Category Archives: software development

Qt will not be ported to Windows Phone 7 says Nokia

Director of the Qt Ecosystem Daniel Kihlberg has posted officially on the future of Qt, Nokia’s cross-platform application framework.

However you spin it, Nokia’s change of direction, relegating Symbian to low-end phones and focusing on Windows Phone as its Smartphone platform, is not good for Qt developers. Kihlberg offers a glimmer of hope for MeeGo though. Whereas CEO Stephen Elop was almost dismissive of MeeGo, saying that a device would be released as part of a learning process, Kihlberg positions it as a source of future disruption:

Nokia also announced it will ship its first MeeGo-related device in 2011, which will rely on the Qt ecosystem – and then will continue with MeeGo as an open source project for future disruption.  Nokia can’t afford to be behind the next disruption again and Qt can play an important role in making sure it isn’t.

But why not port Qt to Windows Phone, which needs a native development stack? Nokia’s Aron Kozak states in a comment:

Qt will not be ported to Windows Phone 7. One of the key benefits of joining an established ecosystem is that there is an established toolchain that everyone uses. All Windows Phone apps will run on all WP7 devices. Adding Qt to the mix would only cause fragmentation.

Unfortunate from a Qt perspective but wise from a developer ecosystem perspective.

In truth, this is near-fatal for the future of Qt at Nokia:

I have to say, Nokia made a bad decision jumping to WP7 knowing that Qt wouldn’t be on it. Now that Nokia did this, they basically went from Qt “Code once, run everywhere” to “Code once, run nowhere”.

says developer Keith Rusler.

The other problem is that developers feel misled:

When Elop came in he said that Qt will be the main framework. Symbian and MeeGo would be unified through Qt. We all stopped working on Symbian C++ and started learning Qt. We have now wasted 6 motnhs of our family’s lives on a dead end. If I knew this was going to happen, I would have started learning Java instead!

Irrespective of the business merits of Elop’s decision, the truth is that its relationship with developers has been deeply wounded. I am not sure how it could have been better handled – except that I think Nokia should have insisted on Qt support in Windows Phone – but I still observe that it has been handled badly. The evidence suggests that Elop under-estimates the importance of nurturing developers in the ecosystems he talks so much about.

MeeGo, Qt, and the new Nokia: developers express their doubts

What are the implications of the new partnership between Nokia and Microsoft for MeeGo, the device-oriented Linux project sponsored by Intel and Nokia? What about Qt, the application framework that unifies Symbian and MeeGo development?

Here is what Nokia says:

Under the new strategy, MeeGo becomes an open-source, mobile operating system project. MeeGo will place increased emphasis on longer-term market exploration of next-generation devices, platforms and user experiences. Nokia still plans to ship a MeeGo-related product later this year.

Nokia is retaining MeeGo but it has moved from centre-stage to become more niche and experimental.

The snag for developers is that there are no known plans to support Qt on Windows phone. According to the letter to developers, Qt developers can look forward to the targeting low-end Symbian devices and at least one solitary MeeGo phone:

Extending the scope of Qt further will be our first MeeGo-related open source device, which we plan to ship later this year. Though our plans for MeeGo have been adapted in light of our planned partnership with Microsoft, that device will be compatible with applications developed within the Qt framework and so give Qt developers a further device to target.

Reaction from developers so far is what you might expect:

By this announcement, I’m afraid you’ve lost many faithful people (developer and consumers) like myself, who’s been a Nokia user ever since I’ve started using cellphones..

and

Wow what can I say, nokia just flat out killed any enthusiasm I had to develop on nokia platforms, I never have and never will use a windows platform. You have just killed QT, even worse killed the most promising OS out there in Meego. Elop is the worst thing that has ever happened Nokia.

and

Weak on execution, you choose to flee. What a sad day in the history of a once proud and strong company.

Nokia could fix this by demanding Qt support for Windows Phone 7.

Trying out MonoTouch – C# for Apple’s iPhone and iPad

I’ve posted an article on trying out MonoTouch, which builds on the open source Mono project to provide a means of developing apps for Apple’s iOS using C# and the .NET Framework.

It is easy to assume that since the .NET Framework is Microsoft’s technology, using a non-Microsoft implementation is risky. Then again, Mono is open source; and the more usage it gets, the better it becomes. MonoTouch is an important development for the project, since it is a commercial project which might actually be making some money for Novell/Attachmate. While it would be nice to get it for free, it is important that Mono makes business sense as well. MonoTouch has given the Mono project a significant boost.

