Microsoft’s president of Server and Tools Bob Muglia has posted a response to the widespread perception that the company is backing off its commitment to Silverlight, a cross-browser, cross-platform runtime for rich internet applications. He is the right person to do so, since it was his remark that ”Our strategy with Silverlight has shifted” which seemed to confirm a strategy change that had already been implied by the strong focus in the keynote on HTML 5 as an application platform.
Muglia says Silverlight is in fact “very important and strategic to Microsoft”. He confirms that a new release is in development, notes that Silverlight is the development platform for Windows Phone 7, and affirms Silverlight both as a media client and as “the richest way to build web-delivered client apps.”
So what is the strategy change? It is this:
When we started Silverlight, the number of unique/different Internet-connected devices in the world was relatively small, and our goal was to provide the most consistent, richest experience across those devices. But the world has changed. As a result, getting a single runtime implementation installed on every potential device is practically impossible. We think HTML will provide the broadest, cross-platform reach across all these devices. At Microsoft, we’re committed to building the world’s best implementation of HTML 5 for devices running Windows, and at the PDC, we showed the great progress we’re making on this with IE 9.
The key problem here is Apple’s iOS, which Muglia mentioned specifically in his earlier interview:
HTML is the only true cross platform solution for everything, including (Apple’s) iOS platform.
Muglia’s words are somewhat reassuring to Silverlight developers; but not, I think, all that much. Silverlight will continue on Windows, Mac and on Windows Phone; but there are many more devices which developers want to target, and it sounds as if Microsoft does not intend to broaden Silverlight’s reach.
Faced with the same issues, Adobe has brought Flash to device platforms including Android, MeeGo, Blackberry and Google TV; and come up with a packager that compiles Flash applications to native iOS code. There is still no Flash or AIR (out of browser Flash) on Apple iOS; but Adobe has done all possible to make Flash a broad cross-platform runtime.
Microsoft by contrast has not really entered the fight. It has been left to Novell’s Mono team to show what can be done with cross-platform .NET, including MonoTouch for iOS and MonoDroid for Google’s Android platform.
Microsoft could have done more to bring Silverlight to further platforms, but has chosen instead to focus on HTML 5 – just as Muglia said in his earlier interview.
Whether Microsoft is right or wrong in this is a matter for debate. From what I have seen, the comments on Microsoft’s de-emphasis of Silverlight at PDC have been worrying for .NET developers, but mostly cheered elsewhere.
The problem is that HTML 5 is not ready, nor is it capable of everything that can be done in Silverlight or Flash. There is a gap to be filled; and it looks as if Microsoft is leaving that task to Adobe.
It does seem to me inevitable that if Microsoft really gets behind HTML 5, by supporting it with tools and libraries to make it a strong and productive client for Microsoft’s server applications, then Silverlight will slip further behind.
I wonder what effect this shifting of positioning by MS for Silverlight will have on MS’s ability to attract devs to Windows-Phone-7. The most important constituency of potential WP7 devs may be the existing SL-for-web devs. But to the extent those SL-for-web devs lose confidence in MS’s word and capability, they may also not take the faith-based plunge into WP7 development.
If MS can change its mind about the positioning of SL today, could they eventually change their mind about WP7 in the same way they did about the “Kin”? If WP7 sales are weak for the next year, might they pull a repeat of the Kin, cut their losses and leave the market to Apple and Google, et.al.?
If some of the new blog posts I’ve seen this morning are anything to go by then the phrase ‘somwhat reassuring’ is a vast understatement of the reaction to Muglia’s new statement. You’d think from some of them that Microsoft ahd just announced “Windows is dead. Silverlight is where all our efforts going” 😉
Personally, I only feel reassured in the sense that I now have something couched in slightly less alarming terms that I can point clients and prospective clients wanting to jump ship at. (I posted my own thoughts in my own blog post here http://irascianwork.blogspot.com/2010/11/silverlight-shenanigans.html )
@Vic my guess – and it is pure speculation – is that WP7 will get a native code SDK in time, alongside the Silverlight one.
Tim
I think people speculate way too much. First you speculate of what _hasn’t_ been said, and then when someone addresses that you speculate that they are in fact not telling the truth.
Soon you will fabricate a truth that will self fulfill and the world will lose out on a great dev platform in favor of HTML/javascript.
