News Blog

In the last blog of the year, I'll share my thoughts regarding the new year. As the overall market is expected to return to a growth path after a recession period, we are likely to see an interesting year ahead of us. I have divided the analysis into three main areas: product, business and technology. After publishing this blog entry, I am also starting my Christmas holiday (see the picture on the right how it looks outside). Therefore, I would like to wish you all a very relaxing Christmas time, and all the best for the new year!

Product: Although a very large part of the mobile business is in the low-end products, the high-end phones always tend to get the largest publicity. There is no question what the single most iconic mobile phone is: the Apple iPhone. Many "iPhone killers" have emerged during this year, but none of those has quite reached the same level of user experience. Now, my crystal ball says that we will see a new iconic product during 2010. It is very unlikely that someone would take over by simply copying; therefore, the next iconic product will not be a copy of the iPhone, but it will provide some other functionality that truly separates it from the crowd. There exist already numerous ideas and concepts, mainly related to location-based services and social networking; what is missing is clever user-centric design to implement these in a cool and highly usable fashion.

Business: App stores have clearly been a big trend in 2009, and some top-notch developers have also gained respectable profits. Apple's App Store has passed the 2 B download milestone, and many analysts believe the business is getting close to $ 1 B mark. We have witnessed the first success stories of ad-based business models - but also shifts in strategic direction like the ad-funded MVNO Blyk did. My expectation for 2010 is that we will see the first example of a totally new business model for phones. Nothing large-scale yet, but a serious start. Google is a usual suspect for this: they have developed or acquired many important building blocks during this year. My own guess for such a business model is a new type of revenue share deal to combine operator subsidy and ad-funded content and applications.

Technology: If you have been following my blog during 2009, it is probably not a big surprise that my favorite technology for 2010 is mobile cloud computing. It will replace many stand-alone apps in the mobile phone, especially those that use heavily other cloud resources - like location-based services and remote teamwork do. The base technology is here today, although some extensions could easily enhance the capabilities further. Simple updates and bug fixes will speed up the innovation cycle. Not all apps will move to the cloud, though. Also, providing non-free apps is conceptually more challenging than with the stand-alone apps model. Here, one solution is to tie the services to an operator-hosted app store, and use operator billing as the payment method.

No comments yet - write one!

Continuing in the spirit of my previous blogs, I decided to write an iPhone-optimized version of the Darteco site. My target was to make it look as much as possible like a native iPhone app. I went through a set of libraries, and found jQTouch. This is a jQuery extension, using a combination of CSS styling and JavaScript actions to "iphonize" web apps. JQTouch turned out to be very easy to use, and to integrate with my application framework based on Ruby on Rails. It took just one day of work to create the iphone.darteco.com site (I'll need to tune the whole system later so that iPhone users will be automatically redirected to this site). And yes, I think it looks very much like an iPhone app, with similar menus and sliding navigation. You can also create an iPhone desktop icon; with that, the native app-like feeling is even stronger, the only difference being that the app is downloaded from the cloud every time it is being used.

So, what were the learnings? With this type of architecture and high-quality developer libraries, we are already entering the era of mobile cloud apps. A user will hardly notice the difference between the cloud approach and a native app. Of course, which alternative is better, depends fully on the type of the app; large local apps do not benefit from this approach, whereas apps that use cloud-based data and are likely to get regular updates gain the biggest advantage. In the case of iPhone, there is also an interesting issue regarding the app store process: mobile cloud apps are totally separate from the process, meaning that they do not need to go through the approval process - obviously, they do not get to be distributed through iTunes, either. An item to monitor is what type of status the mobile cloud apps will gain in the various app stores; will they stay outsize, or will they also get integrated in.

No comments yet - write one!

In order to give an introduction to the technology part, I'll divide the client-side implementations into three main categories: browser-based, other common app -based, and dedicated apps. In all cases, the technology for the server-side implementation can be (almost) freely selected from the available platforms such as J2EE, .NET or Ruby on Rails. Before you read further, try out a demo: with your mobile phone browser, go to mobile.darteco.com, the brand-new mobile-optimized version of this site. Make sure you have JavaScript enabled in your phone - and the site should behave as a very simple cloud app (it works also without JavaScript, but behaves then like a 'normal' web site).

