Update: Why Microsoft Silverlight Will Fail – 1 Year Later


Last April, I wrote a post listing some reasons why I felt Silverlight would not succeed. That post garnered a lot of visits, comments and a few trackbacks. It’s still a highly trafficked post of mine. Furthermore, that post is the number one Google result for “silverlight IDE” and ranks pretty highly for a number of Silverlight related phrases like “silverlight penetration”, “silverlight market penetration”, and a few others along those lines. I’d like to revisit some of those points to see how things have changed in the past year. I was inspired to revisiti this after reading Robert Scoble’s friendfeed topic on this subject. Read on to see the progress.

  • No IDE for me – There is still only one way to compile SIlverlight content. Windows. On the Adobe side of the fence, there is indeed a Linux version of Flex, and with the Open Screen Project, I’d be willing to bet a lot more SWF compiling/creating tools are coming soon.
  • No plug in for Linux - Adobe’s Linux support continues to grow, albeit a little slowly for the 64 bit crowd. As for Silverlight, it’s pretty much a no show, but if you are a particularly industrious Linux user, you can compile yourself a Silverlight version that doesn’t play video or MP3s. Sounds fun, right?
  • Market Penetration = Demand - With Flash Player 9 on 85-90%+ of the clients out there, deploying Flash content and knowing that your audience will see it is a sure thing. They use Flash player. Microsoft still has yet to publish any sort of numbers on what sort of percentage of web users can use SIlverlight content. Does anyone know how many users are out there with Silverlight 1 today? Silverlight 2? I have yet to have a client request Silverlight content on their site or in their project, I definitely can speak for the demand side of things. There isn’t much. There are a lot articles and blog posts on this topic.
  • The Growing Mobile Content Market - I have to admit, I’m not familiar with the progress made in the mobile arena. I see that Silverlight mobile has been announced, but it doesn’t look to be available yet. Anyone know more about the release of this?
  • Maturity – Adobe issued a Flash Player update, “Moviestar”, that added HD video playback, and just recently, Flash Player 10 Beta, with a number of big additions. Silverlight has released a beta of version 2.0 of their plugin which adds a lot of great features including a small but acceptable list of controls for developing RIAs (Silverlight 1 largely requires you to roll your own, similar to Flash player pre version 6). I have to admit, I’m pretty sure that MS’s Silverlight push has to have motivated Adobe to get these crucial player updates out so quickly. I also know that the lack of good controls has soured a lot of potential Silverlight developers.
  • The Developer Community - Not much of a change here. The community for Adobe has only grown more open, with OSS advocates coming to the platform in droves, the Open Screen Project changing things dramatically industry wide. The Silverlight Community has Moonlight, but not much else to hang it’s hat on besides a few apparently sponsored showcase pieces and a few blogs here and there with some XAML and other tests, mostly focused on Silverlight 2.0.
  • CS3 Microsoft still has no creative suite competitor. No big surprise here.

I welcome being proven wrong. I really want to see more growth in this area. Honestly, I do. More demand in the RIA space ultimately means more demand for me and my company’s services. I am attending RIAPallooza at the end of the month, so I will be very interested in learning more about the SIlverlight ecosystem. The agenda looks great, so if you are in the Chicago area, I would recommend checking it out.

In my opinion, having competition in this space keeps Adobe from being slow and unreactive. I have little doubt that without Silverlight, H.264 support would not be in Flash Player 9. Also, I’m willing to bet that it certainly weighed on the decision to open the SWF and FLV formats, etc. I don’t begrudge or wish ill on the Silverlight project, it ultimately may serve to save MS from the collapse of the desktop which few deny is coming. RIA and thick client/semi connected app development is the next wave, with SaaS and freemium services leading the way. I’m glad that there are currently multiple paths and avenues to explore. What I really want now, though, is a nice DeepZoom like Flex component. Something better than Zoomify, please, I want to emulate that animation and the smooth loading/slicing done in that very cool tech demo.

Posted on


40 comments

  1. Peter Witham May 16

    We have to start considering how long until QuickTime progresses enough to start being a competitor to Flash as well.

    SilverLight still has a long way to go, but I’ve gone thru training on v1 and have to say that for anyone who is new to Flash/SilverLight etc. it is starting to look respectable.

    And we know that Microsoft will force the plugin with every Windows install, rightly or wrongly that’s an instant user install base.

