Microsoft: Silverlight still has a future in an HTML 5 world

[Shayla]

At Microsoft?s Mix ?10 conference in March, where Microsoft took the wraps off its HTML 5 vision for Internet Explorer 9, there were a lot of questions about the future for Silverlight.

Six months later ? and one new version of Silverlight (version 4, released in April) later, those questions still persist. Should Web developers target Internet Explorer 9 or Silverlight when creating new Web and mobile applications?

Microsoft hasn?t issued any kind of white paper or definitive guidance. The official IE executive positioning has been HTML 5 is the future of the Web ? but Silverlight?s not dead (yet). This week, however, the Silverlight team finally stepped up to the plate to defend the future of that platform.

Brad Becker, Director of Product Management, Developer Platforms, took the bull by the horns with a post entitled ?The Future of Silverlight? on the Silverlight Team Blog. Becker repeated the party line that HTML 5 is nowhere near done and is a standard in flux. But he also made the case that Silverlight is the right solution for developers looking to create premium media content, as well as content/apps that can run on multiple platforms.

There is still confusion/overlap around business/enterprise apps and consumer apps. Should a Microsoft-centric developer target IE or Silverlight in those cases? Becker?s answer seems to be that it?s up to programmers and they should go with the platform to which they?re most comfortable coding.

I?m hoping Microsoft is prepping some kind of more definitive guidance for developers as to when to target Silverlight vs. when to target IE/HTML 5. (Given how often I?m asked by developers what Microsoft?s stance is here, I bet a white paper or cheat sheet would be welcome.) Maybe once the Softies are ready to talk about Silverlight 5 Microsoft will have figured out its own positioning and politics?

Source : ZDNet

[Mattytommo]

I often have discussions about this at work, as we are looking for interactive content for our company's web app, but are unsure whether to jump on the HTML5 or SilverLight bandwagons. There seems to be no definitive response from either, but Microsoft arguing that "it's not finished yet" as their best defence proves that it's going to prompt a decline in the need to use Flash/Silverlight for interactive content.

One thing I'm not sure of however, is which performs faster. Some multmedia content written entirely in Javascript, is that faster than SilverLight or vice versa?

[hdood]

If you can't do the interactive stuff you want with current standards (not HTML5), then go with Flash. HTML5 is not an option because the standard is not finished yet, and the browser's have not fully implemented it either. In fact, IE7 and IE8 do not support it at all. By going with HTML5, you would be alienating an extremely large percentage of the market. Silverlight is not quite as bad a choice, but it's still a poor choice because it's not as widely deployed as Flash.

[Fraud OS 11]

Of course, it definitely has. People seem to have the wrong impression that HTML5 will do everything (Silverlight and Flash do). Silverlight is ****ing amazingly rich and MS can have their own proprietary innovations put there as not everything can be done the HTML5 way.

[~Johnny]

There's also the nice fact that you can be pretty confident that what ever you write for Silverlight will work exactly the same in every browser - not something you can say about HTML 5. Plus, Silverlight does offer a lot more, it's not a mere flash clone that people play it down to be =/

[hdood]

Plus, Silverlight does offer a lot more, it's not a mere flash clone that people play it down to be =/

Virtually everything I've seen Silverlight used for on the web could have been done in Flash as well though. The only reason to use Silverlight in most cases is because you are a Microsoft shop and don't know how to use anything else. It's dumb to choose Silverlight when Flash will let you reach a wider audience, and in my opinion any contractor who suggests it should be fired and blacklisted.

[Kirkburn]

Virtually everything I've seen Silverlight used for on the web could have been done in Flash as well though.

That's not the same thing. Just because implementations you've seen could be done in Flash doesn't mean all implementations can be, or that Silverlight wasn't a better choice in that situation.

The only reason to use Silverlight in most cases is because you are a Microsoft shop and don't know how to use anything else. It's dumb to choose Silverlight when Flash will let you reach a wider audience, and in my opinion any contractor who suggests it should be fired and blacklisted.

If Silverlight lets you do something better, then sure you should choose it. You're putting way too much emphasis on marketshare over anything else.

PS. Pretty sure MS engineers know how to use Flash.

[hdood]

That's not the same thing. Just because implementations you've seen could be done in Flash doesn't mean all implementations can be, or that Silverlight wasn't a better choice in that situation.

I said that virtually everything I've seen Silverlight used for on the web could also have been done with Flash. If it could also have been done with Flash, then Silverlight can't be a better choice. It's very simple. Similarly, if it can be done in all relevant browsers without using Flash, then Flash is not the best choice.