The two big benefits of a browser-based implementation are easy app downloading (one just goes to a web site and the app downloads and starts) and cross-platform compatibility (only browser with JavaScript support is required). The technology base is AJAX (again, a new name for an old thing...), denoting a combination of HTML content, CSS styling, and JavaScript actions and control, operating in an asynchronous manner. In the case of this demo, all the site content is downloaded in one go, and the navigation is done with a couple of lines of JavaScript. After the download, there is actually no communication between the client and the server - a more complex app would obviously use some protocol to move updated content between them. The drawbacks of browser-based implementation are related to the overall software architecture in a mobile phone: often, only one such app can be simultaneously active, and it does not run in the background. Also, more complex apps can be rather slow - this depends on the rendering and JavaScript engines (see my blog on Google OS).

A good example of using a common app for a cloud service is the mobile RSS reader. RSS feeds use a standardized protocol between the client and the server, and as the total size of download content is typically limited by the service, the technology is well suited for mobile cloud apps. Darteco's web site (the main site, not the mobile version) provides the blog entries as an RSS feed, so one can use the RSS reader in the mobile phone to access those. Again, the pros are easy start-up process and compatibility - many phones have an RSS reader, in addition to the browser. The major drawback is the lack of flexibility; RSS feeds are only suitable for certain types of content, such as blog entries or news clips. However, the limits can be stretched a bit: if you go to www.kylansaitti.fi, my Pet Project, you will see that it provides also calendar entries as an RSS feed.

The third option is to use a dedicated client app: services such as Skype or Google Earth use this approach. A dedicated app gives a lot of flexibility in the design, and potentially offers the best user experience. The obvious cons are the same as the pros for the other two options: one needs to find and download the correct app, and a developer needs to port the app for a number of incompatible platforms. Note that the AJAX technology can be used for these apps as well, like e.g. the new Palm WebOS platform does (see my earlier blog) - this is more familiar to developers coming from the Internet world. My forecast is that the installed base of web programming environments (with extensions and libraries to access phone features and peripherals) will start to grow in new phones and platform, effectively blurring the difference between a browser-based implementation and a dedicated app. However, the main decision factor is not the underlying technology; it will be the quality of the development and deployment tools, greatly affecting the time-to-market and the overall cost side of a cloud services business plan.

No comments yet - write one!

This business domain seems to be full of new names that have been invented to describe 'old' things; I recall Larry Ellison saying that "I've always been doing cloud computing" - and giving Salesforce.com and NetSuite as typical examples that have been around for a long time. Yet, more and more apps actually move to the cloud, or at least they get a (limited) cloud version. With the pressure by Google Docs, Microsoft just announced that Office 2010 suite will be available also as a free cloud version, in addition to the traditional installable package.

I can imagine at least three reasons why this paradigm shift is taking place: First, the new apps are different in the sense that the capability to communicate, collect and share distributed data, and co-work has become an integral part of the application concept itself. Second, moving apps to the cloud enables smooth version updates; a good fit with the modern development processes. And third, it is a 'Blue Ocean strategy': cloud has been selected as the medium to challenge existing leaders (Google Docs vs. MS Office is a good example). Thinking in the mobile context, there is also a fourth motivation: mobile platform space is highly fragmented, forcing the developer either to focus on only a subset of the total market, or to invest on costly cross-platform porting. Depending on the application split between device and cloud (more of this in the second part that focuses on technology), the platform-specific part becomes non-existent, or at least smaller.

Are these four reasons relevant to drive the change in the mobile ecosystem? Number one certainly is, as the cloud is required as part of the concept. Number two and four can potentially make the user experience more seamless: less need to worry about finding and downloading the correct software version (this is trivial with single-product platforms such as iPhone but seems to be less trivial for all others), and a service that stays always up-to-date. This, however, also depends on how the device-side of the solution is implemented. Number three we are witnessing e.g. in the competition between Google services and Android, versus Nokia phones and Ovi. From the software development perspective, my own forecast is that we will see more and more mobile apps that have a strong cloud component. The client-side platforms and apps will not diminish, but they will change in nature: user experience - including things like good usability and cool graphics - will continue to be important, but end-to-end integration and web programming environments will replace monolithic implementations in many areas.

