(X) Hide this
    • Login
    • Join
      • Generate New Image
        By clicking 'Register' you accept the terms of use .

Silverlight and HTML5 - reflections after recent news from PDC

(9 votes)
Svetla Stoycheva
Svetla Stoycheva
Joined Sep 25, 2009
Articles:   25
Comments:   8
More Articles
5 comments   /   posted on Nov 03, 2010
Tags:   silverlight , html5 , gill-cleeren , kevin-dockx , andrea-boschin , zhivko-dimitrov , chris-anderson , walter-ferrari , phil-middlemiss , levente-mihaly
Categories:   Interviews


The recent news about the change of Microsoft plans regarding Silverlight and HTML5 have triggered a lot of buzz lately, and we, as a major portal for Silverlight resources, were of course concerned about how exactly we should interpret them.

We asked the contributors at SilverlightShow to comment on those news, and give opinion on the future of Silverlight and HTML5 technologies. Feel free to share your own opinion below this article, or inside our LinkedIn and Facebook discussions on the topic.

Gill Cleeren, www.snowball.be
Since last Friday, there has been some heated discussion on the future of Silverlight and HTML5, after Mary Jo Foley posted an article, based on an interview with Bob Muglia. In this article, Bob mentioned that Microsoft is “shifting its focus”. Many saw in this article the death of Silverlight and soon the rumor mill started. With every new article, the headlines became spicier and the #silverlight channel on Twitter only contained one message: “Silverlight abandoned by Microsoft”.

However, this included actually nothing new; the Silverlight team already said the same on their blog some time ago. The message is simple (and has indeed “shifted”): HTML5 will at some point be the cross-browser default for websites. Silverlight focuses on building applications that run in the browser AND on the desktop, even on other platforms. Those are clearly 2 different targets and there’s therefore no reason to assume that the two technologies can’t (and won’t) exist next to each other. HTML5 is still a long time away also, perhaps even more than 10 years; it’s far from being a standard. 10 years in our sector is extremely long: no one could have expected 10 years ago what the IT sector would look like today. Vendor-specific technologies such as Silverlight are capable of adapting to changes much faster as well, making them a better choice for the ever-changing enterprise environment.

Kevin Dockx, http://blog.kevindockx.com/
HTML 5 is an important next step in the HTML standard, and almost every big vendor is looking into it – Apple, Google, Microsoft, …  And Bob Muglia was correct when he stated that HTML is the only technology, at the moment, which can target (almost) each and every device.  However, what he forgot to add is that, as every developer will know, using HTML means: browser incompatibilities.  Using HTML 5, at the moment, means: (browser incompatibilities)².  There is no standard, and it’s only expected to be finalized in 2022.  What’s happening now is that every vendor is building browsers which interpret their own interpretation of what the HTML 5 standard should be.  IE9 will be the first Microsoft browser to support part of what is supposed to become a standard – can we expect end users to upgrade to this, knowing a lot of businesses still rely on IE6?  Same goes for Safari/Chrome and Firefox.  I honestly don’t like where this is going – for developers and end-users alike, this screams: “not again”.  Next to that: there are no tools available that can even remotely compete with Visual Studio or Expression Studio as far as developing HTML 5 sites is concerned, meaning: HTML 5 applications will take significantly longer to develop and will have to go through significantly longer testing cycles.

At my day job, we mainly build desktop, intranet and extranet applications for businesses.  Coming from a Windows Forms & ASP .NET background, Silverlight (/WPF) is a giant leap forward – the UI capabilities are obvious, but even more important is the hugely powerful databinding framework and the tight integration with other .NET subsystems.  We build better, more stable applications, in less time, thanks to Silverlight.  Don’t get me wrong though: I truly believe HTML 5 is an interesting development we must keep an eye on, and invest in.  But it’s not ready yet, won’t be for the next few years, and I don’t believe it will succeed in bringing to the web what we already have now with Silverlight – we can already do things with Silverlight NOW which won’t be possible with HTML 5: who knows what we’ll be capable of in 5 years?

For me, the bottom line is: if you’re looking to build a (LOB) application, whether on the web, desktop or windows phone 7 series, the first choice to consider should be a XAML application, now and in the foreseeable future.  If you’re looking to build a website, use HTML.  By which I mean: HTML 4 – HTML 5 simply isn’t ready yet.  If you’re building an application, you should use a framework designed to build applications with.  Silverlight 4 is designed to build LOB applications with.  HTML has always been about designing sites, not applications.

Andrea Boschin, MVP Silverlight (Italy), http://blog.boschin.it/
In my opinion there are a big mistake in the "Silverlight versus HTML5" debate. Many people think these technologies cover similar problems and applications, but I believe while HTML is the elective technology to write web contents (websites) there is not any doubt that Silverlight is more targeted to the application development field. I think having a CLR runtime with high-level languages, XAML with a strong tooling support and a huge set of tools and libraries like Silverlight has, let developers and companies to shorten time-to-market and write code that is more reliable of what they could get using Javascript that is the sole language available with HTML. So since Silverlight has not been made to compete in the web-publishing field, where HTML is king, there is not any fear it doesn't continue to grow in the future. Differently I believe that Flash developers have to begin the migration because there is an evident overlap between HTML5 and Flash. 

