none
Silverlight 6 Wishlist

Answers

All replies

  • - Compile Silverlight to native HTML5, CSS3 and JavaScript for In-Browser-Apps. :-)



    • Proposed as answer by willrob68 Saturday, May 25, 2013 9:36 AM
    Wednesday, April 13, 2011 3:41 PM
  • Jack, you're a trouble maker, lol.

    SL 5 hasn't even shipped, but the early bird gets the worm(and they tell me Apple is full of them).

    Wednesday, April 13, 2011 4:54 PM
  • +10 on compiler/converter to HTML5.

    But, without that... Android and more alternative platform support.

    Wednesday, April 13, 2011 5:15 PM
  • @samcov: I agree. :)

    @mtiede: Yes, it would be great. I've just inserted into dotnet.uservoice.net, vote for it here: http://dotnet.uservoice.com/forums/4325-silverlight-feature-suggestions/suggestions/1713161-compile-silverlight-to-html5-css3-and-javascript-?ref=title


    Wednesday, April 13, 2011 5:42 PM
  • Android and more alternative platform support.

    Yep. They've totally missed the point on that. Balmer had his little "Developers, Developers, Developers" tantrum a few years back, but what exactly are they doing for developers today? Here's what they could be doing. Amazon has an Android store, why doesn't Microsoft? Obviously they'd need a Silverlight runtime for Android, but then it would be trivial for me to port my app. Now that would be doing something for developers.

    Wednesday, April 13, 2011 5:46 PM
  • That's hat I use as workaround today:

        <sys:String   x:Key="ClosePath" >
            F1M171.964,69.0358L169.982,67.0534 166.988,70.047 164.01,67.0694 162.02,69.0599 164.997,72.0375 162.028,75.0071 164.01,76.9895 166.98,74.0199 169.973,77.0135 171.964,75.0231 168.97,72.0294 171.964,69.0358z
        </sys:String >
    
        <Style TargetType="Candy_Controls:PathButton" x:Key="CloseButton">
    
            <Setter Property="PathData" Value="{StaticResource ClosePath}"/>
    
        </Style>


     

     

    Wednesday, April 13, 2011 11:50 PM
  • my wish would be direct database access at least in full-trust. RIA services are very helpful but still I'm so much faster in WinForms where I can query data directly from the database.

    Thursday, April 14, 2011 12:02 AM
  • support no html -5 csss .

    iphone native app support

    Reporting tool


    Thursday, April 14, 2011 1:29 AM
  • SL5/SL6 wishlist: SelectionBackground and SelectionForeground properties for RichTextBox control.

    Thursday, April 21, 2011 5:48 AM
  • I guess I have to whine about Android... AGAIN!

    This was disturbing!  I went to my local Ralphs Supermarket here in Southern California, and they had two large stands selling Android Phones and contracts like loaves of bread!

    What the Flying Crap is that?

    These guys are in it to win it, and we better get our stuff together, and I mean soon!

    Friday, April 22, 2011 3:11 AM
  • Auto generate the logic for editting relational tables in datagrid.

    replace all IDs with a field of the parent table.

    Friday, April 22, 2011 4:12 AM
  • Thomas: you hit the nail squarely.

    I have been in a mire of fearless depression since Mix's Keynote.  It's one thing to take the rug out from under us, but another to not have any backup plan/framework for us to continue.

    That is a double-whammy.

    I think what we're looking for is a Microsoft Web Toolkit (MWT).  It would work precisely like WPF/Silverlight but compile into javascript/css/html.

    This should be the future of Microsoft.  However, watching the two stooges up during the keynote (it would be 3, but scottgu continues to be the sole voice of reason) gives me no measure of confidence.

    In the meantime, I've been looking into assmire that is jQuery and co.  Ahhh... the stench of sprawling never-ending javascript files with 2px-high scrollbar carets.  Mmmm... how I miss thee.

    Friday, April 22, 2011 11:34 AM
  • Have you investigated Script#?

    See the video here - Script#: Compiling C# to JavaScript using Visual Studio

    Friday, April 22, 2011 5:36 PM
  • It would work precisely like WPF/Silverlight but compile into javascript/css/html.

    That's like handing someone the ingredients of an omelett and asking them to give you a car. If it was possible to do something like that, we wouldn't hate HTML so much. First off C# and VB.NET have features that the morons on the JavaPuke committee have never heard of. In WPF we can create our own controls, in HTML there are what 5? If you to define your own radial container control that rotated a child based on an attached property, how exactly would that be expressed in HTML? The shift in strategy was driven by total effing morons within Microsoft. Any real programmer knows you can't build a foundation on crap, and that's what HTML is. Microsoft tried HTML Applications before, and they sucked, the addition of a Canvas tag isn't going to change that.

    Just think HTML is already woefully behind Silverlight, and comparing it to the full WPF is just an insult. When HTML is basically where it is now in three, five, or even seven years, are we seriously still supposed to be developing against it? Think about it, if the HTML morons were to start tackling data binding today, does anyone realistically believe it would be ready before 2016? And that's being optimistic. If history is any guide, they'd have a joke of a first draft sometime in 2022. Are we honestly supposed to wait a DECADE for data binding in web browser apps? And please, I'm not talking about some sort of hacked up JavaPuke library, I'm talking native runtime support.