No comments yet - write one!

In my earlier blog, presenting the leadership book by Anne-Liisa Palmu-Joronen, I promised to get back to the topic. I spent a few interesting evenings with Gary Hamel's The Future of Management, trying to see if I could find some new matches between that and my own recent thinking. Although I have liked many of the ideas spotted by Palmu-Joronen and a couple of other new books and articles, I must also say that I haven't got the feeling that I truly believe they will solve the challenges faced by today's business leaders; the reason is that they tend to isolate the leadership challenge, without going holistically deep into the companies themselves (like Hamel does). The main theme of The Future of Management is how to re-invent the company and liberate the existing innovation capacity by removing the obstacles created by old-fashioned management style. Here, I will extend this by sharing my vision how to build a future high-tech company - companies based on the vision presented here are certainly not very common today, so let me know if you have seen one.

The vision starts with a small start-up, possessing a novel business idea, and great passion to implement it. Now, what typically happens next (assuming the business idea makes sense), is a VC providing financing (and optionally help to refine the strategy), or a large company acquiring and integrating the small start-up (which many entrepreneurs regard as a nice exit point). Why not having something between these two extremes: a host company providing A) seed financing, B) existing brand, C) global account management, and D) financial and administration services to the start-up, without performing a true integration? This would help the start-up to strengthen and speed up their sales process, and provide it with the crucial financial data and peer comparison, yet it would be able to run the operations with the original small-company feeling and passion. What should not be included in the "host-company services portfolio" are things like strategy, operational processes or HR. When the start-up grows, it gains a larger share of the total revenue over years - or it gets divested as a separate business. In both cases, the key is the original mission and passion, carefully nurtured in the heart of the small company. In this vision, leadership and motivation come from the origins of the start-up, and they should rather be preserved, not re-created later.

No comments yet - write one!

During the last month, two interesting acquisitions or mergers have taken place within the Finnish mobile software outsourcing industry. First, a leading Chinese IT solutions and service company Neusoft Ltd acquired the mobile arm of Finnish Sesca Group, including close to 300 people in Finland and Romania. The remaining part of Sesca Group, employing 300 people, will continue focusing on outsourcing for energy production and process industries. Second, two Finnish companies, Flander and Ardites, joined forces with Chinese Symbio Group. The new company, Symbio, will be have global headquarters in Beijing, San Jose and Tampere (Finland), and a presence in 22 global locations, including five off-shore development centers in mainland China, Taiwan and Bangladesh.

The strategy of Neusoft Ltd looks similar to what several Indian IT outsourcing companies have pursued earlier: expand the business scope from traditional IT outsourcing to mobile domain, and acquire on-shore and near-shore presence and talent to get closer to customers. With more than 15 000 employees, Neusoft already has off-shore scale. As the operations in Finland closely resemble those of Wipro and Sasken, it will be interesting to see how smoothly the integration will proceed in this case.

On the other hand, the new Symbio seems to have higher ambitions what comes to strategic value network. The new company will have special focus in services and solutions for the enterprise domain. With talent pool in mobile development and user experience, and existing enterprise customers such as IBM and Microsoft, Symbio aims to become a leader within this area. It also looks like the on-shore vs off-shore ratio will be higher than with e.g. Neusoft, although Symbio also has a footprint in Bangladesh, one of those potential countries for next-cycle major off-shore expansion. As a summary, Symbio will likely be competing more directly with companies like Digia, Elektrobit and Teleca.

No comments yet - write one!

In my previous blog, I shared some learnings from the development time. Now, have a look at the Internet service itself: www.kylansaitti.fi. As the service is - at least right now - only available in Finnish, I'll give a bit more insight into what it offers, and what the targets are. As I noted earlier, the service may later turn to become a business of its own, but that will be a different story.