By my side I ensure you, I will continue to write my applications with Silverlight, because I really do not want to return to the hell of the "multiple-browser development" I remember in my past days when I wasted much of my time fighting with different Javascript and CSS dialects, strange HTML issues, and with huge differences in how browsers support HTML. Do you really think HTML5 will solve them?

Zhivko Dimitrov, UX/UI Design consultant
The real power of Silverlight has never been in its uniquely rich interactions, or in its compatibility - flash has those, HTML5 will too. Silverlight's strength is under the hood - developer productivity tools and the power of C#. HTML5 so far offers nothing to match these - it will provide a cross-browser platform for rich interactions, but without the programming language to enable complex business scenarios at the enterprise level. Flash did this 15 years ago and it never really made it to the enterprise solutions market, I expect to see something similar with HTML5 - it might become the choice of platform for lightweight design-oriented applications, but Silverlight will preserve its business application development grounds.

Chris Anderson, Silverlight developer, author of Pro Business Applications with Silverlight
Microsoft has invested heavily in Silverlight over the previous few years, including building a number of strategically important products on top of it (Lightswitch, Pivot, Azure portal, Windows InTune, Lync client, Windows Phone, RIA Services, etc).  That alone has always fuelled my expectations that Silverlight has a bright future as a development platform, as these products demonstrate Microsoft's faith in the technology, and their expectations that it will remain a core part of their entire suite of development platforms well into the future.  As a technology, it has matured rapidly (4 releases in 3 years), and I believe it has reached an acceptable state of maturity where a wide range of applications can now be built upon it without any ugly hacks and workarounds required (although it could always do with further improvements).  I for one would be glad to see a slow down in release cycles now that it has reached its current level of maturity, as it is difficult for many developers to keep up with its pace (even more so for authors such as myself).  

Ultimately, I've always been of the opinion (as per Bob Muglia's statement), and have said this publicly many times before, that Silverlight isn't generally suited to building websites upon (neither is Flash in my opinion), instead being better suited to business and media applications. I've always believed this (writing about business applications in Silverlight before pretty much anybody in the pre-Silverlight 2 days where the focus was primarily on media, and subsequently went on to write
a book on the topic), and will continue to do so.  HTML has *always* been the best solution for where reach is of high importance, and Silverlight is the best solution where providing rich and consistent user experiences in a narrower range of environments is your focus. That said, HTML5 is not even close to being ready as a technology (with little in the way of tooling), and has been stated by many to be somewhat akin to the functionality Silverlight 1.0 provided.  We've come much further than that, and I certainly want to keep moving forward rather than taking what I see as a huge step backwards.  I'm sure that even when HTML5 is finally ready, you will still be able to build applications in Silverlight much easier and in much less time.

The phrase "our strategy has shifted" used by Bob Muglia however is one of great concern to me (bringing up memories of Windows Mobile and IE6, which stagnated once Microsoft's strategy shifted away from them).  This is a statement that has damaged the Silverlight brand, and unfortunately will continue to do so for quite some time.  I'm still waiting for an adequate statement from Microsoft that will quell my concerns regarding this shift (particularly after the precursor to this controversy generated by Scott Barnes), but in the meantime I'm maintaining my faith in Silverlight simply because it's a fantastic technology, with huge potential in its current form - well ahead of any alternative.  I honestly can't see it going away any time soon, and I'm sure that the recent outrage from the Silverlight community will dictate to Microsoft that we're not prepared to let it slip away quietly.  In which case, this controversy may even result in a positive impact upon Microsoft's strategic plans.  It's in Microsoft's best interests to not let Silverlight die, and I think they are well and truly aware of that now.

Walter Ferrari, Silverlight developer
I think that in all of this "stuff" the positive thing was the reaction of the people.In almost all the community blogs and portals  I'm seeing a lot of comments, sometimes outcries, sometimes crushing remarks, sometimes sadness, sometimes anger.
Well, this is what matters for me. If Microsoft wanted to "measure" the critical mass of the Silverlight developers this is the answer. We are not few in number... and I'm confident about the future.

Phil Middlemiss, http://silverscratch.blogspot.com/
I think the fuss over Bob Muglia's PDC conference quote is an overreaction. I would only worry if I thought Microsoft has an "either Silverlight OR HTML5" mentality - and I don't think they do. A lot of clarification and commentary has come out since the PDC conference, and it appears that all Microsoft is saying is that they are going to be providing appropriate support in the future for a complementary technology. For a technology company, that makes sense - especially given the importance of HTML5. But the important word is "complementary"; there are so many good reasons to use Silverlight in the right situation that will not change with HTML5. And HTML5, as a standard, is still a long way off.

Levente Mihaly, @leventemihaly

While I can make predictions about Silverlight's future, HTML5 is a big question for me at the moment. Silverlight is clearly a better technology, it's easier to develop for, it has better features etc. Only time will tell if HTML5 will be picked up by the masses. The tools not there yet, there are huge differences between implementations between browser's and based on it's slower development cycle it'll be always behind Flash/Silverlight. It'll have one advantage though, it won't need a plugin. Right now the best tools to create feature-rich websites are Flash and Silverlight. If HTML5 will mature it can displace the plugins in feature-rich websites, and that's a good thing.