    Why is it that Adobe is all for Flash on lots of devices, but Microsoft TOTALLY AND COMPLETELY ignores the number one request by ITS DEVELOPERS?

    http://dotnet.uservoice.com/forums/4325-silverlight-feature-suggestions

    More platforms and better binding. And what do we get, no more platforms, and virtually no more binding support. Thanks for listening Microsoft.

    Friday, April 22, 2011 5:50 PM
  • Actually Jack... it is possible to do this.  It would just take a LOT of effort and manpower/hours. :)

    Please look at GWT. Google does it with no problem.  I am currently researching this technology to fully understand how it works (and if it is what I think it is).

    I'm not sure where/what all those Silverlight engineers within Microsoft are doing now, but this (a Microsoft version of GWT -- Microsoft Web Toolkit) is precisely what they should be doing.  Since HTML5 is a (somewhat) standard, they should be catering to this and creating an adaption infrastructure that translates Silverlight (or sub-WPF) applications into a target platform, such as HTML.

    Actually, what Microsoft should be doing is creating a packaging architecture that caters not only to HTML5, but to Droid, iOs, and BlackBerry as well.  They should not be asking developers to regress back into JavaPuke (as you call it... and agree!) and HTML as a development paradigm.  It is Microsoft that should be "protecting their investment" in preserving the integrity and superiority of .NET/WPF architecture, and taking on the burden of translation between platforms.  This is what they are good at.

    Developers should simply develop applications.  Microsoft should be creating:

    1. Tools
    2. Frameworks
    3. Adapters to other Platforms

    They do a great job with the first two (in fact, it is the BEST thing that they do!), but the 3rd is where the pain point truly is, and precludes us developers from our supreme maxim of "Develop Once, Run Anywhere."

    The next "Big Announcement" is in September.  Let us pray for adaptation.  Otherwise, I just might defect to GWT.

    Monday, April 25, 2011 9:43 AM
  • Actually Jack... it is possible to do this.

    No, it's not. You might be able to get something that uses a subset of WPF controls that outputs some sort of html equivalent, but anything remotely close to a 1 to 1 translation is impossible. Like I said before, HTML simply doesn't have the infrastructure to support Silverlight functionality.

    Please look at GWT.

    It's a web toolkit, hardly anything revolutionary. Does it support databinding? Threading? Extension of native controls? More than one container, i.e. Grid, Canvas, StackPanel, MyPanelThatDoesExactlyWhatIWant, DockPanel, etc, etc? Thread locks? LINQ? Strongly typed variables? Sound? Printing? Encryption of files on the client and sent via web service to the server? Isolated Storage? Etc, etc, etc.

    (a Microsoft version of GWT -- Microsoft Web Toolkit) is precisely what they should be doing.

    Please no. The world doesn't need ANOTHER web toolkit.

    Since HTML5 is a (somewhat) standard

    Ahahahahahahahahhaha. One of the HTML standard bodies has said, "we're not even going to try and define versions anymore." That's how pathetic they are. The only thing they can agree on is not to agree on anything.

    Actually, what Microsoft should be doing is creating a packaging architecture that caters not only to HTML5, but to Droid, iOs, and BlackBerry as well.

    Microsoft developers want a SILVERLIGHT port to Droid and iOS, Blackberry is dead. As a developer if you had support for Windows 7 phone, Droid, and iOS, would you really be missing any significant opportunities? No? So that's TWO more platforms. TWO. We're not talking a monumental effort here. Adobe is doing it.

    preserving the integrity and superiority of .NET/WPF architecture, and taking on the burden of translation between platforms

    Yes, by actually porting it to those other platforms, not attempting to convert it to puke via some horrid translation.  

    but the 3rd is where the pain point truly is

    Well, we're in agreement here, we just disagree on how to best address #3. I think the VAST majority of Microsoft developers want a Silverlight port. The old Microsoft would have embedded Silverlight into IE9 and called it HTML 7. That way corporate developers who don't give a crap about running in more than one browser could standardize on Silverlight. That's LEVERAGE. The old Microsoft used it, the new clueless Microsoft is DELEVERAGING itself by sucking up to standards that buy them NOTHING. Well played.

    Monday, April 25, 2011 12:55 PM
  • I'd like much better support on the Mac. The performance is poor in comparison to the PC plugin version. GPU support on Mac would be a start.

    Jack - I have to say that I agree with many of your points! Especially the one about ignoring the highest voted feature (more platforms). Maybe MS need to remember what the E stood for in WPF/E!



    Monday, April 25, 2011 12:58 PM
  • LOL Jack... I appreciate the enthusiasm.

    I think you underestimate the flexibility of Javascript.  I absolutely hate it as a development language, but it is incredibly flexible.  Just look at jQuery (and jQueryUI).  You can do a lot with it.  I'm sure if you put a few teams of Silverlight's smartest developers together in a building for a year, they could come up with an adaptation to HTML5.

    My point with GWT was just conceptual: developers develop in Java and *Google* carries the burden of compiling *down* to Javascript.  Developers focus on Java, and Google takes care of the dirty work.  This is what I feel Microsoft should do.

    As far as all the bells and whistles of what GWT doesn't do (which I think you'd be surprised on what it can do), it's just a matter of time to get it ported.

    Like what geedubb said: the premise of Silverlight isn't a just plugin.  It goes much deeper than that; it is meant to be a development platform that caters to the supreme developer maxim: develop once, run everywhere (someone called it a development brand... I like that, as well).