    If I have to choose, I’d still develop Flash/Flex over SilverLight :)

    Regards,
    Peter Witham

  2. John May 16

    Hands down, my issues with Silverlight are the lack of Mac tools and the lack of any decent design tools. I work in a MS shop as a designer and the MS Expression suite is a load of crap. I use Fireworks extensively for all my design work and then use the free Infragistics FW to XAML extension. Heck – even our own developers won’t use Expression Blend except to maybe to draw a gradient.

  3. DannyT May 16

    I think you’re grossly underestimating the Developer Community point here. The Silverlight developer community will be a subset of the .net developer community which is absolutely massive. Even if only a small percentage of this community jumps on the Silverlight bandwagon it will easily compare with the size of the Flash/Flex community. This community also does have a very strong open and standards focused subset despite the common perception.

    I choose to focus our efforts on Flash/Flex for UI technology but do not rule out Silverlight as it will become a viable platform and there will be many agencies using it and many customers requiring it. Silverlight will not kill off Flash/Flex by any means but I certainly can’t see it disappearing any time soon either.

  4. winoo May 16

    My opinions as a .Net developer:

    1) A quick check between AS3 and C#: AS3 has no multithreading. C# does. AS3 has no overloaded functions. C# does. AS3 has no generics. C# does. AS3 has no LINQ. C# does. I’m sure AS3 still does animation pretty well (SilverLight is not far behind there either if at all) but RIA is going to be a serious platform. W/ so many things missing, AS3 is not that impressive a choice as a robust platform language.

    2) SilverLight fits in the existing proven .Net solution stack, which makes it the ideal choice for the massive .Net developers out there. The same will be true for JavaFX among J2EE developers. Picture this: If I already have my Sql Server layer ready and a WCF layer in the middle, why would I use sth outside of .Net world at the presentation layer. Someone has to use violence to get me use Flex over SilverLight there. Moreover, I doubt the server side support Adobe offers is mature enough to match what .Net or J2ee world has for years.

    3) Flex has more developers, for now. Once SilverLight 2 becomes gold, there’ll be a wave of .Net developers invading this RIA world. After JavaFX is ready, a second wave of Java developers will rush in. Both waves out-# the current Flex developers by a long stretch. By then, Flex’ popularity among developers will be gone. Adobe will be squeezed by both M$ and Sun.

  5. Rob May 16

    Hey winoo, I develop with VB.NET, and I mostly agree with your first paragraph, but it will take 3-5 years for Silverlight to impact Flash. .NET is good for a lot of stuff (Visual Studio’s text editing is infinitely superior to the Flash IDE), but it’s no replacement of Adobe Flash, but I also agree that if Microsoft can make Adobe do more stuff to keep Flash way ahead of WPF, everyone wins.

  6. Peter Witham May 16

    “therell be a wave of .Net developers invading this RIA world”

    Sounds like a Balmer kinda quote to me :) , no offense or flaming intended, it just sounded like one.

    The problem here is self evident, there will always be both sides claiming the winning solution over the other. At the end of the day I always use what ever the client wants to get the bills paid, regardless of which is best.

    Regards,
    Peter Witham

  7. JmD May 18

    To winoo (as well as the rest of the bunch):

    To be honest, as an end user I dont care what road the developer took to get the app up and running. I care about the final destination – the app itself. And in this department Flex / Flash is way beyond what Microsoft has to offer in terms of RIA. The simple truth is that the pure number of available apps from the Adobe camp is so undeniable overwhelming compaired to Silverlight based ones. And thats no surprise given the number of years each technology has existed. But apart from that – again from a users point of view – the things Ive seen done with Silverlight dont impress me much, with a couple of rare of exceptions. Flash / Flex apps simply are better and look better. And they look and work the same regardless of what OS I happen to use.

    ///JmD

  8. Jody Brewster May 21

    Just to let you know, the first part no IDE for me isn’t necessarily true. I can build Silverlight apps in Python on my mac in TextMate or whatever.

    Anyway, I agree with one of your points which is market penetration. But Flash has a problem in this space… bad reputation. If Microsoft can build Silverlight without the reputation Flash has grown, they may just make a dent in the market penetration.

