Last week saw a significant debate around Microsoft PDC announcements and the push towards HTML5, rather than Silverlight. Here are a few links (in case you haven't read much about the topic) with personal comments. Before I get to those, however, let me state that I have a distinctive impression that Microsoft has a real lack of direction these days, and even at a recent conference with many Microsoft speakers I notice way more criticism towards the company than the average (which is quite low), but also totally different suggestions on the actual solution. That's not a good sign for me.

But let me get back to the commented links:

  • The first article predates the Silverlight discussion but it is very interesting for the Delphi camp. Tim Anderson wrote about " Lessons From Evernotes Flight From .Net ": the company moved one of the few WPF tools out there back to native Windows, using Visual C++. Of course, they should have rather used Delphi, to keep using the higher level and more sophisticated VCL model, rather than MFC, but the key point here is taht there are companies moving back from .NET client to native Windows. Certainly interesting, even if probably not a trend (also because native .NET clients are quite limited in numbers).
  • Later, Tim wrote about Silverlight Dream Is Over , reporting from PDC the impression that Silverlgiht was notably missing, that Bob Muglia had mentioned a "strategy shift" to HTML5. A significant comment is that "if Microsoft itself is downplaying Silverlight’s role, it will tend to push developers towards Adobe Flash". Notice that over the last week there were even rumors about deals (or even a merger) between Adobe and Microsoft, and it is easy to guess that Silverlight vs. Flash might be part of it. In general, though, while Flash is gaining more and more platforms (including phone platforms), Microsoft is focused on Windows, Mac, and Windows Phone. Too little.
  • A second post from Tim from PDC was "Microsoft big on Azure, quiet on Silverlight": Microsoft demonstrated HTML5 Canvas hardware acceleration and has apparently the most HTML5 compatible browser: this is a big shift!
  • A separate personal comment is that using the keynote at SDC in the Netherlands, which focused on Silverlight, i had once more the distinct impression that too many Silverlight success stories relate to video streaming. Which is interesting, but it is also what HTML5 is really likely to kill...
  • Julian Bucknall of DevExpress (long time Delphi friends, but active producer of Silverlight components) has a blog post about the same topic, and like 100s of other blog post, "A Story for Halloween: Is Silverlight a Zombie". Julian comments: "WPF and Silverlight were two of the future tracks for .NET (the third being ASP.NET MVC) at PDC. Failure of either or both , will mean that .NET’s future will be tied to ASP.NET MVC". I'd say too little for a platform that was supposed to revolutionize development.
  • An official Microsoft blog post on " The Future of Silverlight " tried to do soem demage control and give a different spin to the whole matter, basically pointing out that HTML5 and Silverlight can work in parallel. Discussing standards versus "innovation", they claim HTML5 is adding support for features previously provided by plugins, later covering areas in which HTML5 won't be good enough. But this is very weak, and is can be read as "mission accomplished", so we can now drop or reduce the role of Silverlight. The excuse that you need out of browser applications is very weak, given there are other options and they don't address most phones, where applications are more and more popular. I'm biased but don't think Silverlight is the best replacement for a native Delphi Windows application.
  • I found a very interesting comment in a reply to the blog post by a "former internal Silverlight developer at Microsoft", The Rise and Fall of Microsoft UX , saying that "I’ve been saying to a colleague that MS has an inconsistent strategy for years. Its like they have the classic “too big” syndrome where they have so many autonomous divisions with differing visions of the future, yet to the outside world they look schizophrenic." This is 100% my position. I don't know if Silverlight is alive or dead, if there will be further investments or not. But I don't care. If Microsoft changes stratagy every could of years (also within the ASP.NET realm), I won't suggest to many of my clients to do a two-years development project on any of these platforms, there are just too many chances that the technology will be obsolete before they ship. Truly native HTML can be hard, but it will certainly last longer and be a much better long-terms investment. The same can be said for most work you do in Delphi (but certainly not all of it).
  • Finally, one of the best post I've read is "Silverlight Shenigans" by Ian Smith. One of the key points is "Yes, HTML5 and JavaScript do suck somewhat. The tooling isn't there either. And the real mistake Microsoft have made in rushing to announce the 'change in strategy' is that they've done so too early. The tools aren't there for HTML5, Microsoft haven't got them in place, and are clearly some significant time way from having them in place."

Having written comments to the individual link, there is not much I have to add... other then reiterate that Microsoft seems to be lacking vision and long term support for its technologies. After dropping all possible Java support, VisualFox Pro (still a rather lively platform), the dynamic languages on top of .NET, moving their bloggers to WordPress, and not down playing the role of Silverlight, the overall trust in the development story at Microsoft is getting to a new low. But that's a personal opinion, of course.