limited to theme, colors and image, but of the whole software …this is our vision here at open-plug

Thanks moPocket for the video!

Ok it Seems a little bit SciFi for now, but after all look at the VK 2020 or the latest korean “card size” phone are not so distant (ok appart the flexibility ).
But what is even more surprising is the software depicted in the video : it is adapting to the user and any context … so far from our little applications and their static icones, a BIG step further compared to the previous post about live UI.

:: mobiface :: next gen mobile interface thoughts  : Wow great idea! The background image is going more cloudy when the network strength is going down…neat!

Today, in the cellphone space (as it was for the PC years ago), two main applications/services launching UI paradigms are fighting for dominance:

• Icones that launch applications : All architectures needs a kind of “glue” or shell, called explorer on Windows. This is the path choosen by high level OS : great for dynamic downloads, new applications. Look at you windows/linux Desktop, your PalmOS explorer , your s60 or pocketPC one…same concepts of text/icon point and click : very poor for services interactions, UI consistency.
• Ad hoc UI : the “Menu” Application in many proprietary OS  : great for integration, with user interaction fine tuning, ad hoc customization by operator, wrong for code maintenance and dynamique expandability.

Read the rest of this entry »

found here : bLaugh » Archive » Cell Phone Overkill

Really this is happening…our customer (phone manufacturers) are asking for features that even us, as geeky as we are, don’t know how to use! (found again via the Kathy Sierra: http://headrush.typepad.com/creating_passionate_users/ )

This shows also that voice is and continue to be the “killer app”…of a phone (how bizarre how bizarre). The industry is truggling to find a new one (camera is a good idea, I like the concept of “personal data gatherer”).

My guess is that a “one fit all user device” (as in the PC land) is really simply irrelevant : see the latest SonyEricsson offering, see how many segments you have for cell phones, opposed to the PC market:

  Read the rest of this entry »

Here is a high level presentation of the mobile handset industry channel….from what I’ve understood, and from my own Open-Plug/Texas Instrument perspective.
This is by no way a yet-again-prediction-about-the-mobile-phones-market: no I won’t say here that the next big thing are Ultra Low Cost handsets …more than very high end smartphones

First the main actors of this drama (ok, I do my best to mimic the gfx of the great schemas from the Kathy Sierra’s blog):

In nearly square boxes, the companies that are part of the flow, from the handset idea till the real device in customer hands:

• OEM (Original Equipment Manufacturer: wikipedia definition): In this industry the OEM term is perverted a little, the OEM are the companies with Brand Name (in fact value added reseller or VAR), like Nokia, Sony-Ericsson, Siemens, Philips etc. More and more they are outsourcing handset design to ODMs and production to an EMS (that can be the selected ODM itself).
• ODM (Original Design Manufacturer: wikipedia definition): Design House that are technically creating the device, based on OEM requirements, or in other models directly from operators requirements, or from their own to sell their design to OEM/operators. Well known ODM are HTC, Compal, Wiz4com, Cellon, Foxconn, BenQ (Chinese and Taiwanese companies for the most).
• EMS (Electronics Manufacturing Services: wikipedia definition): The factories that are actually building the devices…and and we are in a 400 millions units market… those are BIG factories.
• Operator/Carrier (did you say telco?): Another kind of beast for the handset part of the mobile industry equation. They can be either the key driver for handset specification (see Docomo), or a simple distributor.
• Distributor : retail shops that can sell “unbranded/unlocked” headsets that can be used with any service provider (operator), or directly resell operator branded phones with a service plan.

On the supplier part of the flow:

• Platform Vendors : Here is the main hardware supplier of the chain (Ok I’ve intentionally omitted all the zillion of different other hardware suppliers involved: flash, screen, boards, plastic, etc), the one that sells the heart of an headset, its chipset. This chip is comprised of the CPU and the associated RF/Analog baseband/Digital Baseband. In this area the undisputed leader is Texas Instrument (used a lot by Nokia, Sony-Ericsson, Motorola and many ODM), followed by Philips Semi Conductor (main client: Samsung, and many other OEM/ODM). Many other platforms like ADI, Infineon, etc. are also represented. Nearly all those platforms are now backed by an ARM CPU.

And now the most interesting (hehe, ok at least for me) SOFTWARE! which is now the undisputed … bottleneck of the industry. Traditionally the Platform Vendor were integrating all the needed parts of the software…without making any money from it. But with the explosion of the number of functionalities, software actors are becoming more and more prevalent.

• Integration House: A first answer to those integration issues was to outsource it to other companies with large and dedicated software resources.
• IP vendors: We spoke about features…Here are the backers of those features. Need a Java VM? asks Esmertec, a wap browser? Openwave, Jataayu, Teleca, An predictive input text method? AOL/T9 or Izi.
• OS/FWK vendors : The hardware part being so heteroclite, and the time to market so hampered by software IP integration, this new kind of players are emerging, becoming a key enabler for IP vendors by reducing their integration and porting efforts, thus the time to market. On the OS side, Symbian, Microsoft and Palm are the leaders of the … little market of the smartphones (forecast are, in the best case, around 15% of the market). On the framework side (for all the other phones) key players here are TTPcom, OpenWave, Teleca (obigo) plus some new comers like SkyMobile Media … and of course Open-Plug .