Using SQLite 3 with Unicode in Delphi

I’ve updated my SQLite3 Delphi wrapper for Unicode in Delphi 2009 and higher. Previous versions of the wrapper ducked the issue by using Ansi strings throughout.

image

I actually used Embarcadero Delphi XE for the development, but I would expect it to work in Delphi 2009 and higher, since it was in that version that Delphi first properly supported Unicode. Converting older Delphi projects is meant to be seamless, except in cases where you are using pointers or doing interop with native DLLs; of course this wrapper does both.

The SQLite 3 API expects either UTF8 or Unicode strings. To be more precise, some functions have Unicode versions indicated with a “16” suffix, and some do not, in which case they expect UTF8 if they accept string values. Although UTF8 strings support Unicode characters, most characters generally occupy a single byte just as in Ansi strings, so one of the things I discovered was that I could not simply rely on PChar, Delphi’s null-terminated string type, which from Delphi 2009 is a Unicode type with double-byte characters. Instead, for cases where the SQLite 3 API expects a UTF8 string, I have used PAnsiChar as before.

It is all somewhat confusing, and there are a few cases where Delphi does not do quite what you would expect. I recommend Marco Cantu’s paper Delphi and Unicode [pdf], one of the best resources I found. This article by Nick Hodges on Unicodifying your code is handy too.

Finally, in the example I keep an object in memory and it is easy to end up with code paths that do not free it. I love this feature of Delphi (since at least 2007) which informs you of your mistake when the application closes:

image

I have uploaded the code and you can find it linked here.

Update: My assumption that Delphi 2010 would work the same way as Delphi XE was incorrect. I have some code that reads a blob field containing a UTF8 string and returns it as a string. I read the stream into a byte array and then cast it to a UTF8String:

str := UTF8String(bytes);

where str is a UTF8string variable. This worked in Delphi XE, but in Delphi 2010 I got garbage. I have modified the code to use a TStringStream and now it works in both.

Microsoft’s new President of Server and Tools–Satya Nadella, from Bing division

Microsoft has appointed a new President of Server and Tools to replace Bob Muglia. He is Satya Nadella, 43, and has been leading the Online Services Division, the bit in charge of Bing, MSN and adCenter. Before that, Nadella led Microsoft Business Solutions, focused on the Dynamics CRM application.

image

It is a big role, and no doubt it is significant that Nadella has a cloud-oriented background within Microsoft. In his first memo in the new role, Nadella says:

I like to have my head in the clouds and feet on the ground (something I learned from Qi).

I track Microsoft financials in a small way, and have observed that while Muglia’s Server and Tools has turned in stellar figures for several years, the Online Services Division by contrast has delivered large losses.

Then again, competing with Google in search is not easy, and Bing is perhaps doing as well as can be expected.

CEO Steve Ballmer has also announced that Amitabh Srivastava, the well-regarded senior vice president in the Server and Tools Business is leaving Microsoft.

Puzzling moves, which suggest to me significant strategy disagreements between Ballmer and the old Server and Tools leadership.

MeeGo NoGo: things look bad for the Intel/Nokia Linux project

A sad post yesterday from MeeGo contributor Andrew Wafaa suggests that MeeGo on netbooks may no longer happen:

Basically by all accounts MeeGo is stopping all work on the Netbook UX. Yup, all our hard work is now almost for nothing 🙁

This is remarkable. The original Moblin project, sponsored by Intel, was all about bringing an excellent user experience to Linux on netbooks. The first netbooks ran Linux, but met resistance from a general public familiar with Windows; yet Linux is more suitable for netbooks than Windows in its present form.

Moblin is different. It’s a friendly way to get the most out of your netbook. It doesn’t work like most other computers because it’s optimized for enjoying media, interacting with your social networks and the internet.

wrote Moblin Community Manager Paul Cooper back in 2009, when netbooks were hot.

The problem: tech trends sometimes outpace corporate planning. Moblin was a good idea in 2008, but nothing was delivered; and by the time it looked like it might be ready, the market seemed to want tablets – or Apple iPads – rather than netbooks; and whatever problem Moblin was addressing was already solved by Google Android.

image

Two years later, in February 2010, Moblin merged with Nokia’s Maemo, creating a new project called MeeGo. The new focus would be tablets and smartphones:

The power and capability of handhelds has reached astounding levels – netbooks have been a runaway success – and connected TVs, tablets, in-vehicle infotainment, and media phones are fast growing new markets for devices with unheard of performance. Our goal is to develop the best software to go with these devices.

said Intel’s Imad Sousou.

So where are the MeeGo smartphones? Well, maybe we will see one at Mobile World Congress next week. But Nokia is in disarray. According to a leaked memo from new CEO Stephen Elop:

The first iPhone shipped in 2007, and we still don’t have a product that is close to their experience. Android came on the scene just over 2 years ago, and this week they took our leadership position in smartphone volumes. Unbelievable.