    I agree with you Jack that the politics stuck it to us this time around.  Which is just a shame.  Like you said, we aren't even being listened to... and probably worse: dismissed.

    Tuesday, April 26, 2011 8:28 AM
  • I think you underestimate the flexibility of Javascript.

    Nope, it defines virtually nothing, so people can do unbelievably horrific hacks with it. Consequently it's a maintenance nightmare.

    Just look at jQuery (and jQueryUI)

    An almost necessary add-on to JavaPuke to compensate for a ton of missing functionality that should be in any modern language.

     

    I'm sure if you put a few teams of Silverlight's smartest developers together in a building for a year, they could come up with an adaptation to HTML5.

    I could take a bunch of brain dead monkeys, shoot them up with crystal meth, and use them as crash test dummies. Following those tests, they could come up with an adaptation of POS5. In realistic terms, Silverlight is the functional equivalent of POS23 (honestly that's how far behind POS is.)

    This is what I feel Microsoft should do.

    Why? Another horrific web toolkit? I can just imagine trying to debug that garbage. Call me crazy, but I don't want my compiled and packaged code converted to crap where something is ALWAYS lost in the translation. And that assumes that all browsers are rendering the crap identically. HTML isn't part of the solution, because it's the problem.

     it is meant to be a development platform that caters to the supreme developer maxim: develop once, run everywhere

    Well it could be if Microsoft actually listened to its developers. I wasn't at MIX but from blog posts I suspect it wasn't a hit. I really wonder if Microsoft heard the thud. Fortunately the Windows phone group hasn't been inflitrated by the supreme morons yet, so I can do WPF style development for the phone and make money on apps. Given Microsoft's new "skip Silverlight development on the PC" strategy, I'm quickly losing interest in web applications. And to anyone who says, what are you talking about, look at all the new stuff they put in v5, my response is that it's obvious Microsoft has significantly scaled back development. Thanks for the 3D, but where is the:

    Native menu control

    MultiBinding

    PLINQ and TPL

    FlowDocument

    JPEG XR

    All of these things have been highly requested, not to mention Android and XBOX support. What we got in v5 is clearly not the result of a year's worth of work from a dedicated team. The 3D support seems like it was just enough to keep on par with Flash, but almost nothing more. Either that or it was a side benefit from work being done for the phone version. My original request for PathGeometries as resources is an issue that people have been asking for a fix since v2, and yes there is a work around. Unfortunately that work around is worthless in the designer.

    Like you said, we aren't even being listened to... and probably worse: dismissed.

    Unfortunately you're probably right. Sometimes I wonder if Microsoft has hired Scott McNealy and Ed Zander who are now working to secretly destroy Microsoft from within. It's one of the few theories which explains Microsoft's inexplicably moronic shift in strategy.

    Tuesday, April 26, 2011 1:18 PM
  • LOL!  Wow.  I appreciate the discussion here and would like to join you in the energy towards the trainwreck that was Mix11 and the "strategy shift" that has been employed (I'm also digging the conspiracy theory. ;))

    Consider this workflow:

    1. Developer creates .NET application.
    2. Developer creates adapter package.
    3. Developer deploys adapter package (to target platform: Legacy HTML/HTML5/Droid/iOS/WPF/?).

    The "adapter package" in this case is the theoretical Adaptation Framework I'm pawing at here.  The developer does not have to worry about Javascript hacks.  In fact, the developer doesn't even ever have to touch Javascript.

    The Burden of Adaptation is handled by Microsoft.  Microsoft engineers are the ones who handle all the Javascript hacks and implementation in an Adaptation Runtime (still working out the theoretical concepts here).

    The developer develops once (in a native environment with Xaml, Behaviors, Triggers, etc), and then creates adaptation packages that are created through an Adaptation Compiler/Runtime/Pipeline/System that is maintained by Microsoft (or other 3rd party technology houses such as Novell/mono).

    Your hatred for Javascript is absolutely understandable.  No way do I (having gone through the mire that is javahacking) want to go back into that wretched land.

    All of us .NET developers, having tasted the greatness of the WPF paradigm, cringe in disgust at the thought of returning to standards-land.

    What I'm suggesting is a fundamental shift (a TRUE strategy shift) in how Microsoft applications are created: through client adaptations.

    I'm actually going to create an overview/diagram of what I'm thinking.  Maybe that will clear it up for you (and me!).  Stay tuned.

    Tuesday, April 26, 2011 1:37 PM
  • (I'm also digging the conspiracy theory. ;))

    No conspiracy required, stupidity explains it easily enough.

    What I'm suggesting is a fundamental shift (a TRUE strategy shift) in how Microsoft applications are created: through client adaptations.

    Microsoft's old strategy was to improve defective frameworks, hence all their fantastic proprietary additions to HTML. Not surprisingly, those additions were roundly adopted because they made sense. Microsoft INVENTED AJAX. Would the new braindead Microsoft do that? Nope, that might offend Richard Stallman.

    Maybe that will clear it up for you (and me!).

    I get what you are saying, but it can't work due to the limitations of HTML/JavaPuke. If you want to go through the exercise, just start looking at the Silverlight documentation and try to figure out how you would implement things in JavaPuke. You'll be able to tackle some of it, but your JavaPuke will be HUGE, and that's only what you were able to convert.