  9. Tim May 31

    I mostly agree with winoo, even though I am not so confident that M$ or Sun will take over the RIA world ;-)
    The main advantage of SL is its native integration in the .NET stack. Every (well written) .NET application could switch its GUI layer to SL with little to no changes in the other parts of the application. No need to integrate another layer for data transaction, “issue” which hasn’t been completely solved with Flash/Flex (AMFphp for instance, doesn’t offer full compatibility btw server and client data types ).
    But like JmD said, who cares what language was the application build upon ? we just want it to be useful !

  10. javaguy May 31

    I think you’re overlooking something.

    Do you know what the most popular IDE is? Visual Studio…

    It’s highly likely .net shops will remain .net shops. I think Flex is the more polished product now, but I wouldn’t discount the 800 pound gorilla. They’ve got too much money and are investing heavily…it was 2003 when Visual Studio .net arrived, and look at where .net is now.

    Correct me if I’m wrong, but it’s the Flash Player itself that is not multi threaded. Big Oops.

  11. Peter Bromberg May 31

    No offense but it kinda sounds like your’re trying to “convince yourself” against Silverlight with many of the points you make. There will certainly be an IDE for the Linux crowd as the MONO / Moonlight implementation is moving ahead quickly (with Microsoft’s blessing, I might add). Same for the “plug-in”.

    Likewise, market penetration, mobile (Nokia is doing it), the community and everything else will come. You have a product that isn’t out of beta yet, so trashing it this early on in the game is kind of moot.

  12. Darren Jun 1

    The pro-Silverlight comments here are a great example of why Silverlight might succeed. They don’t really understand the rival technology but have already decided that they’ll be using Silverlight because it’s from MS.

    An example from winoo “I doubt the server side support Adobe offers is mature enough to match what .Net or J2ee world has for years”. The truth is that Flex is server-side agnostic but if you want to use Adobe LiveCycle, you’ll be using J2EE as your backend. But you can just as easily use the free WebORB product and employ .Net on the server-side. This is with real-time messaging, two-way communication of typed objects, etc. If you use LiveCycle, it will even automatically sync your data on both the client and server sides. Or you can use straight web services, languages such as PHP, Ruby, and Python on the server-side, the list goes on.

    This technology is available now. You can start building a Flex/.Net solution today with all the features that Silverlight might deliver in a few years time. Open your eyes, MS fanboys.

  13. Mike Jun 1

    Won’t Microsoft just eventually incorporate the Silverlight plug-in into Internet Explorer? I would think that would be a pretty quick solution to the market penetration question.

  14. A Director Jun 4

    @Darren

    Your obviously not a manager, nor are involved in budget decisions. Because if you were, then actual business decisions like the following might have occurred to you:

    1. cost to convert a .net shop to Flex including software, employee trainiing and lost productivity. If you have 50 developers in total, and estimate $20k per developer, you’ve already hit $1million
    2. Unless I did a huge bull*hit sales presentation loaded w/ pretty powepoint and buzzwords my stakeholders couldn’t understand(read Accenture), I could never justify the cost to the stakeholders. That $1 million would never pay off.
    3. If it were my money / company, I certainly wouldn’t pay.
    4. It costs a lot less to retrain a already .net employee to flex then to get an experienced .net developer to get up to speed on silverlight and WPF

    Btw – I’m not a .net guy and don’t run a .net shop. But I do run a large IT team for a fortune 500 company

  15. Darren Jun 5

    @A Director

    Well, I’d obviously argue that if your team is that big, you should not be letting the server-side .Net guys do user interface work. And you shouldn’t be letting any of them do design. That’s if you want a usable, stable, quality product. Work at a detailed spec beforehand and get both your teams to work to that, each using the technology that is best suited to the job. In the example you gave, that would be .Net on the back-end, Flex on the front-end, and WebORB in the middle. The front-end guys are going to have work closely with the designers, who’ll all be using Adobe products. Unless you want to retrain all your designers to use Blend? Flex is a front-end technology, let the .Net back-end guys continue to use .Net.

  16. netdragon Jul 1

    As an objective member of the advertising industry, I can tell you that unfortunately Silverlight won’t fail. There are already major publishers that rely on Windows Media Player that are switching to it with their goal being by fall or end of year, and from what I’ve heard the next version is relatively stable and bug-free. One major publisher has only a few Silverlight bugs that are keeping them from completely dumping Windows Media Player.

