Everything and the Mobile Software Universe…

Big news in our world! Check the press release.

Price (the offer values the company to 100 Millions Euros) is not so high compared to Trolltech technical and community assets (but high …looking at the actual company revenues of 7 Millions Euros). This is not a dot com acquisition. Period.

The next game would be to understand why.

Trolltech is providing a native development environment called Qt, which is a set of “OS services” (memory management, Thread, etc…) and is famous widget library. This environment has been ported on Desktop Linux, Windows, MacOS and embedded Linux: “Qt/embedded”, now called Qtopia Core , on top of which a nearly complete phone application stack has been built, Qtopia.

The framework allows C++ development but recently a java version surfaced: Qt Jambi

Trolltech provides also (and sell) some development tools: a RAD, QtDesigner, qMake a command line tool chain, a plugin for Visual Studio and some internationalization utilities.

While huge adoption in the mobile phone market remains to be seen, Qt is at the earth of one of the biggest OpenSource piece of sotfware: the KDE Linux Desktop (…father project of the now famous webkit browser engine).

So crossing with Nokia current strategy and ths interesting quote from the Nokia PR:

“Trolltech’s deep understanding of open source software and its strong technology assets will enable both Nokia and others to innovate on our device platforms while reducing time-to-market. This acquisition will also further increase the competitiveness of S60 and Series 40.”

Kai Öistämö, Executive Vice President, Devices, Nokia

Here are the different bets:

• It is widely known that the proprietary S40 is difficult to maintain and extend/modernize, porting Qt as a companion framework may allow Nokia to open it’s most widely used platform (S60 is negligeable compared to S40 market share) to third party developpers … and open source developpers.
• Nokia wants to have cross platform technologies to merge S60/S40 and desktop environment, so take advantage of the HUGE Qt developper pool.
• Nokia desperately needs a credible platform and a set of APIs to counter Android in the web services area…and the Java Qt makes sense here.
• Does Nokia has some ambitions for KDE to use it as its base OS for its forthcoming “Personal Computer”, touted as the next big thing and next strategy of Nokia?

Be prepared for a S40, a S60 Qt port …. and perhaps an opening of the S40 platform, at least for selected third parties.

Anyway I quite don’t get this …

• In Hildon regards, the Maemo Tablet OS running on the nokia Internet Tablet (n770, N800 and N810). This one is based on GTK, the Qt archrival on the Linux Desktop, uses a Mozilla based browser, so is in the opposite technical direction: will it be cancelled as it is to run a Qt based Tablet OS?
• For KDE Desktop: Dealing with a little company like Trolltech is something, having Nokia as the main backer of its framework is something else. How the OpenSource community will react?

What do you think Nokia has in mind?

Thomas

Ok I admit, I have an iPhone. I love it,  blablabla you know the song already. It has its flaws, but as an old time mobile software engineer I’m really stroked by one BIG fact: The applications I use the most on it are fully web based!: My IM messenger (JiveTalk), my english/french dictionnary (Ultralingua), my mail and rss reader (special version of gmail and google reader) … even my all time favorite mobile game Bejeweled is web based!

What a shock I wasn’t prepared for that: when Steeve Jobs told us that the only way to add application will be (at first) through the web browser I was the first to laugh, only raw C++ is meaningful for applications, a web browser is a mere toy compared to a real application framework.

How wrong I was. And here is why. (and no it won’t be only about the iPhone)

• Unlimited and affordable data plan, and efficient bandwidth and coverage: I’m in Europe (France) and here network coverage and edge (2.5G) are very efficient.
• Webkit and Mozilla : Webkit engine tends to begin the defacto mobile web browser (check what pleyo is doing) embedded in S60, MacOS, Android…the only other credible contender is the Nokia Mozilla version (my Nokia N800 is simply unbeatable for web browsing).
• Raise of ad-hoc web services framework: the famous and numerous web widget frameworks (webwag being one to be noticed), and Yahoo GO for example.
• And the biggest one which is vastly under looked: modern websites, sorry webservices, are fully Model/View/Controller (ruby on rails, but above all struts2, etc.) what does it means in human readable language? : it is VERY easy to adapt the content/services of  a web site to different browsers / way of presenting data. Look at the plethora of “iPhone” optimized sites (ebay, dailymotion, facebook, etc)  that have popped up everywhere in few months.

Those approaches have something in common

• Need of a reliable wireless data link
• Well architectured network backend to provide optimized business data and adapted rendering data (the last one is not mandatory, check RSS for example were the business data has no notion of representation in it).
• An “On Client” web service framework: a browser with standard and added proprietary APIs like the iPhone Safari, a limited and fully proprietary engine like YahooGO, or a full OS with the complete stack like Android and … the iPhone OS (OK, don’t forget the “old” high level OSes like S60 and WinMobile).

Everything seems to be in place, and from what we saw above a good web service client platform would have to:

1. Be fun to use and compelling, tailored for each user
2. Be VERY efficient for the phone common tasks (phone call, address book)
3. Offer a nice and easy way to deploy data representation and flow control from existing web services backends…with good performance and relatively wide access to the underlying platform and datas

For me the first two doesn’t have to be understated (just try a WinMobile phone for a few months to understand what I mean ), as the device remains a phone, a communication machine and voice is still the undefeated champion for communication. This is where the iPhone is groundbreaking at a first sight…and also where I’m not sure of what Google Android will deliver (call me skeptical if you want…).

The third point may bring a lot of optimism … as it implies that we doesn’t need a single platform anymore, but a bunch of deployment possibilities, tailored for each devices/client or even each services . Android and  the iPhone may be seen as such a platform with at least two of those deployment possibilities: the browser and application native development, here Android is much more friendly to Java/Web programmer that the iPhone. But we could perfectly imagine devices with more deployment options or other completely different but close enough to web development standards to allow fast adaptation of web backends….why not an iPhone with an Android sandbox?

At the end the famous “cloud” (the network) is really shaping the “on device” clients, allowing more and more diversity and at there won’t be a “one fit all” solution…

Thanks Steve Jobs for being the first to have put in place all the elements of the chain, dealing with carriers, content provider, services providers…and coming with a great consumer electronic design.

Google wants to go further? not sure for now, but the US 700 MHz auction have to be followed very carefully cause if this spectrum becomes “free” of the carriers, we don’t know how fast it could go!