    Tuesday, April 26, 2011 1:52 PM
  • Image Compression please, I cannot print MBs of Bitmap, my printer sucks. Is there any way to compress  the Silverlight created huge bitmap to cool jpegs ??? :(

    Tuesday, April 26, 2011 2:25 PM
  • ...

    PLINQ and TPL

    FlowDocument

    ...

    Thanks for the info. I just searched the web for this. Unfortunately it didn't made it into SL5. Sometimes it's really annoying to wait so important features. However, if the community would bundle all it's efforts into a single open source project that implements the most missing features in high quality then we wouldn't suffer so much. e.g. for TPL you find two implementations on the web but I dont't feel comfortable to use them in real world apps. I don't say that they are badly implemented but I don't believe that a single person could provide the necessary support. With other feature it's the same, you find implementations here and there but I would prefer to have a well tested and supported toolkit.

    So who wants to start the  Silverlight_MS project. :) (MS stands for MISSING STUFF)

    Tuesday, April 26, 2011 7:45 PM
  • So who wants to start the  Silverlight_MS project. :) (MS stands for MISSING STUFF)

    Call me crazy, but I don't do volunteer work for MULTI BILLION DOLLAR CORPORATIONS. (Just look at all those MySQL saps who did volunteer work to make Larry Ellison richer, I'm sure he appreciates it. Maybe he'll send them a thank you card from one of his yachts.) As Microsoft developers we shouldn't have to do this stuff.

    One other point on this, there are simply some things that can't be done effectively in non-privileged code. The menu is a perfect example of this. If the user clicks outside the menu, only the underlying system can effectively track it without resorting to hacks.

    Tuesday, April 26, 2011 7:55 PM
  • "Going a step further, Adobe is also enabling developers to test their applications with emulators and tools for Android, Blackberry and even Apple's iOS. Apple's SDK for iOS runs on only Mac OS X, yet Gruber stressed that a developer could use Flex and Flash Builder 4.5 to build an app for iOS, without the need to be running a Mac."

    Full article here - Adobe Extends Flash Builder for Mobile and PHP

    Wednesday, April 27, 2011 2:08 AM
  • "..., without the need to be running a Mac."

    Hmmm....  I thought that was against Steve's rules.

    Wednesday, April 27, 2011 8:13 AM
  • Full article here - Adobe Extends Flash Builder for Mobile and PHP

    Wow... that's just depressing.

    Wednesday, April 27, 2011 9:43 AM
    • Full Richtextbox support that allows saving the Byte[] of anything (including inlineUIContainers).
    • Better DOM support.  Any Trusted SL App should be able to gain unfettered access to the HTMLDocument and do anything it wants with respect to what the browser will allow. The same support as is currently in WPF.
    • An improved HTML DOM wrapper, get away from MSHTML and continue and improve System.Windows.Forms.Webbrowser.
    • Ok, I'll bite, allow C# to compile into Javascript byte code (IL) equivalent.
    Wednesday, April 27, 2011 10:03 AM
  • Full article here - Adobe Extends Flash Builder for Mobile and PHP

    Wow... that's just depressing.

    Exactly what I was feeling...

    Wednesday, April 27, 2011 2:34 PM
  •  

    Full article here - Adobe Extends Flash Builder for Mobile and PHP

    Hmmm...looks like further proof that Javascript and HTML 5 win. 

    Maybe it's time for us to look into Adobe?  Which I already did 7 years ago and rejected.

     

    Wednesday, April 27, 2011 3:37 PM
  • I was thinking the same thing.  Adobe just covers the client-side, obviously.  They don't have a .NET framework for service-side (and goodies like WCF Ria Services).  You would be giving one thing up for another.

    This is just a mess.  You would think that over a decade of framework development that there would be a clear answer to stay on top of dynamic market forces.  Microsoft just has not been adaptive enough.

    Wednesday, April 27, 2011 3:43 PM
  • One thing that's clear, it's way too early to call the winner.

    I need HTML 5 to be solid, because for the smaller stuff, RIA(SL & Falsh) are just plain overkill.  Add to that the fact that MS failed to hear us when we said they needed a really robust text engine in SL, way back in Silverlight 2 beta!

    HTML wins when we're talking simple text oriented sites, well, it is the Hyper Text Markup Language after all.

    However, after you get past simple to moderate complexity, you need an web application, not a web site.  These two are twins, they are designed to work together, not compete.

    Don't fear HTML, embrace it, because what I'm seeing is that customers need both, and that's been our strategy from the beginning because SL never had the install base to be used on the internet anyway.

    Finally, the real shake out will occur when MS unleashes Windows 8 Tablets.  IMO, they will dominate like nobodies business if they can get them out by Christmas.

    Why?  Well, cell phones are a consumer play, but tablets are an Enterprise play, just as computers were.  The large companies will drive acceptance of tablets, and they're waiting for one that does everything the iPad does, but satisfies the real needs of a business user for work and play.

    That tablet, IMO, is the Windows 8 tablet... stay tuned!

    Thursday, April 28, 2011 5:32 AM
  • Good point of the MSFT buisness focus, but big Elephants do move slow as IBM did in late 80's and 90's.  However, once they arrive there's no mistaking their footprint.  SL to me is currently in a wait cycle, or is being sunsetted.  The puny release of SL 5.0 was nothing really.  SL needs to be WPF in a browser.  The only restriction should be that the developer must have a trusted root certificate to achieve unfettered client and cross domain access.  Imagine what that would bring to the table. 