  17. netdragon Jul 1

    Btw, I forgot to mention to you that besides the fact that MS video has only slightly better video quality than Flash, it’s very very expensive for publishers with tons and tons of existing video clips all the way back to the start of the web to switch to Flash. It’s cheaper for them to just switch to Silverlight and then they have support for some type of Rich Media. I don’t see Silverlight being used for Flash banner ads, but for video players, I think it will have at least half the market share.

  18. Daniel Jul 25

    @Darren

    Truely spoken! Backend-Developers should never do UI-Design … that’s why SL may fail: All of their apps look crappy, cause all the creative minds are still in the adobe world. The customer only sees the product and for him it’s 99% design and not code. So the “side” with the better designers will win :)

  19. Biff Jul 25

    It’s interesting that very little of the discussion focuses on users. From their perspective, this is just another “rich” application plugin that will slow down their browsers, distract them with, ahem, flashy animations and unwanted sounds, etc. With the large footprint of .Net developers, I do expect that Silverlight will become very common by version 3 or 4, and I’m sure that I’ll eventually be forced to install it to access some important content, but I can’t believe there is actual user demand for it. Hey, if I were running MS, I’d do the same thing, but it’s hard for me to believe that the browsing experience will be any better, particularly since Flash and QuickTime aren’t going away either. Just another thing to slow the browser. Just another set of security patches to download.

  20. Tim Jul 28

    amen Biff, amen.

  21. mb Jul 28

    You guys are killing me.

    I’ve never seen a line of business app built in Flash. Maybe some cool online sites like Webkinz or Disney, etc., but there is more to RIA than video and gaming. Silverlight 2 opens the door to LOB RIA’s, and if Flash had done that well, we’d see a bigger presence of them out there.

    There is room for both technologies. You keep building your entertainment site in Flash, have fun with that while the SL guys race past, it will be far more accessible than Adobe, count on it!

  22. LoL Aug 22

    mb, you’re a funny guy. Maybe you’ve never heard of Flex, but that’s Adobe’s answer to more traditional and systematic web apps… and in its relatively short life, is already millions(!) of developers ahead of silverlight.

    Bottom line is, silverlight has been out over a year now and they’re too embarrassed to even release penetration numbers. Short of them stuffing a silverlight installer in the next Windows itself, they have extremely limited means of building that penetration to even a paltry 10%. No users will ever clamor for it. Flash/Flex looks and functions better and in more standardized/recognizable ways, AND it doesn’t require them to download some virtually unknown plugin anymore.

    And if/when they become capable of force-bundling it out to a large enough number of people to offer incentive to developers, it’s reasonable to assume everyone will have already moved on to the groundbreaking enhancements of the next flash player, and forgotten Silverlight ever existed.

  23. Arvidson Aug 26

    mb, you could not be more wrong. I work with NikeiD and NikeStore and they are all Flash (not even Flex as they were build before Flex becoming really freat to work with).

    So far I have not seen a really cool Sl site. Sure they work but they just don’t look and feel as good as what you can do in Flash/Flex. The reason for that, I believe, is Flash CS3. Having developers doing UIs will never turn out good. Until MS will have a great tool for doing exectly this, SL will not be a success…

  24. Mark Kovalcson Dec 15

    Let me start off by saying that I am an ASP.NET developer who is very comfortable with JavaScript, and who uses Photoshop and other Graphical tools. I also agree with comments that suggest that the typical developer is not typically as artistic as many people who deal with graphical content. However that has no impact on Silverlight being viable.

    Silverlight will allow these same developers to create much more advanced web applications leveraging tools that they are already familiar with. These developers outnumber Flash developers by orders of magnitude. I know many ASP.NET developers who hate JavaScript and are very excited about Silverlight.

    Silverlight will succeed well beyond Flash for a number of reasons.

    1. It is much larger in scope than Flash. We are talking about a Compact .NET framework running in a browser.

    2. Distribution is a simple key click download just like Flash, which is very little effort.

    3. Silverlight is gaining momentum fast. In October there were estimates on market penetration in the US approaching 30%

    4. The Adobe Flash development tools are expensive to those who haven’t already invested in them and there are a huge number of developers with Visual Studio who suddenly have the ability to create silverlight applications without learning a completely new toolset and who would never have the opportunity to use Flash tools.

    5. There is a community growing quickly and already starting to put source out there for more elaborate solutions using Silverlight.

