A brief article on the launch of Silverlight 3 and why the interest in Silverlight is eclipsing that of WPF.
The excitement behind Silverlight has been gaining in momentum throughout this year, with one of the highpoints being the MIX09 conference in March this year where the announcement of Silverlight 3 beta took centre stage. There are no signs of this momentum having dropped with the final version of Silverlight 3 being release tomorrow, on July 10th.
I will be attending the UK launch of Silverlight 3, together with our MD Gary Scott, at the MS Offices in London.
… for some reason, we in the UK are 12 days behind the rest of the world 
I came to Silverlight through its big-brother WPF about half a year ago. I was looking for a new challenge and WPF certainly delivered, the framework has an elegance that leaves its predecessor WinForms looking clunky and second-rate. Silverlight shares this same elegance, some of the finer details are missing (Triggers, 3D, double-click events!), however there are enough similarities that a WPF developer can pick up Silverlight in a matter of hours, furthermore developing applications that target both frameworks is quite straightforward.
Since making the move to Silverlight I have done very little with WPF. I have found that my Silverlight blog posts and articles have proved far more popular than their WPF counterparts. Also, our customers here at Scott Logic are showing an increasing interest in Silverlight, but rarely mention WPF. So why is this?
Most companies within the financial sector follow a similar archetype; with their desktop software being most typically developed with WinForms (or Excel!), and a seemingly equal split between companies whose web offerings are developed on a Microsoft platform (ASP.NET) or Sun Java (J2EE).
If we think for a moment about what WPF could mean for companies such as these; personally I think the answer is unfortunately very little. WPF revolutionises the way in which the UI is assembled and connects to your business objects, as a developer I think it is great! However, is that enough to convince the management that they should ditch their WinForms development and invest in WPF? I don’t think it is. Whilst the developers will enjoy the switch to WPF, what’s in it for the end users – skinning? rounded borders? (blurry fonts? – ahem). If there is no perceived end-user benefit or reduced development cost, there are no compelling reasons to adopt WPF. Interestingly Microsoft’s Tim Sneath mentioned that the early expectations for WPF were that it would be “primarily for consumer software”, although he does allude to adoption of WPF for enterprise software – sorry Tim, I just don’t see it happening from here!
The story for Silverlight is very different. Delivering interactive web applications has always been a struggle, the use of AJAX to retrofit interactivity on top of what is essentially a static presentation model, HTML, is a constant battle. However, interactivity and a rich UI is something that all end-users want … and Silverlight certainly delivers. This is the reason why Silverlight is gaining interest not just from the developers, but also from their managers (and their manager’s managers). Interestingly those companies that develop their web-offerings with Sun Java seem to be as interested in Silverlight as their Microsoft ASP.NET counterparts.
With Silverlight starting on what looks like a winning streak, what is the future for WPF? Some people are of the belief that Silverlight will eventually kill of WPF, however I do not share that view. Personally I think that WPF will be eclipsed by the shadow of Silverlight for a few years for the reasons given above. However, I think that WPF is the Tortoise to Silverlight’s Hare; as companies start to adopt Silverlight their developers will gain the expertise necessary to develop WPF applications. My thinking is that in a few years time, when these companies embark on new desktop software developments, it is not the features of WPF that will steer them towards WPF, but rather their developer skillset.
The immediate future for Silverlight looks very bright.
Or is it? … Unfortunately not. Our customers want to use Silverlight, but so often they cannot because they know their users do not have the required plugin to run Silverlight applications. The problem is that the end-users will probably not install Silverlight until they have a reason to, a sort of Chicken or the Egg problem. With statistics indicating that approximately 25% of users having Silverlight installed, this is a big problem.
So, back to the Silverlight 3 UK Launch, if there is a QA session, there is only one question that I really need to ask, and that is:
 |
Ian, as a developer I am sold on Silverlight. The release of version 3 put the icing on a cake that is already quite sweet. But what is Microsoft doing to help user-adoption of the plugin? |  |
Regards, Colin E.