    As far as IDEs?  Adobe Flex 4.5 is looking a lot like VS2010 but they'll never be able to have class like support.  That is, unless JQUERY becomes the Javascript library of which reflection would power Flex intellisense.  Now that could be a SL killer indeed.

    Is MSFT telling us that's the direction they're headed? HTML 5 and Javascript using JQuery as class libraries? Hmmmmm....

    Thursday, April 28, 2011 12:24 PM
  • Don't fear HTML, embrace it, because what I'm seeing is that customers need both, and that's been our strategy from the beginning because SL never had the install base to be used on the internet anyway.

    HTML is the problem, there's no point in embracing it. That's like defending your abuser. The old Microsoft would have either bundled Silverlight with Windows or simply called it HTML 7. Corporate developers would flock to it in droves because they don't care about multi-browser support. The longevity of IE6 is proof of that. Why is it that I've NEVER heard an Adobe rep say don't use Flash, but Microsoft employees say don't use Silverlight all the time. It's moronic. Why not, "use it for whatever you want, it's hands down the best web platform by far." The navigation framework is proof that at one point they had grander plans for it. Bundling with Windows and extensive use on Bing and installed base would be a non-starter. In fact, they were making great progress, and then for some inexplicable reason, with two seconds left, they punted from the 1 yard line down by 6.

    With Windows phones and tablets, Microsoft developers have the nice option of simply withdrawing from internet development. Microsoft had given us the internet option as well, but this pathetic v5 release shows they've temporarily (hopefully) lost their minds for HTML5. If Microsoft were to put out a solid App store for Windows then the loss of the internet option wouldn't matter much at all. As for the new strategy, are there any Microsoft developers who are actually excited about it?

    Thursday, April 28, 2011 3:04 PM
  • Couldn't agree more.  The move away from Silverlight is "inexplicable".  I hope something good is coming down the line.  Otherwise, it looks like there was some management squabbles and someone who supported SL LOST.

    I still hold out hope that they will support Android or something, but I don't hold out a LOT of hope.

    The whole thing seems a little bizarre considering how good we know the SL platform to be.

    I plan to use it for EVERYTHING I can and only resort to HTML5 if all else fails in SL.

    Thursday, April 28, 2011 3:49 PM
  • Jack, HTML 5 is only a threat in some peoples minds, it's still html, with all of it's problems.  as a matter of fact, with version 5, the old problems are back, and they're worse than ever.

    Have you seen IBM's Silverlight & Flash killer(their words, not mine), Maqetta?  IMO, it's a joke, but who knows.

    As for SL5, I think you guys must be looking at it the wrong way.  There are a LOT of new features that are awesome and I need them NOW!

    ICustomTypeProvider for creating TRUE dynamic objects without IL, is very powerful. 

    ImplicitDataTemplate is off the hook for some of the systems we need to implement.

    CustomMarkupExtension is another way to have code interact with XAML that I love.

    Sure, if you're a WPF person, these things aren't big deals, but I never knew WPF until I started SL programming, and it looks like heaven to me.

    Of course, those features don't get much press, and I am excited about the 3D stuff too, but those little things will have a BIG impact on application development, IMO.

    Finally, I do agree that phones and tablets give us another option, but as the standard internet turns more towards applications, the differences between HTML and RIA will become even starker.

    Thursday, April 28, 2011 5:34 PM
  • +1 @samcov

    Thursday, April 28, 2011 6:48 PM
  • HTML 5 is only a threat in some peoples minds

    Right. The only problem is some of those minds are making high level decisions within Microsoft. Those dolts have no clue what customer loyalty is, or binding, or threading, or MVVM, or LINQ, or typed languages, etc, etc. They hear, HTML5 has a Canvas, woohoo all the problems are solved. That's how brain dead they are.

    IMO, it's a joke, but who knows.

    Aahahahha, another open source HTML editor. Just what the world needs. Yes, it is a joke, and a very tired one.

    There are a LOT of new features that are awesome and I need them NOW!

    Yes, there are new features, but it's clearly a half hearted effort. The delta between 4 and 5 is nowhere close to prior upgrades. Where are the new controls, significantly improved binding capabilities, or slew of requested features? Kudos for the 3D, that's awsome without a doubt, but...

    but as the standard internet turns more towards applications, the differences between HTML and RIA will become even starker.

    Right, which is why I'm not really super pissed. I have the option of just saying, screw internet development, it's just not worth the headache. My app for Windows 7 phone is actually making money, which really wasn't my goal, but it's a nice side benefit. So I'll stick with that until the hype of HTML 5 subsides and people realize it's still a turd.

    Thursday, April 28, 2011 7:30 PM
  • I would rather stick with the current status of SL than switching to HTML5, 6, 7... HTML was never ment to be used for applications and that fact will never change. The T in HTML still stands for text, right!? So let me know when HAML gets released. Oh wait a second, we already got XAML.

    Thursday, April 28, 2011 8:03 PM
  • Sunday, May 01, 2011 1:07 PM
  • Your goal involves converting Silverlight/WPF to HTML. It won't work. How do you convert XNA 3D to HTML and expect to have any sort of decent performance? For the foreseeable future there are only about three or four major platforms, Windows and Mac OS for desktops, Silverlight is already on those (no one gives a crap about Linux desktops.) That leaves phones, iPhone, Windows phone, and Droid. Do a port of iPhone and Droid, really not a significant effort for a company of Microsoft's size and resources, and guess what, they're done. Poof. All of their developers are delighted. Why waste time even trying to develop an approach that turns Silverlight/WPF into crap?