We have some brilliant sources of innovation inside Nokia, but we are not bringing it to market fast enough. We thought MeeGo would be a platform for winning high-end smartphones. However, at this rate, by the end of 2011, we might have only one MeeGo product in the market.

Perhaps Nokia will progress MeeGo smartphones with renewed vigour; but what looks more likely is that Nokia will embrace a rival platform, maybe Google Android or Microsoft’s Windows Phone 7.

That might well be alongside MeeGo, rather than replacing it, but Nokia needs to focus its energy and I would guess that MeeGo will lose out.

It may be the beginning of the end for a promising project that has progressed too slowly.

Update: Reuters is reporting that “two industry sources close to the company” say Nokia has ended development of its first MeeGo smartphone

Nokia plus Windows Phone 7 – would that be a smart move?

The rumour is that Nokia’s CEO, ex-Microsoft Stephen Elop, is planning a major strategy announcement on Friday February 11. The obvious move would be to embrace a new Smartphone platform, since neither Symbian nor MeeGo look likely to catch up with frontrunners Google Android or Apple iPhone. Could Elop be planning to partner with his former company and embrace Windows Phone 7?

It is a fascinating proposition. Here is the case in favour. For both Nokia and Microsoft, Android is the key competition in this market. The momentum behind Android is deterring both phone manufacturers and operators from investing seriously in Windows Phone 7. Microsoft’s phone is well-regarded, but has made little impact on the general public. Nokia could change that; it could make beautiful Windows 7 phones and get them to the mass market.

Microsoft has also done a good job with the developer tools for Windows Phone 7, with Visual Studio 2010, Silverlight, XNA, and the .NET Framework.

On the other hand, if Nokia were to adopt Windows Phone 7 for its high-end phone platform, would it not alienate its own development community, which is oriented towards Linux and C/C++? I think it would, unless Nokia insisted that as part of its deal with Microsoft, Windows Phone 7 would also support native code development with Qt, Nokia’s cross-platform application framework. This would be great news for Microsoft as well, though it might not recognise it. Windows Phone 7 needs to allow native code development, and Qt is ideal for the purpose. Qt already supports Windows CE, which underlies Windows Phone 7. If Nokia could present Windows Phone 7 as just another platform for Qt, the deal would be palatable for existing Nokia developers.

If Nokia were to announce this, it would transform the prospects for Microsoft’s Smartphone OS as well as helping Nokia to make a renewed impact.

Now for the case against. I am not sure that Qt on Windows Phone 7 would be acceptable to Microsoft, which might prefer to keep developers locked to Visual Studio and .NET; and Nokia has an easy alternative, which is to adopt Android instead. Qt support is still an issue, but there is already an independent project to bring Qt to Android. The combination of the Android and Nokia brands has obvious appeal, whereas taking on Windows Phone 7 would be risky.

The biggest shadow over Windows Phone 7 is cast by Microsoft itself. I do not doubt the commitment of the team which builds it within Microsoft, nor the quality of the developer tools. I do question though whether Microsoft as a whole sees a long-term future for Windows Phone 7 and its “Metro” user interface. The strong hint at CES was that Windows 8, rather than Windows Phone 7, is the basis of Microsoft’s tablet strategy; and if that proves to be the case, then Windows Phone 7 may gradually be displaced. Another puzzle is how Microsoft intends to use “Jupiter”, a rumoured new user interface library for Windows that may well be designed with mobile and touch control in mind. Maybe full Windows with “Jupiter” is the future of Microsoft’s mobile platform, rather than Windows Phone 7? I discuss this in more detail here.

There is enough uncertainty around Windows Phone 7, and enough buzz around Android, that Google’s mobile platform looks to me more attractive than Microsoft’s from Nokia’s perspective. I do not dismiss the Windows Phone idea though; it would be a bold and interesting move.

I expect this post to be very out of date soon, if not by Friday, then certainly by early next week at Mobile World Congress.

Update: A Nokia and Microsoft partnership is looking more likely since Google’s Vic Gundotra tweeted:

#feb11 "Two turkeys do not make an Eagle".

Updated SQLite wrapper for Embarcadero Delphi (and Free Pascal)

A while back I worked on a Delphi wrapper for SQLite 3, which I published on this site as an open source project. Others amended it to support Free Pascal and Lazarus, so you can use it on Linux and on the Mac. I’ve not touched it for a couple of years; but recently received an email requesting support for the SQLite 3 backup API. There are only a few functions involved so I added them to the wrapper, and also updated the Sqlite DLL to version 3.7.5.

