Do You Still Believe In Silverlight?

16 commentsWritten on October 30th, 2010 by
Categories: Silverlight

Unless you're not really paying attention to what's going on in the Silverlight world, you probably already heard that Microsoft has confirmed a shift in strategy for Silverlight:

I asked Bob Muglia, the Microsoft President in charge of the company’s server and tools business, that very question and got what I consider to be the clearest answer yet about how Microsoft is evolving its Silverlight strategy. “Silverlight is our development platform for Windows Phone,” he said. Silverlight also has some “sweet spots” in media and line-of-business applications, he said.

It's the development platform for Windows Phone (more on that in a minute). It has some sweet spots in media and business apps. Things sure can change a lot in a short time. Silverlight was at first positioned as an ideal technology for media. After that, it was increasingly being marketed as a great platform to develop business applications, cross-platform ones even. Now, despite the fact that nothing has been taken away from the platform's capabilities, it has some "sweet spots" in both areas where it was supposed to be a great technological choice. Every single feature or capability that once formed Microsoft's message that it was great for business applications is still there. And they won't take that away either. The mere fact that their message about Silverlight has transformed so much now, is incredibly telling. If you're a Microsoft customer who's invested a lot in Silverlight for business applications, you have some pretty good reasons to be upset about this.

After all, given how the stated goals of Silverlight have evolved in the past 3 years, who's to say those "sweet spots" will remain interesting enough for Microsoft? I have no doubt that they're going to heavily invest in making Silverlight a great development platform for Windows Phone (whether they will succeed at it is an entirely different matter though), but what you need to develop great mobile applications differs a lot from the needs you have when building great business applications. Will the needs of developers using Silverlight to create business applications be met by Microsoft? Let me rephrase that: keeping the manner in which they're now promoting HTML5 in mind, will they also keep putting additional resources into improving some niche areas of a technology who's primary purpose is to facilitate mobile development? Maybe they will. Maybe they won't. I sure as hell wouldn't bet on it.

Now, unlike a lot of people, i don't think that Silverlight is dead. Far from it even. Silverlight has a future as long as Windows Phone has one. Its real future lies in the mobile world now, and no longer in the browser or on the desktop. It's going to keep working in your browsers, and its going to keep working on your desktops. But you really need to wonder how many of the needed improvements for those 2 environments you'll see being introduced to Silverlight, and how long it'll take. It's quite reasonable to assume that improvements in Silverlight which target the mobile development story will be receiving higher priorities than those which target the browser or desktop development model. After all, that is what the new Silverlight strategy implies. And given Microsoft's need to conquer a respectable position in the mobile world, and their ever increasing support of HTML5, it only makes sense for them to do so.

As for Silverlight being a suitable platform for mobile development, i can't help but wonder whether it really can live up to the high expectations that people will set on it. Anyone who's done any serious Silverlight development knows that it is a technology which suffers from a lot of memory usage problems. Partly because there are quite a few memory leaks in Silverlight's own controls, but also because the Silverlight programming model easily leads to a lot of dangling references which will lead to serious memory problems in a memory constrained device such as a mobile phone. Unless they get rid of the memory problems very quickly, i can easily see Windows Phones suffering from a lot of problems as people start using more and more 3rd party applications on their phones simultaneously. Well, it would have to support running multiple applications simultaneously before it'll be a real problem, but you know that's only a matter of time.

So, as for me personally, i don't really have a lot of belief left in Silverlight as a technology. I don't see it advancing much more in the browser or on the desktop, and i think it's got some real, difficult challenges ahead in the mobile world. What do you think?

  • http://www.inquisitivechap.com Stefan

    Well I’m just back from the beta MCP exam 70-506 TS: Silverlight 4, Development. I hardly use SL4 these days, but the exam was relatively easy as the technology itself is quite, well.. intuitive.

    In order to get HTML5 (+jQuery + some data backend + some session management + some REST way of accessing data + etc) application right you need to learn a lot, experience it and then code, test and deploy… Mind I haven’t used SL4 on anything bigger than a pet project, but it was sooo easy to setup everything and get things done.
    No issues with jQuery data binding, playing with CSS3 trying to get things where I wanted then, integrated authentication supported out of box, etc.

    I’m not a web guy, but I think that SL will stay with us, enterprise quys looking for an easy to deploy way of delivering apps that have usability requirements close to what was required on desktop (Ctrl+C, Ctrl+V mainly ;-)

    But sure, then next big thing is something that will allows such a nice dev experience for the web.
    It always seems to just behind the corner (GWT anyone?)

  • http://irascianwork.blogspot.com Ian Smith

    Thank you for restoring my belief that not every Silverlight blogger is a Microsoft shill determined to bury their head in the sand, ignoring anything and everything that doesn’t promote their narrow world view of things, whilst simultaneously attacking those who point out the obvious as “haters”, and criticising independent tech journalists who repeat word-for-word what Microsoft higher ups tell them as “unprofessional”.

    What a sorry state Microsoft “community” (now seemingly completely controlled by Microsoft, and focused solely on Microsoft suck-ups wanting MVP renewals instead of any kind of honest or real world analysis) is in. At least if the tweets and blog posts of the last few days are anything to go by.

    This is by far and away the best blog post I’ve seen on the whole heated “Silverlight vs HTML5″ debate.