Tags: News, silverlight, WPF
Hi Richard,
That is some interesting feedback there. I might just take you up on that bet. Have you seen my latest post:
http://www.scottlogic.co.uk/blog/colin/2009/07/news-from-the-silverlight-3-uk-launch-and-an-answer-to-my-question-on-silverlight-adoption/
I would guess that it will just under 50% by end of year
I agree with quite a lot of what you say – perhaps it is more an indication of the caution in the financial domain at the moment.
Regards, Colin E.
P.S. SL does not have all 3 trigger types; property, event and data – that is what I was referring to.
Hi Colin,
I have to agree that security is a concern for many enterprises, but I already know of one leading investment bank that is adopting Silverlight for a trading front-end. It’s advantages as a LOB platform are numerous and its not just about “shiny/animated” buttons etc. Speed of prototyping in SketchFlow and speed of development with decent C# developers is incredible. SL3 represents a paradigm shift in the designer/developer workflow. I am constantly amazed at how quickly Silverlight apps can be built.
The penetration argument, I believe, is moot and it’s all been done before. Flash vs Java vs DHTML etc.. Flash was at version 5 (with no actionscript) before it had 90% penetration, which it hit circa June 2002 (and that was just PCs, did not include PDAS, mobiles etc). At that same time Flash 6 was at 30% penetration!
The iPhone has revolutionised the market for mobile apps and Silverlight 3 is now in a position to really capitalise on that market. Support for touch-screen apps in Silverlight has a strong future too. Windows Mobile 7 will increase penetration of SL apps also.
I would bet my house we will see 50%+ penetration by year-end.
Just my two-pence worth.
Regards,
Richard
PS Small errata: You will find that Silverlight has had triggers since version 2.
Silverlight 3 UK launch … and the future of WPF? – Colin Eberhardt’s Adventures in WPF…
Thank you for submitting this cool story – Trackback from NewsPeeps…
Hi Oleg,
Amusing, illogical, both of those things it may be. For consumer applications I think your argument holds – if you tell the end-user that they are going to get a really cool application, and to use it they just have to download this little 4Mb plugin, then, yes, they will download it.
However, for business applications it is a little different. Often the end-users will use your software from their place of work, in tightly controlled IT environments. Here the installation of a new plugin is not something the end-user can do themselves, it is a matter for the IT team. If Silverlight is not already on their list of approved software, getting approval can be a tiresome and long-winded process.
It really is only a small obstacle, and some of our customers say “If we make this cool application, people will install Silverlight”, however, most do not want to take on that risk – no matter how small it might be.
There are other factors too. Some managers might see the 25% – 30% Silverlight penetration figures and start to wonder if this indicates problems other than simply the time taken to adopt a new technology. Perhaps … they see the low adoption figures as an indication that Silverlight just isn’t any good?
In the current climate, Financial companies have to be more careful than ever, and minimising technology risk is part of this.
Anyhow, thanks for your views – I am always interested in what other think.
Regards, Colin E.
Very well put article.
Frankly, I am always amused by this argument
“Our customers want to use Silverlight, but so often they cannot because they know their users do not have the required plugin to run Silverlight applications”
It seems so illogical. In fact, for twenty years of development I never met a customer who refused to install something if they want to use some program.
Sometimes, they fail to install, that’s true, when the deployment is buggy and plainly just does not work – and they give up. Sometimes the price was too high. Sometimes hardware requirements. Sometimes the program was not needed.
But I never met something like that:
- “Are you graphic designer? Here is your free Photoshop. Works fine on your configuration”
- “Does it require installation?”
- “Yes. It will take 10 minutes and you should not do nothing.”
- “Thanks then, but no, I will continue to use Paint”.
Nevertheless, I know this is a very popular argument right now. Honestly, I just don’t get it.
Silverlight 3 UK launch … and the future of WPF? – Colin Eberhardt’s Adventures in WPF…
Thank you for submitting this cool story – Trackback from DotNetShoutout…