For business-oriented people, the simplest description could be "Intranet for real communities". First, it aims to serve non-virtual communities that exist in the real world as well. Typical such communities are small towns or villages, parts of bigger ones, as well as associations and clubs. As small local companies have typically a tight link to their surrounding communities, they are part of the target group as well. Kylänsaitti is mainly serving people who live in the area; However, also other groups are likely to find it useful. These include summer-time visitors, people who consider moving to the area, or people who have moved away and wish to preserve the link to their earlier community. For individual communities, the service offers Intranet-like features such as calendars, event lists, two-way email system, maps, news, discussion forums, and guest books, in addition to static content. As people are often members of multiple communities, they may also utilize the index-page services to easily find all relevant events and news from all of them.

One of the main targets of Kylänsaitti as a service is to find ways to enhance and enlighten the life of real communities. The administrator of a community site may set the edit rights for individual pages, so that the whole community can join in to create the content. Ease of use has been one of the key design criteria, as not many of the users are expected to be experts in Internet applications. The application itself integrates with other open-interface services: Google Calendar for calendar services and Google Maps for mapping and geotagging.

When you go the list of communities in the service, you will see it is very short: we are in the very beginning. Therefore, you will have a chance to witness how this type of service evolves. Will it attract more users and communities? How will the interaction between communities look like? Will the system continue working, if the number of communities keeps growing? Come and visit the site every now and then, and you will see how the pet project progresses forward.

No comments yet - write one!

Since the beginning of Darteco, I have also been working on a personal pet project. My aim is to create an Internet service from scratch, using Ruby on Rails (RoR) as the development platform. My target has been to get a good insight into one of the most modern and versatile web development systems - and to establish a fully functional, professional-looking and useful Internet service in a pilot form. The service itself may later turn to become a business of its own, but that will be a different story. I'll tell more about the service in my next blog; in this one, I'll focus on the project part.

So, let me share with you some of my findings and learning points during the project: Quite some time ago, I used to write lot of code, but that was before the era of web apps and web programming. So when I started my RoR app, I had to get into the fundamentals. Luckily, I came across a book called RailsSpace (by Michael Hartl and Aurelius Prochazka, Addison-Wesley Professional Ruby Series). This book is mainly about the Rails part, yet providing a sufficient level of Ruby language intro as well. I can highly recommend the book: it is easy to follow, and it gives plenty of practical examples on how a typical social networking app works.

I am a long-time user of Apple Mac, so I was happy to discover that Mac's OS X comes with the Rails framework pre-installed. That required just some straightforward updating. The next thing I needed was a code editor or IDE (integrated development environment). I first took Aptana Studio, an Eclipse-based open source environment into use. However, after some time, I found the Mac implementation unnecessarily heavy and somewhat unstable, and I switched to a rather simple Rails editor called TextMate (available for Mac from macromates.com). It's not free, but I think it's well worth the price tag. And with these equipment I started my project - The result, as it looks like right now, you will see within a week.

No comments yet - write one!

Google announced mid-July that they are working on a new lightweight OS, to be bundled with their Chrome browser. The main target product segment is netbook PC, where browsing is a major usage scenario. With an open source -based new OS, Google aims to reduce the cost of netbooks, thus increasing the potential user base for Google services and cloud apps. According to Google, the Android OS will stay on the current trajectory and continue focusing on mobile phones.

When the Chrome browser was released, one of the new key architectural elements was the high-performance JavaScript engine called V8. Chrome also offered better integration with cloud apps, effectively blurring the boundary between locally installed traditional apps and cloud apps - at least from the user experience point of view. Therefore, the new OS announcement seems very logical, emphasizing Google apps strategy based on cloud-hosted apps and data. I would assume Google will continue optimizing the rendering and JavaScript performance of the Chrome & OS combination, as well as further developing the system libraries to improve asynchronous operation (for occasions where a network connection is not available). As Google's Vic Gundotra also stated that the current mobile apps model is a dead end, expect to see the same type of architecture coming later to Android as well - making the development environment similar to that of Palm Pre (see my previous blog).

No comments yet - write one!