Silverlight's future as I see it is in:

  • line-of-business applications
  • replacing WPF in a lot of cases with its out-of-browser capabilities and more lightweight framework
  • live streaming apps and video sites where something extra is needed (like DRM)
  • becoming a cross-platform framework, running on phones, tablets, cars etc.

Personally I'll wait a little before start learning HTML5, when the platform and the tools are more mature then I might need to give it a shot. The Iphone is a good example, because it's not so easy to develop for, but since there are millions of devices out there, no one get around it.



  • -_-

    RE: Silverlight and HTML5 - reflections after recent news from PDC

    posted by SL developer on Nov 03, 2010 15:32

    All the stated above is correct, but missing the point.

    All Silverlight developers knows how awesome is Silverlight, the problem is that these developers need to convince there manages/clients that MS is not 'shifting' from Silverlight. MS should give a strong and load message about the next years roadmap (and I don't mean the pail statements in the SL team blog).

    MS just shot itself in the foot, developers are forced to stop investing in Silverlight, these are the same people that are supposed to develop apps for the Window Phone 7 ! 

  • -_-

    RE: Silverlight and HTML5 - reflections after recent news from PDC

    posted by Franck Jeannin on Nov 03, 2010 16:27

    Turning Silverlight into a platform to build application actually makes a lot of sense.

    .NET has become too big and complex to grasp for newcomers (it was easy to learn incrementally over the last 10 year but I would not want to have to start from scratch today).

    Luckily, Silverlight is a very compact, cohesive and consistent subset of .NET (80% of the features for 10% of the size) that someone can learn from scratch (once you are comfortable with it, nothing prevents you from upgrading your skills to the full .NET).

    When HTML5 is finalized and widely adopted (several years from now), it will still lack some of the features that Silverlight has today (fine-grained multithreading, ability to process binary content efficiently, etc.).

    For instance, it would be impossible to build Ormetis in HTML5 (see what is possible in Silverlight bearing in mind that there is no server involved here, the code is 100% client code http://www.youtube.com/watch?v=X8oj7kQyQhs ).

    Furthermore, JavaScript remain with HTML5 the Achilles’ heel of the Web compared to .NET languages like C# or VB (see www.exgeekblog.com).

    Finally, .NET is the only platform that let you use the same tools, skills and languages to build code for devices, PCs and the Cloud.


  • -_-

    RE: Silverlight and HTML5 - reflections after recent news from PDC

    posted by Doctor G. on Nov 03, 2010 17:41

    IMHO I think the issue is this: Imagine your are Telerik or Infragistics, and you have just hear that an MS VP state that Silverlight is still alive, but it is the no longer choice for the WIDEST deployment of RIA apps. That means the huge investment (marketing, engineering, sales, etc.) we just put into our companies Silverlight Controls is now for a niche market that is Microsoft only (don’t tell me that Silverlight on Apple <SL V1.0> or Moonlight are real choices).

    Even those of us who are individual developers were counting on the widest market for our work. Right now, that ground is being ceded to Adobe Air [Android, Mac, PC, etc.) (and I used to be a Adobe MVP in Flash :-( and I decided to invest in Silverlight skills).

    I think this is the pain. I can no longer say that my products are really going to be write once, run anywhere. That space for Microsoft tools is the currently vaporware HTML5.

    Does that clarify why this is such an outcry about the decision?

     Postscript: HTML5 will never have the deep handles into the device and platform than Air or Silerlight OOBs have. This is key in the huge mobile market. We need better unified interaction with devices, sensors, etc. You, Sirs, are loosing sight of a key market discriminator provided by SL -> its ability to provide a level of EXTREME RIA (not just industry standard HTML5  RIA) across platforms. That is a tremendous market (and Ray Ozzie told you this when he left!, and I know Ray since college, listen to him).


  • -_-

    RE: Silverlight and HTML5 - reflections after recent news from PDC

    posted by David Roh on Nov 03, 2010 19:24
    What I find extremely frustrating is the deafening silence coming from Microsoft management after all of the negative developer responses - I have 3 plus years invested in Silverlight which I do believe is the very best way to build business app and web sites (if Microsoft would just fix a few issues like text handling and SEO).
  • -_-

    RE: Silverlight and HTML5 - reflections after recent news from PDC

    posted by Thanigainathan on Nov 06, 2010 19:59

    Hi All,

    If HTML5 becomes complete then waht happens ?

    Companies will start building the tools to build HTML 5 based applications. For example if MS would make a tool then they will do it and definitelt they will build a system which has the C# or VB.Net or any other lang which will generate the intense Javascript to do the task of animation or business application.

    Well the problem will be cross browser. Instead of adopting to this its better to adopt to install the silverlight plugin. We can tell this to the clients if they are feared about the MS decision on Silverlight.



Add Comment

Login to comment:
  *      *