    We're not asking for a port to 53434 different platforms. Heck, I could care less about iOS. Let me run on Android and they can even profit on it by having their own Android store, just like Amazon does. Why is it that Amazon can do a port to virtually every platform out there, but Microsoft can't?

    Sunday, May 01, 2011 5:40 PM
  • Do you mean Amazon or Google?

    Obviously some functionality gets lost in translation, but it is possible to translate a great deal of functionality out there to ensure most requirements are met.  Just look at GWT.

    There's a lot of smart peeps in Microsoft and doing a conversion/adaptation/translation like this is right up there with what they do best.

    I'm not sure why *you* as a developer even care about what can be translated or not.  This is up to the organizations (such as Mono) that have to do all the dirty work (so that you don't have to).  Never say never jack... you can do a lot w/ JScript. ;)

    Sure, you're not going to get 100% of the functionality.  But if you can get 80-90%?  That's enough to persue it IMO.

    Sunday, May 01, 2011 6:10 PM
  • Do you mean Amazon or Google?

    I meant Amazon. Kindle runs on everything, and now they have their own app store.

    Obviously some functionality gets lost in translation

    So why bother?

    is right up there with what they do best.

    No it's not. In the past they would identify weaknesses and address them, and if that meant proprietary extensions, so be it. If you chose to play in their world, that was the price you paid. And guess what, Microsoft developers didn't care, because they could deliver solid solutions faster, and didn't give a crap about standards. ONLY LOSERS CARE ABOUT STANDARDS.

    I'm not sure why *you* as a developer even care about what can be translated or not.

    You live in this magical world where apparently 1 to 1 translations between disparate systems are possible. And you say, it's up to such and such to do the dirty work. That's just not living in reality. Here is a sample. I want you TODAY to develop a system that in JavaPuke opens a local file, encrypts it, and sends the results to a web server. Oh, and do it on a background thread. It's IMPOSSIBLE. So how would anybody convert a Silverlight solution that does that to JavaPuke?

    Never say never jack

    I can say never with the current implementations because they simple don't provide the necessary API calls. Yes, those could be added, but that's a completely different story.

    you can do a lot w/ JScript

    The problem isn't all with the language. It's with the DOM and the browser APIs as well. Here's another example? Can JavaPuke access the operating system's native networking stack? Nope. So how would you translate a Silverlight app that does to JavaPuke? The devil is in the details, and you've consistently ignored that fact and said, "that'll be up to the smart guys at Microsoft to figure out." The whole reason for all the additions to HTML 5 is that you can't do those things in HTML 4, regardless of how smart you are. And guess what, HTML 5 with all its improvements is still a joke compared to WPF. The delta simply cannot be overcome by cleverness.

    Sunday, May 01, 2011 6:32 PM
  • I appreciate the can'titude, jack.

    You keep bringing up these wild edge cases (not to dismiss them, because yes, what you outline is a challenge, I agree).  Don't look now, but even to this day C# and VB.NET continue to diverge in key aspects of their featureset (and even now with the next version of .NET are they focused on converging all languages - that's versions later).  But well over 90% (maybe way more) of functionality exists between the two languages today.

    Of course you're going to have difficult aspects of COMPLETE adaptation.  In the use case you mention above, I actually think all of that would be possible, save read from the local hard drive.  It's a matter of research and implementation.

    All I'm saying is that Google is doing all of this already on a very basic level with GWT.  Developers today are developing their applications in their native languages and they are being translated down to platform equivalents.  All the "dirty work" is already being handled by "other smart people" who are much better at doing such a thing.  It is happening RIGHT NOW at a competing technology firm -- and if Google can do it, I bet Microsoft can do it better (that's totally IMO) -- otherwise, it might be time to start brushing up on your Java.

    Sunday, May 01, 2011 9:25 PM
  • But well over 90% (maybe way more) of functionality exists between the two languages today.

    They're both built on top of the MSIL and were never meant to be identical.

    I actually think all of that would be possible, save read from the local hard drive.

    And the OS network stack? You're waving your arms again and magically making problems disappear. And 3D performance worth a damn via a Canvas tag and JavaPuke??? I'm really beginning to wonder how much programming experience you have if you believe these things can be overcome by pure force of will.

    It's a matter of research and implementation.

    No, it's not. I don't know the full details of GWT, but it may be based on a significant subset of Java. You simply can't overcome lack of threading or synchronous networking by being clever.

    It is happening RIGHT NOW at a competing technology firm

    No it's not. Your repeating that over and over isn't going to change the fact you can't do the impossible. Here, a two second search resulted in this. http://www.manning-sandbox.com/thread.jspa?threadID=19404 They talk about how you can't use log4j because JavaPuke doesn't support local io.

    it might be time to start brushing up on your Java.

    What's there to catch up on? It hasn't added anything since its initial crap release.

    Sunday, May 01, 2011 10:54 PM
  • You're waving your arms again and magically making problems disappear.

    And you're waving your arms and making problems magically appear.  We're both just sitting here bickering and waving our arms if we're not doing what matters most: writing code.

    I'm really beginning to wonder how much programming experience you have if you believe these things can be overcome by pure force of will.