The new Palm Pre smartphone has an interesting development platform on top of the Linux-based OS. The core piece is Mojo™ application framework that executes the webOS applications. A webOS application is strikingly similar to any standard web app written using a combination of HTML, CSS and JavaScript - therefore, the potential developer base is very large, compared to platforms that require native programming. The webOS apps run locally (although they may call e.g. cloud services), as opposed to web apps that use the web server - browser paradigm.

The apps themselves follow the popular MVC (model-view-controller) paradigm, providing a separation of app logic, data and visual presentation. In addition to the runtime environment, the Mojo™ app framework includes UI widgets and services that help writing functional and eye-catching apps: The UI widgets are included by simply calling a JavaScript library from the HTML file, and the widget set covers all standard UI elements from simple buttons to more complex widgets such as data pickers. Services are either local phone functions exposed to webOS apps, or remote services accessed through web services interface. On-device storage is provided by high-level HTML functions that access an underlying SQLite database engine. The simplest webOS app contains just three files: appinfo.json file that holds the app metadata, index.html file that holds the actual app, and icon.png that holds the graphical icon for the app launcher.

Palm is currently providing only a limited and controlled access to the SDK, signaling that some parts may still be under development. The tool chain seems to follow a fairly standard approach, including an Eclipse-based plugin. Overall, based on the available information, the architecture and implementation of the framework look clean, solid and expandable; I will therefore eagerly wait to see how the system performs in the finalized form.

Image © Palm Inc.

No comments yet - write one!

Approximately ten years ago, Psion made a conceptually (though certainly not business-wise) revolutionary miniature laptop that they called Netbook. The device had an 8-inch VGA screen, and it was running the EPOC operating system - the baseline-to-come for the Symbian OS. The same name, Netbook, became the synonym for all those small mobile-oriented laptop PCs that started to emerge during last year. As a result, Psion made a claim that the term is a registered trademark they still own, and therefore it cannot be used as a generic marketing term.

What happened next is interesting: instead of the case going to an official court process, it went to a "Google process". Google investigated the claim, and determined that the term Netbook is indeed a protected trademark. Google then informed their advertising customers that the term can no longer be used in the ads. Of course, the effect was not that dramatic, as consumers could still search for "Netbook" and got directed to the respective pages, even though the term was not used in the ads. In this particular case, the trademark owner was a real company with real products - would the case have been different if the trademark owner had been a patent troll or similar? Will this case create a baseline how to handle certain IPR-related disputes in the Internet era?

At the end, the whole case was solved by Psion and Intel entering an agreement early June which allowed the term Netbook being used in marketing, and stopped any further court actions.

2 comments

Several ex-Nokians have decided to write a book. A recent entry to this category comes from Anne-Liisa Palmu-Joronen, who worked as Head of Legal Affairs and member of the Management Team in Nokia Mobile Phones EMEA (1988-2005). The book, written in Finnish, is called "Nokia-vuodet, mitä johtamisesta voi oppia" (could be translated as "My years at Nokia; learnings regarding management and leadership". Translation is mine - the same applies to the rest of the blog as well).

The book heavily criticizes the management by objectives in the form where it tries to create an illusion of infinite growth of performance. As Palmu-Joronen says, the whole system turns to be like "trying to run excellent 100 meters races, one right after another". While the management processes in many companies are highly optimized, there are still huge gaps when it comes to the leadership side. As a solution, the author proposes a concept of "Leadership through trust" ("Luottamusjohtaminen"). The foundation is a two-way agreement between the employee and the superior, creating the ethical promise they both agree to pursue in all their actions. Growing into the new level of leadership requires strong personal self-awareness; The key ingredients of leadership through trust are virtues, and the ethics built on them. Leadership is the primary theme of the book - The part covering Nokia gives little new to the public picture.

In many aspects, I tend to agree with the author: too often, we sincerely believe that our busy calendars prove our efficiency. I believe that human performance is a very complicated thing, and the true maximum potential (with respect to both quality and quantity) of great individuals and teams is seldom reached. What I personally plan to do is to go back to one of my favorite books, Gary Hamel's "The Future of Management", in order to see if the "softer" touch provided by Palmu-Joronen helps to get something new out of that; Hamel talks a lot about trust as well, though the context is slightly different. Therefore, I expect to return to the topic later.