    I’ll admit up front that I have been a MS developer for a number of years. That doesn’t mean that I dislike the Adobe and formerly MacroMedia products. Dreamweaver was my html editor of choice a few years ago before I started writing all of my html in Visual Studio.

    However it doesn’t take a rocket scientist to understand that when you suddenly give an Army of developers a powerful tool that has a low cost of admission in terms of learning curve and doesn’t require them to purchase any software, that it is going to gain ground fast.

  25. Harry Jan 5

    Silverlight will succeed for these reasons:

    1. Some developers will believe that it costs thousands of dollars to be a Flash Platform / Flex developer.

    2. Some developers will believe that Adobe is evil. Microsoft’s Scott Barnes is making good progress spreading this message.

    3. Some developers will believe that Flash is small and dinky. They believe that the bigger your executable is, the better the program is.

    4. Some developers don’t want to learn anything new, and will be assured that Silverlight is easier for them. (Yes, those afraid to explore new things will lead the way.)

    5. Silverlight will be embedded with every copy of Windows sold, making its penetration 100%.

  26. Richard Jahmarkt Mar 2

    This site shows market penetration for SilverLight, Flash and a number of other plugins:
    http://www.statowl.com/flash.php
    http://www.statowl.com/silverlight.php

  27. Harry May 13

    Hey, I was wondering if you could post another update.. now it’s been 2 years since your first prediction and it appears that Silverlight is at 26% penetration.

    It’s cool if you don’t have time to do another post though. I know that Adobe people are more interested in spending their time developing great new things like Gumbo and Catalyst, instead of spending all their time smashing their competition.

  28. JustAWebDeveloper Jul 31

    I am seeing More and More Silverlight stuff out there. Netflicks watch instantly just went Silverlight for instance. I am a longtime flash supporter/developer, but I am now beginning to recommend Silverlight to my clients that have the need to interact with a lot of data in their applications. It is simply better then flash in this circumstance, and now that it is supported in Linux and Mac clients, there is usually little resistance on the part of my clients. Most non data-centric applications still get a flash recommendation from me, but who knows what the future holds as I learn more Silverlight technology.

  29. free microsoft point Sep 1

    it will easily compare with the size of the Flash/Flex community. This community also does have a very strong open and standards focused subset despite the common perception

  30. postscripter Sep 9

    Am a .net developer , unfortunately till now very little number of websites are using Silver Light, and I stumbled upon many silverlight based websites that has plenty of bugs…I feel that silverlight is still not mature. What surprises me most is Micrsoft still uses Flash in some of their pages…for example the Microsoft online clincs are Flash not silverLigt!

  31. Jared Jan 18

    I’m not a fan of silverlight personally. It says that I need to get them to go on xbox.com and view everything, yet when I try to install it, it says I have the most updated already. Not sure which is the source of the problem.

  32. Free Xbox Live Codes Feb 19

    Silverlight has been out over a year now and they’re too embarrassed to even release penetration numbers. Short of them stuffing a silverlight installer in the next Windows itself, they have extremely limited means of building that penetration to even a paltry 10%. No users will ever clamor for it.

  33. Dominic Mar 1

    I see it as there is more .Net developers than Flash developers (attributing to the fact Windows is the most popular OS (and keep in mind that creating flash animations isn’t exactly a ‘developer’)). It’s already been a more than a year since you wrote this, and Silverlight is still not dead. Actionscript 3, compared to C#, is less than anything I have seen from a developer’s point of view. And the Flash IDE sucks for programming. Flex sucks too compared to Visual Studio, face it.

    But yeah, Adobe got the head start, and I doubt Silverlight will ever been as popular, but it`s getting there.

  34. Virtual Worlds For Kids May 30

    If Flash is going to have a had time surviving without any Apple mobile device market share, how can Silverlight make it?

  35. free itunes code Sep 30

    I totally agree Microsoft Silver light will definitely fail. It was going to the moment they released it.

  36. Free Amazon Gift Card May 21

    Looks like it’s failed already!

  1. Seb Lee-Delisle » Blog Archive » Why Silverlight is going to fail
  2. shane.korin » Blog Archive » Why Silverlight Will Fail
  3. 2 Years Later… An Update on Why Microsoft Silverlight Will Fail | Visualrinse | Design and Development by Chad Udell
  4. 2 Years LaterĶ An Update on Why Microsoft Silverlight Will Fail | Iona.LABS

Leave a reply