    :)

    Monday, May 02, 2011 11:55 AM
  • I personally am an 100% adopter of IE7,8,9 and could care less about open platform browsers.  MSFT should at least bring full blown WPF to the browser even if it is entirely only on the IE stack.  Want to win the browser wars, then provide WPF to IE only.  Open up everything for trusted root certificate holders and you're all set. 

    Bottom line is XAML everything it's touted to be or not?

    Monday, May 02, 2011 12:34 PM
  • I personally am an 100% adopter of IE7,8,9 and could care less about open platform browsers.

    OK, now here we have a MICROSOFT developer. His employer BUYS products like Visual Studio, his concerns and not Richard Stallman's might be of interest to Microsoft.

    MSFT should at least bring full blown WPF to the browser even if it is entirely only on the IE stack.

    Ah, old school profitable Microsoft approach, use the browser/OS for leverage.

    Want to win the browser wars, then provide WPF to IE only.

    So you are arguing that a browser that conforms to standards, like every other browser, provides no competitive advantage? What about all those poor Microsoft haters who won't be able to render pages optimized for IE in their browser?

    Bottom line is XAML everything it's touted to be or not?

    Probably overhyped. I mean, how could you NOT want to be designing pages that are a combination of HTML and SVG with no animation support. And all that databinding nonsense that actually delivers separation between data and presentation, that couldn't possibly have any advantage over CSS files randomly sprinkled throughout the universe.

    Monday, May 02, 2011 1:05 PM
  • And you're waving your arms and making problems magically appear. 

    Well that's an admirable attempt at jujitsu. I'm pointing out all the problems with your conversion approach, i.e. JavaPuke doesn't support the necessary APIs to make it possible, and it's ME who is making problems appear? If you're seriously looking for feedback, you shouldn't play that kind of game.

    Monday, May 02, 2011 1:46 PM
  • Point taken.  From my perspective, you're trying to play the "how much of this is really possible" game, which will take a lot of time/analysis/resources -- an effort meant for a team, not two opinionated developers dueling on a forum post.

    But sure, let's run some of its course then: As far as most business applications requirements go, how much of WPF can be ported to javascript?  I'm going to start at 100% and then you can chip away at it.  How much will we have left? 95%? 90%? 85%? 80%? Way less?

    I'm talking requirements for business applications.  Not video games (although a great deal of "force of will" can get this done as well, I suspect).

    This game is meant for a team of people with a great deal of resources to figure out.  We're both eyeballing in this case.

    But sure, go ahead.  Tell me what % of .NET can't be ported over to JavaScript and I will continue to engage you in this game.

    One thing is for sure, the % is good enough for Google, because they have already done it (sorry to force the issue, but it's true).

    My point is that this letter is to an organization who has the resources to figure out the PRECISE AND EXACT % of portability.  There has been no messaging or communication that they have looked into it or have even THOUGHT of it.

    The message has been: "Silverlight is now Windows only, now you have to deal with a hellish technology you dealt with over half a decade ago.  Good luck."

    Monday, May 02, 2011 3:54 PM
  • I'm going to start at 100% and then you can chip away at it.

    The point is, it isn't 100%. So why should Microsoft waste time on something that its customers, its loyal developers don't really like or want? MIX went over like a thud because Microsoft gave its customers something that they didn't want. I haven't heard or read a SINGLE Microsoft developer that likes the new strategy. Not a single one.

    Not video games (although a great deal of "force of will" can get this done as well, I suspect).

    Did you just graduate? Developers haven't even really tackled 3D in Silverlight let alone HTML because it needs dedicated runtime support. And besides, now you're hedging more by stating "business applications." Racing to the bottom, one of the key problems with HTML, and why real programmers avoid it whenever possible.

    Tell me what % of .NET can't be ported over to JavaScript and I will continue to engage you in this game.

    It's less than 100%, so there's no point in playing.

    There has been no messaging or communication that they have looked into it or have even THOUGHT of it.

    People have been throwing this idea around for a while now. Most of the time it just gets rejected. The dopes at google ran with it a bit because they have an irrational love of HTML. It'll be funny when they realize, "hey, we have a search engine, an os, maybe we should have our own language." They're probably already working on it.

    Monday, May 02, 2011 4:52 PM
  • irrational love of HTML

    More like irrational love of maximum market reach.

    Monday, May 02, 2011 5:03 PM
  • More like irrational love of maximum market reach.

    Well there's that, although they are getting to the point where they can start setting some de facto standards. Their rejection of H.264 in favor of WebM is a classic Microsoft style play. Don't like it? Tough. I don't know much technically about SPDY, but I applaud them for identifying a problem and trying to remedy it. Again, their willingness to reject standards is BEAUTIFUL. I'll take de facto standards over something designed by a committee any day.

    Monday, May 02, 2011 5:36 PM
  • Todays blog of Mary Jo clearifies this little war here. If we like it or not, but Silverlight will not become the big coss-platform WPF world problem solver. But hey, we can still hope that the ligth of the moon will fully shine everywhere.