So after the presentation of the actors…here are their interactions!

We will present here some of the main channels that are best describing the process of designing and selling a mobile phone.

• OEM oriented: the OEM is pushing its ideas/designs till the distribution channel.
• ODM oriented: ODM are selling their designs to OEMs.
• Operator oriented: Operator is defining the device design/requirements.

• A: The “historical” channel. An OEM is making its devices form A to Z, from design to production, using only platform vendors for their silicon factories, providing their own hardware design and not using “standard” platforms. Nokia is perhaps one of the last of its kind (but recently outsourced some handset to Foxlink if I remember well).
• B: An OEM specifies a handset and then go shopping to get an ODM which will be able to design the handset an produce it (perhaps via an EMS).
• C: In that case an ODM designs an handset and sells the design to an OEM that will simply stamps its brand on it (HTC is working like that).

In those channels the operator is only seen as a distribution channel, only able to validate or not an handset. Anyway more and more Carriers are asking for branding and customization…and it’s not so easy to convince Nokia to replace its brand by a Vodaphone or Orange one.

• D: Very similar to B, but this time the carrier specifies itself the handset and is feeling strong enough to get ride of and replace a big handset brand name (OEM), see the HTC operator branded phones (orange, O2).
• E: The Operator is “fully” specifying the handset, like in D but will associate its branding with an OEM one (that itself will most of the time outsource the design and production to an ODM/EMS).

Keeping all that in mind, a next post will describe the software flow and what are the impact and implications of those channels, the eco-systems around hardware platforms, and the various software strategies deployed in the industry

This great post asks the good question, “does mobile OS matters?”.

A solid, robust OS is only a tool to achieve “function”: ok it helps to achieve that, but do you know the OS in your iPod/Blackberry/phone/car etc? Having a big OS HAS drawbacks, like CPU overload, memory consumption … for nothing related to the device “function”. The only real advantage is that an OS offers a common developpement framework, that is a clear abstraction to the hardware and allows easy developement for a wide range of devices … but after all is that the role of an OS? Take a look at Java: it offers such a development abstraction, and relegates the underlying OS to a mere commodity (hum, … ok sort of), it also has a BIG drawback : this hardware abstraction is too strong/big/too generic for mobile devices (performance issues, used RAM and above all access to device hardware functions). After all, what is “really” needed for mobile device?

1. Perform main function perfectly, with no compromise to second order functions: your phone has to be a great phone even if it also does PIM….
2. COST.
3. Design/Ease of use/ fun, again Never underestimate the power of fun.
4. Time to market.
5. Higher and higher function/features complexity, WITH NO SACRIFICE to the first, second and third rules.
6. Did I say COST?

To satisfy all, you need to make compromises (like market segmentation in function and cost).

So coming back with my point of interest: Mobile phone industry. Open Plug has been founded around this idea: “only Developpement tools and APIs matters”, and above all, as said in the article “only device function matters”.
…and this industry (devices, telepony…mobility) is so big that the today’s big player won’t allow a complete industry stole ala microsoft: the big guys there are Cell phone operators, phone manufacturer, content providers…
Nowadays software developement became the undisputed bottelneck (along with mass production) when building a cellphone. To help with that (and take the lion share of a billion device market), software houses are pushing high level OS adoption in cellphones: see efforts from microsoft, symbian, palm …. anyway it adds software and hardware cost (bigger CPU, even adding CPUs, more RAM, bigger battery) and in many case it really lowers ease of use, and the first rule is many times sacrificied.
Linux may be a great free alternative BUT Linux alone is only a kernel, you can’t do any application cause there is no FRAMEWORK on top of it: no sound, graphic, telephony handling, etc. You definitively need something: Motorolla has tried to do that, Trolltech (Qt/Qtopia) is doing that, and more recently PalmSource is doing that. This last, surprising move, really shows that the main PalmOS advantage was its APIs on top of which many applications have been written, on top of which you could really make a mobile phone with coooool features (see my treo650, ok there cost is fairly high). The underlying OS (who know the kada kernel???) was really a commodity, they even tried to make a new kernel but ultimatelly failed before selecting Linux: the kernel war is over, the big players are there, too late to compete.
Here at Open-Plug we have a similar approach than Palm: building a framework on top of …. a subset of the linux API and not really on Linux itself, so all the framework may run on Linux (for High-End Phones) or, porting this Linux API subset on top of low-end/ proprietary OS/platfrom. The whole thing, coupled with an efficent component based technology and powerfull development tools may seems very similar to java in fact … but in that case:

• No virtual machine, so no CPU penalty
• Using ASM/C/C++ : maximum RAM/CPU efficiency (if well coded )
• Direct access to harware
• Remotely deploy solutions on previously “closed” platforms.
• Use existing well tuned proprietary OS if needed, so no additional hardware cost to the platform.
• Allows a great customization by the players (Mobile Operator, OEM, ODM) that really NEED AND WANT to offer services/function: see how it is difficult to ask symbian, and so Nokia to add a little vodaphone live logo, not mentionning a complete user interface definition for a certain kind of device like full operator branding, phone specialization like music phone, kids phone, etc.

Sounds too good to be true ? Hey, ok I’m biaised and enthusiatic about what we are building here, but we are making progress and you will see our phones soon .