If Silverlight lets you do something better, then sure you should choose it. You're putting way too much emphasis on marketshare over anything else.

It's the only thing that matters.

PS. Pretty sure MS engineers know how to use Flash.

I'm not sure what point you're trying to make. Flash isn't a better choice because Adobe uses it on its websites, it's a better choice because it will let you reach a wider audience.

[Guest]

Silverlight is currently installed on over 60% of all machines... thats impressive.

http://riastats.com

[jakem1]

Silverlight is currently installed on over 60% of all machines... thats impressive.

http://riastats.com

Be careful. Those sorts of stats might make hdood question is irrational hatred of MS ;)

[hdood]

Silverlight is currently installed on over 60% of all machines... thats impressive.

http://riastats.com

Impressive in context maybe (or maybe not, considering Microsoft's sneaky distribution), but 8.6% for v3 and 53% for v4 on the sites polled is not impressive when compared to 93% for Flash 10 and 2.8% for Flash 9. It should be obvious which platform is the better choice.

[jakem1]

No, it appears he still has that irrational hatred of MS.

[hdood]

No, it appears he still has that irrational hatred of MS.

How on earth is it "irrational hatred" to point out that it is better to choose the solution that can reach over 90% of the market over the one that reaches 60%?

[andrew_f]

For me, part of the 'open web' is the ability for anyone to try and expand on standards with their own thing. Flash and Silverlight (although I'm not a huge fan of either) are welcome in my book, because it'll keep the web moving forward.

[jakem1]

How on earth is it "irrational hatred" to point out that it is better to choose the solution that can reach over 90% of the market over the one that reaches 60%?

Your hatred is irrational because you just turn up in every Microsoft related thread and bash them for no reason.

Your 90% marketshare argument is absolutely pointless in this case. 60% vs 90% is irrelevant, especially considering that, like Flash, Silverlight is free and widely available. There are no limitations preventing its wider acceptance and installation is just as quick and easy as Flash installation. Given these facts, it makes little sense for developers with .NET experience (and there are exponentially more .of them out there than there are Flash developers) to learn Flash from scratch when they would find Silverlight a lot easier to learn. I mean, are you seriously suggesting that you'd rather take the time to learn a whole new development language and familiarise yourself with a whole new set of tools rather than just add a link telling your users to download a runtime package?

I could understand if you were arguing that Silverlight doesn't support key features that Flash does but your marketshare argument is particularly weak.

[hdood]

Your hatred is irrational because you just turn up in every Microsoft related thread and bash them for no reason.

That's how you choose to interpret it. I don't know if you're a Silverlight developer with a grudge or something.

Your 90% marketshare argument is absolutely pointless in this case. 60% vs 90% is irrelevant, especially considering that, like Flash, Silverlight is free and widely available. There are no limitations preventing its wider acceptance and installation is just as quick and easy as Flash installation.

Flash is already present on 93% of systems, including a large number where Silverlight cannot be installed by policy. It will always be a better choice than gambling on people caring enough that they choose to go to the trouble of also installing Silverlight (if they even can.)

Given these facts, it makes little sense for developers with .NET experience (and there are exponentially more .of them out there than there are Flash developers) to learn Flash from scratch when they would find Silverlight a lot easier to learn. I mean, are you seriously suggesting that you'd rather take the time to learn a whole new development language and familiarise yourself with a whole new set of tools rather than just add a link telling your users to download a runtime package?

Who cares what a .NET developer wants? If he wants to develop in .NET, then he should. However, if you run a company that wants something interactive on the web, you don't hire the .NET developer, you hire someone who can reach as many people as possible.

You're like one of those people who were always arguing that sites shouldn't support IE6 even when it had 30% of the market, "just because." It's not about ideology.

[Argi]

I'd like to see HTML5 do adaptive stream like Silverlight. Never going to happen, though.

[Elliott]

I'd like to see HTML5 do adaptive stream like Silverlight. Never going to happen, though.

While not technically HTML5, Apple's proposed and open HTTP Live Streaming supports adaptive streaming.

[Argi]

While not technically HTML5, Apple's proposed and open HTTP Live Streaming supports adaptive streaming.

That's good, and I didn't hear any major issues with the event stream last week but the penetration of that technology is just Apple at the moment.

Honestly, I was all for HTML5 video when news started going around about it but since then all I've read is more and more on what it can't do so I feel the future is still in plugins at it stands. HTTP live streaming seems pretty nice but as with anything it's just adoption.