Although I used Delphi XE to work on the wrapper, I did the build for the repository with Delphi 7, running in a virtual machine, because I know this version still has plenty of use. The problem though is that Delphi 2009 introduced full Unicode support, causing compatibility issues. My wrapper is compromised because it uses the old AnsiString, so that it works with all Delphi versions. I have in mind to fix this so you get full Unicode support when I have time to do so.

There is no support for Delphi’s data binding, and the wrapper appeals to developers happy to code their own SQL. Of course it works like lightning.

Red Gate to charge for .NET Reflector, runs into storm of protest

Tools company Red Gate is to discontinue the free version of .NET Reflector, a popular tool for debugging and decompiling .NET code.

image

The tool itself is amazing. It takes advantage of the fact that .NET code is not compiled to native code until runtime. The code that is distributed is in .NET “intermediate language”, which means it can easily be decompiled. Reflector makes this as easy as opening a file. This is invaluable for debugging when you do not have the original source code, though a further implication is that if you want to protect your source code you need to obfuscate it.

Reflector was created by Lutz Roeder, who shared it freely with the community. In 2008 Roeder sold Reflector to Red Gate, stating:

Red Gate will continue to provide the free community version and is looking for your feedback and ideas for future versions.

Red Gate developed Visual Studio integration and some further features, offering a free version as well as a paid-for premium edition. Now it has decided this is no longer viable. Watch this video as distinctly uncomfortable CEO Simon Galbraith answers the question “Didn’t we promise that Reflector was always going to be free?”:

Right now owning Reflector doesn’t make commercial sense. Further development of Reflector doesn’t make commercial sense. Reflector’s a tool that needs to stay up to date. We need to spend money on it. At the moment we can’t do so in a commercially justifiable way.

We’re really regretful that we ever made such a statement that we were going to try not to charge for it. In hindsight we wish we hadn’t done that. It was never a promise by the way. Two years ago we thought we could make a success of it without having to charge for it, it turns out that wasn’t the case.

Galbraith says that Red Gate had expected two benefits from Reflector: sales of other tools to Reflector users, and up-sell to the premium version. Neither has really happened, he says, adding:

We know that people are going to be cross with us.

What has particularly annoyed users in the feedback forum is that the existing free version is time-bombed, and will expire on May 30th. So users will be forced to upgrade.

i have been a happy (paying) customer of red-gate for some time now – sql tools, .net tools.
i have told many other developers of your products – happy to explain how good they are and how awesome red-gate is.
i won’t be doing that any more. IMO you have managed to destroy your company reputation such that purely on principle i won’t be recommending you to anyone anymore.

says one user.

Reflector has actually been time-bombed for years. I have a download from 2008, and if I run it I get this message:

image

If I click Yes, I am told that automatic update is impossible and directed to the Red Gate download page. If I click No, the dialog closes. In either case, the Reflector executable deletes itself. This is not so bad if you can download another free version; but following the change of policy that will not be the case.

Red Gate might not have made money from Reflector, but now it has the opposite problem: bad PR because of withdrawing an existing and popular free tool.

The price for the new Reflector will be just $35.00; not much for such as useful tool. There is nothing wrong with a software company charging for its work. This is a difficult transition though, and the question is: would Red Gate have been better off releasing Reflector back to the community and ceasing its own investment, rather than making a renewed effort to make it a viable commercial product?

How Microsoft’s Office Web Apps were written in C# and compiled to JavaScript, maybe

While researching another product I came across this 2009 tweet from Microsoft’s Nikhil Kothari:

Office 2010 web apps – perhaps one of the most ambitious script# projects!

Script# is loosely equivalent to the Google Web Toolkit, but whereas GWT compiles Java to JavaScript, Script# compiles C# to JavaScript. According to the site:

Script# is used extensively by developers within Microsoft building Ajax experiences in Windows Live, Office to name just a couple, as well as by a external developers and companies including Facebook.

I had come across the project before, but was waiting to see if would evolve beyond what looks like a personal project for Kothari. It is hosted on http://projects.nikhilk.net rather than on an official Microsoft domain, and the latest release is 0.6.2. In other words, it does not have the look of a project that you would recommend for production work, interesting though it is. Nor is there much public activity around Script# that I can see, though there is a CodePlex site dedicated to improving its JQuery support.

Seeing Kothari’s tweet though raises several questions.

  • Did Microsoft really use it for Office Web Apps, a high profile project which is a key part of Microsoft’s cloud computing strategy?
  • Is there another, more up-to-date version of Script# that is used internally and which may one day burst into the public arena?
  • How might it impact the Silverlight vs HTML5 debate, if Microsoft comes up with a C# to JavaScript compiler in Visual Studio that lets developers code in .NET but deploy to cross-platform JavaScript?

I am sure there are readers of this blog who know more than I do, so by all means let me know.