    I could say a lot more about what I think, whether it’s the “lie” that Ray Ozzie was “moving” to the consumer division (since revealed to actually be a complete departure, albeit without an agreed date), the complete repetition of a three year old keynote at this year’s PDC (with only the word “Silverlight” changed to “HTML5″ this time around), the cluelessness of a CEO when it comes to mobile or tablets who refuses to learn anything from the mistakes he made four years ago with the iPhone and much more recently with the iPad, the… well you get the drift.

    The real story here is that Microsoft have shafted (again!) thsoe developers who swallowed their “vision” and have spent the last 3 years investing their own time and money getting up-to-speed in a complex, rushed-to-market, alarmingly buggy technology they’ve now decided to abandon (just as they were beginning to get on top of it – oh! the irony!) because of their inability to market it correctly or follow through on for the real world marketplace.

  • http://damieng.com Damien Guard

    It’s not *the* development platform for Windows Phone 7 – you can also use XNA and indeed have to if you want 3D.

    [)amien

  • Pingback: Tweets that mention Do You Still Believe In Silverlight? | The Inquisitive Coder – Davy Brion's Blog -- Topsy.com

  • Mike

    I’m seriously confused by this move and really question if Microsoft knows what they are doing. Marketing Silverlight as the development platform for WP7? You might as well put a fork in it. The market share is already dominated by iPhone and Driod. There is no compelling reason to make the switch to wp7. That aside, I thought the real power of Silverlight was developing internal Line Of Business applications that could run in most browsers without issues. Where I work we are stuck with many different versions of IE. We have some users that can’t upgrade, because some thrid party apps don’t work (that is another story in itself). I would imagine that many other companies are in the same position. So where does this put us in Microsofts future strategy? Can’t take advantage of HTML5? Oh we can, just use a different browser. That’s a brilliant marketing strategy, push all your users to a different platform.

    From this you have to wonder what Microsoft’s long term direction is. Take the announcement of Lightswitch, it generates Silverlight applications. So let me get this clear, Microsoft, you spent money creating an app that generates an app in a technology that you are completely changing direction on? Sounds like a good use of money. Maybe they are trying to position Ligthswitch as the development tool of choice for WP7. Further announcing this at PDC (a developer conference) and showing the new features of IE9 and how it supports HTML5, doesn’t make much sense either. At a developer’s conference, one would think, you should show off tools developers would use to aid in the development of HTML5.

    Davy, are you guys creating new apps in Silverlight? What about the apps that have already been written?

  • http://codedreams.blogspot.com Mike Strobel

    My feeling has been that Silverlight needs to go away. Microsoft’s former strategy of Silverlight being “the” UI platform has done serious damage to the .NET ecosystem, and most of the damage has been felt by WPF. Their strategy readily encouraged ignorance as to the capabilities (read: limitations) of the Silverlight platform, and WPF’s development and adoption rates have suffered immensely as a result. Silverlight was almost never the best choice for desktop applications, and it’s time to end this nonsense once and for all.

  • James L

    Hopefuly this will stop wpf from being held back. Maybe they’ll start using outside of the dev division! Lets not forget that html5 is still a few years away from really happening. I wonder what they have in store for us? Must be pretty well advanced for this sort of public message to be made. Its the right move of course

  • http://elegantcode.com/ Jan Van Ryswyck

    I started learning more about HTML(5), CSS(3) and JavaScript starting about a year ago. I sort of figured that the intrinsics of real web applications have always been centered around these big three, and will be for a long time to come. I sorta expected a fall or at least a decline of Silverlight, but not as spectacular as what we’re seeing here. Just see what happened with Flash the last five to seven years, and the outcome for Silverlight wasn’t that hard to foresee. Anyway,
    I’m glad that Microsoft has finally come to their senses and embrace the standards of the web (only time will tell, but fingers crossed).

  • http://davybrion.com Davy Brion

    @Mike

    “Davy, are you guys creating new apps in Silverlight? What about the apps that have already been written?”

    we are, though i don’t think it’s a good idea for many of the reasons i’ve described here:
    http://davybrion.com/blog/2010/09/keep-your-eyes-on-the-road/

    As for apps that have already been written… they’re going to keep working. The only problem is that Silverlight itself will improve very slowly from now on, and i’m guessing not all that much.

  • / pho

    In an enterprise environment for LOB apps I think silverlight makes a lot of sense; easy mindset switching for developers that get shuffled between WPF and web projects is always a plus. But other then that I never did see any future for it…

    Proprietary web technologies will not make it in the “open web world”, with consumers demanding content on more devices than just their windows/osx machine.

  • Peter

    Silverlight isn’t dead, HTML5 is far from being an adopted standard.
    See what the SL product group has to say http://team.silverlight.net/announcement/the-future-of-silverlight/

  • http://davybrion.com Davy Brion

    @Peter

    i didn’t say it’s dead, but that doesn’t mean it has a bright future ;)

    and what the product group has to say is irrelevant… Lotus Notes product groups probably still have a lot of great things to say about their product as well

  • JeroenH

    Muglia responds here (not directly to this blog post I presume, but still ;-)

    http://team.silverlight.net/announcement/pdc-and-silverlight/

  • http://davybrion.com Davy Brion

    @JeroenH

    all i see there is a mix of confirmation and damage control ;)

  • G-Mac

    Muglia is now set to leave Microsoft. Gotta be careful what you say, I guess.

    So Silverlight is still “all that” then, afterall. Who knew?

  • http://davybrion.com Davy Brion

    @G-Mac

    i still doubt that very much, but we’ll see how it unfolds in the next 2 years or so :)