[wellofsouls]

Silverlight is currently installed on over 60% of all machines... thats impressive.

http://riastats.com

well, not that impressive considering Microsoft has been pushing it via Windows Update.

Virtually everything I've seen Silverlight used for on the web could have been done in Flash as well though. The only reason to use Silverlight in most cases is because you are a Microsoft shop and don't know how to use anything else. It's dumb to choose Silverlight when Flash will let you reach a wider audience, and in my opinion any contractor who suggests it should be fired and blacklisted.

From the technological and developer point of view, Silverlight is vastly superior to Flash. First, there are the whole range of .NET programming languages you can choose to develop with, and most are superior to ActionScript, and mostly bug-free compared to the gazillions of bugs and quirks in the Flash controls/ActionScript interactions. Yes, Flash can do most things Silverlight can do, but a lot of times Flash has to do it with tons of controls, strange hacks, and messed up code, while Silverlight can do it with clean, well-organized and easily readable code. Not to mention Visual Studio being a vastly superior programming IDE than Adobe Flash Professional.

The .NET platform is also much more efficient and secure.

As for which one to use, I'd say use Flash when you are building a public web site, since it is the de facto standard out there. But if you are building an internal web app for a company environment, Silverlight is hands down win. As for HTML5, it won't be able to replace either Flash or Silverlight. Maybe HTML7 or HTML8, but then based on the current trend, developing complex web apps in HTML, CSS and JS is just a pain in the @ss compared to using the .NET platform.

[pezzonovante]

Flash is slowly dying. Just like native Win32 API based programs are almost dead. Silverlight and .Net are the future of computing.

[hdood]

As for which one to use, I'd say use Flash when you are building a public web site, since it is the de facto standard out there. But if you are building an internal web app for a company environment, Silverlight is hands down win.

Exactly. When dealing with the public you want something that is accessible to as many as possible. What's best from the technical side and what has the best tools is secondary. What you use internally doesn't really matter, since you have control over all the systems there anyway.

Flash is slowly dying. Just like native Win32 API based programs are almost dead. Silverlight and .Net are the future of computing.

It's a bit of a stretch to call native win32 code almost dead, when most software is native win32 (including all of Windows itself). .NET might have a bright future, but today it is a minority platform. Out of the 11 programs I currently have running, only one is .NET.

Silverlight and .Net are the future of computing.

I'm all for sticking it to Adobe, but Flash being replaced by Silverlight would not be much of a victory. It just makes things more closed and gives Microsoft more power.

As for the difficulty in developing for HTML5, that's more down to a lack of tools. If someone managed to release awesome tools, it would be a victory for all. At least years down the road when HTML5 is actually implemented in the browsers most people use. It's not something that is "right around the corner," so any discussions about replacing Flash with it are way premature.

[HawkMan]

Except sivlerlight is infnintely more open than flash.

it's faster. uses less resources, works better, can do a lot more. and so on.

[jakem1]

You're like one of those people who were always arguing that sites shouldn't support IE6 even when it had 30% of the market, "just because."

Well I'm not sure how you reached that conclusion but if we were to follow your argument to its logical conclusion we'd all be using IE6 because it was the market leader so there was no point supporting an alternative.

It's not about ideology.

The only problem is that it's always about ideology for you. You ignore the facts to simply push you anti-MS position. You were the one that started spamming an article about Silverlight and HTML5 with nonsense about Flash and then completely ignore the fact that Silverlight is in many ways a better alternative, either because it's more capable or because there are more people that are familiar with related technologies. Nobody said that there's no place for Flash but to dismiss Silverlight's massive install base and the large pool of talent available to develop for it is just plain stupid.

Hopefully there aren't more people with as narrow-minded an attitude as yours because it's that sort of defeatism that has lead to Flash being the bloated monster that it is today. You're whole argument boils down to supporting the market leader rather than the better product.

[pezzonovante]

It's a bit of a stretch to call native win32 code almost dead, when most software is native win32 (including all of Windows itself). .NET might have a bright future, but today it is a minority platform. Out of the 11 programs I currently have running, only one is .NET.

How many developers learn the Win32 API these days? I would think less than 1%. Even the legendary Charles Petzold no longer prefers the Win32 API. The last edition of Programming Windows was released in 1998. There's absolutely no reason to choose the Win32 API instead of .Net for building new applications.

I think 5 years from now, Win32 API developers will be an almost extinct breed. Hence I believe Win32 is gone.