I think all that has been said by Microsoft so far is that Visual studio/expression blend will support HTML5 going forward as well. Which is all good things.
HTML/javascript is so many years behind in tooling and support that it makes me cringe everytime you are faced with it.
You get the reach but to a very high cost sadly, why can’t we have both worlds? No CANVAS tags in the world will fix this.
Niclas,
I agree that speculation has its risks, but I think you are also speculating, except not as explicitly or openly. You are implicitly speculating that MS will always tell you well in advance what they are going to do.
Responsible, informed speculation is what CIOs (and developers and users) have to do all day long. Every platform choice involves a multi-year bet, and hence implicit “speculation” about the future.
Should we speculate about the severity of global climate change? Whose word should we take? Critical, public speculation and discussion is a major strength of any open organization or society.
Vic Klien
Its got to hurt. Think about it… the only two choices for developing mobile apps for WP7 are SL and Flash (after first WP7 release). An SL mobile app will only run on WP7, so if WP7 fails -a fair possibility- then your SL app is a loss. OTOH, a Flash mobile app can be marketed to WP7, Android, and other mobile device users. As it stands now, the choice is almost a no brainer. MS is making a serious mistake if they don’t at least support SL on Android.
Supporting WP7 on Android is an interesting question. Note that Apple makes no attempt to create a “cross-platform” system, but rather focuses all their attention on making their stuff work well on their own hardware. MS apparently makes a good licensing fee when a WP7 device is sold, but if they made a SL browser plug-in for Android, they’d be giving it away for free. They don’t even support SL running inside WP7’s own browser, at least not yet.
Today’s press has stories about WP7 devices selling well in Europe. The comments section of the “Register” (often sarcastically anti-MS) story was quite positive about WP7. I plan on taking a look myself at BestBuy here in the US next week.
Vic,
You could call that speculation, but that validates what I said before, you speculate of what I haven’t said instead of taking what I say at face value.
I don’t speculate, I try to spend my energy on building the best user experience with the best available technology today, if the platform I build on today is not there tomorrow (remember we speak years) it is rarely a huge problem to reimplement the wheel. The really hard part is to come up with the correct application and features to start with, not to duplicate them.
Also application, especially in the user space, change quickly, what is hot today is not hot tomorrow, that rate of change is probably faster than the rate of change of the platform.
Change will happen no matter what I think or want, I gear my development skills at adapting quickly to change instead of trying to make a informed decision about something I cannot be informed about, thus wasting effort and energy. Thus your speculation that I believe MS will tell me upfront is actually the opposite, I expect MS not to tell me up front and it isn’t much of a problem most of the time, and less so in the web space.
This is not that same as saying I promote taking stupid decisions or not trying to see the future as best as you can. Upfront design vs incremental design debate in XP if you so wish…
I just don’t like the headline chase that happens, how about you put some of that energy building a couple of SL apps, that will persuade MS more than any article in the world. That is if you think SF is good and should stay.
Cheers,
Niclas
Niclas,
Your analogy with up-front versus incremental design is a useful one. I agree that speculating forever or too much wouldn’t make sense. At some point you have to make a bet.
Normally I don’t spend as much time “speculating”, but this week has been a reassessment point for me. I’ve already invested 2+ years in learning about WPF and SL and spent 6 months this year as part of a contract team working on a SL4 LOB app. In January I plan to get another SL-LOB contract. I am getting frequent recruiter calls. (Well… at least I have been :).
Despite their poor messaging at PDC, MS has said and done enough recently (e.g. LightSwitch) to give me confidence they will continue to invest in SL. As much as I’d like? No, but enough.
Vic Klien
We’re developing a high-end project management suite, and found HTML/JS impossible to use for responsive interactive scheduling calculation, so we investigated Silverlight. We found that Silverlight outperforms HTML/JS by well over 100x in computation and rendering. The user experience of our Silverlight app offers clear advantage over the competitions HTML/JS apps both in performance and capability (for example, none offer real time cross-project resource leveling). The net-net of it being HTML/JS was simply inadequate for our application.
Silverlight provides a true client-side programming model, whereas HTML/JS is server centric, thus, they are two entirely different beasts. We certainly recommend Silverlight over HTML/JS for business app development where interactive graphics and complex client-side math is involved. On the other hand, HTML has been, and will remain the choice for generic web presence, for example, general use websites, advertising, etc.