Image © A-L Palmu-Joronen

No comments yet - write one!

Yesterday, Intel announced a definite agreement to acquire Wind River, for approximately $884 M, representing a premium of 44%. The two companies have already worked in partnerships, jointly developing e.g. the Moblin Linux platform for embedded devices. The announcement states that Wind River will continue serving their existing customers and hardware architectures: this means that they will be developing embedded software for many other platform providers than just Intel. Wind River is an active Android partner, helping e.g. Kyocera to develop Android-based phones for CDMA networks. As the announcement also states that the deal "will boost Wind River’s Intel-architecture focused sales", one could guess there will be a shift of strategy regarding the various platforms and platform vendors.

What are the potential implications of the deal? Both Wind River as well as Intel are members of the Open Handset Alliance (the consortium behind Android), so they will supposedly advance the use of Android on Intel platforms - whether this will in certain device segments compete directly with Moblin, remains to be seen. Wind River is also a core member of LiMo Foundation, whereas Intel is not; as this is one more Linux-based embedded platform, we will see whether the deal will lead to any change of strategy in this area. With these Linux consortiums pursuing the fast-growing Netbook market (in addition to phones and MID products), there will be an interesting competition emerging also against the MS Windows platforms.

Image © Wind River

No comments yet - write one!

Mobile app store is not a new innovation - several stores have been active, but prior to Apple iTunes store launch last year, they never enjoyed a wide-spread success. Difficulties in finding and purchasing apps, together with the obvious challenges in interoperability, have greatly limited the customer base. Apple changed all this: With only one device platform (the iPhone and iPod Touch) to support, the latter problem was solved. And then Apple provided the same level of user experience they offer with their other products and services, making the purchasing process a simple task. The total download volume has now passed the 1 B mark, so the other entrants need to get their machinery well oiled in order to catch up.

During this year, other players have attacked the same market: Google with Android App Market, Microsoft with Windows Marketplace, RIM with BlackBerry App World, Nokia with Ovi Store, Palm with App Catalog, and finally, Sony Ericsson, who joined the race yesterday. The developer common business terms seem to be 30% revenue share to the store owner (RIM is happy with just 20%), plus a fixed annual fee in most cases. With the store fee, developers get the marketing and sales support, as well as certain level of testing and validation performed by the store owner. Especially Nokia has the advantage of a very large installed base of devices, compared to Apple; however, this is certainly not a single unified platform, effectively requiring multiple versions of an app that should run across the whole product range. The same applies to all the other stores - together with volume and variation, comes the fragmentation.

During May, Jonathan Schwartz announced Sun's plans to open a Java Store, containing Java and JavaFX applications. Again, Sun has a very large installed base of devices as a potential market: alone within the mobile scope, there are roughly 2.6 B devices running Java applications. Unfortunately, this does not mean all those phones would run identical Java apps; due to evolving standards, hardware limitations, and time-to-market reasons, device and platform vendors have created a widely fragmented product base. As for most of the stores, this will be one of the main challenges for the Java Store as well: How to make sure a customer receives an application that will happily run in her phone, without the need to crawl through unnecessary technical details?

Image © Apple Inc.

No comments yet - write one!

As this is the first Darteco blog entry, I'll give a quick outlook, regarding the targeted audience and planned content. With the background that I have, there will be strategic, business and technology content alike - though I don't intent to dig too deep into the technology content, keeping the material palatable for a wide audience. With the exception of cited references, all comments and opinions are solely my own, reflecting my reactions on various topics within and around the mobile industry. Sometimes, I expect to go back in time and bring up an example from the past experience; However, the intent is always to help understand the current state and potential next steps. I also expect to include management and leadership aspects, as they often are such a key to make the difference between high and average performance in the mobile world. When writing comments, please keep in mind the normal blog etiquette. I'll reserve the right to delete spam-like content.

With this, I warmly welcome you to read and actively comment the Darteco blog! You'll also find my Email address on the Contact section, in case you opt for that channel to deliver your feedback.

No comments yet - write one!