    "I asked Microsoft officials what the changes mean for the future of Silverlight and was told by a spokesperson that it’s “status quo” with Silverlight, in spite of today’s org changes. Microsoft officials have been signaling since last fall that the company’s Silverlight strategy has shifted, and its emphasis, going forward, will be on Silverlight as a development tool, and HTML5 as Microsoft’s cross-platform app/content play."

    source: http://www.zdnet.com/blog/microsoft/microsoft-reorg-scott-guthrie-to-head-new-azure-application-platform-team/9324

     

     

    Monday, May 02, 2011 8:26 PM
  • Wow!  What a blow.  ScottGu will be missed -- greatly.

    will be on Silverlight as a development tool, and HTML5 as Microsoft’s cross-platform app/content play."

    What does that even mean???  How is Silverlight a development tool?  That doesn't make sense.  A tool in development?  When I think of "development tool" I think of something like Resharper, or even Visual Studio -- something that makes development easier and faster.

    In any case, Microsoft is re-orging, and there is a big announcement in September.  Let us hope for the best.

    Monday, May 02, 2011 8:47 PM
  • http://www.zdnet.com/blog/networking/chrome-11-the-best-browser/1006?tag=content;selector-blogs

    Just pathetic. Microsoft is getting their asses handed to them in the browser space, and they're still chasing standards. I wanted to move back to IE9, but it still loads tabs slowly, or just randomly hangs. And since it doesn't offer a single compelling feature that Chrome doesn't have, why use it? Can we look forward to the next version in two years? Now if IE9 had HTML+, i.e. Silverlight, I wouldn't be looking back. Instead, what can I do? Wait for it, I can change the colors of a path in some crap SVG. I'm just jumping up and down with joy, honestly, I can barely contain myself.

    Well that's it for me tonight, I'm going to go watch the MIX keynote again, puts me to sleep in no time.

    Tuesday, May 03, 2011 2:36 AM
  • Maybe for our SL 6 Wishlist it should just be that SL 6 will come about...  With Gu gone  (isn't that a solvent ;-) ) and SL de-emphasized and buzz from Microsoft about SL seeming to die down, I just hope there ARE future versions of SL.  But hopefully, that is just my eternally negative view on things...

    Tuesday, May 03, 2011 7:54 AM
  • Maybe for our SL 6 Wishlist it should just be that SL 6 will come about

    Silverlight as a web browser plugin is dead. The bottom line is, if Microsoft isn't actively promoting it, why should any IT manager stick their neck out and use it? Not only is Microsoft not promoting it, they take the opportunity to say don't use it whenever possible. That leaves Silverlight as a plugin relevant in a only couple use cases:

    DRM protected video - Relevant to Netflix and maybe a couple other major providers

    3D - Isn't even applicable until v5 ships

    Everything else - Use HTML 5. Their "shifted strategy" argument is that while it might be nicer for developers to target Silverlight, HTML 5 even in its current vaporware form, is the way to go.

    Bottom line. If you're using Silverlight as a browser plugin today, you should start rethinking your plans. Either go with HTML 5 or Flash. With Microsoft ACTIVELY downplaying Silverlight, its installed base is going to dwindle, making it significantly harder to argue that the plugin requirement isn't prohibitive.

    Tuesday, May 03, 2011 12:45 PM
  • hmmmm...  Looks like SL5 was the sunset release then doesn't it.  They've shifted focus and left us here hoping for more.  I guess it's time for me to bite the bullet and move on over to flash and javapuke.  At least there, I can get things done without waiting for SL to open up.  SL is dying for desktop delvelopment.  The interesting part is that MSFT has already moved on....we still can't believe it.

    Remember, according to ZDNET Article:

     

    I asked Bob Muglia, the Microsoft President in charge of the company’s server and tools business, that very question and got what I consider to be the clearest answer yet about how Microsoft is evolving its Silverlight strategy.

    Tuesday, May 03, 2011 6:42 PM
  • Silverlight is our development platform for Windows Phone

    No reason they couldn't do a few minor custom extensions to HTML for things like accelerometer, compass, and gps, and ditch Silverlight there as well. When those things get officially adopted by the HTML committees, they could add a quirks mode to the phone browser to support the legacy versions of those APIs. Now that I think about it, I'd be surprised if this isn't their current long term goal.

    I'm pissed.

    Yah, Silverlight losing out to a couple clowns showing SVG color manipulation during MIX made it crystal clear that they're sunsetting Silverlight. What a lesson in corporate stupidity.

    Tuesday, May 03, 2011 6:52 PM
  • EASY TO INSTALL PLUGIN!!!!!


    I loose so many visitors because of this.  

    I've got a super busy site with over 100,000 free xxx videos.

    Despite showing videos and images of the sites content and a slick custom plugin installation page people still will not take that extra step and install the thing.

    Furthermore, the "Get Silverlight" install and spec page is not localized.

    My sites live stats display like a stock ticker.  Every time I look at it I keep telling myself to switch to Flex.....

    Tuesday, May 03, 2011 8:32 PM
  • Hey Jack, if SL is for the phone does that mean WPF is for the IPad?

    Tuesday, May 03, 2011 10:49 PM
  • Multivalue convertor should be there as soon as possible. As I have shifted from WPF I like it very much.

    Wednesday, May 04, 2011 3:29 AM
  • For multi binding support you can use this library

    http://www.scottlogic.co.uk/blog/colin/2010/08/silverlight-multibinding-updated-adding-support-for-elementname-and-twoway-binding/

    It's pretty good and stable. 

    Wednesday, May 04, 2011 4:56 AM
  • I think multibinding should be supportive like it has in wpf.

    Wednesday, May 04, 2011 5:04 AM
  • I want MDI (Multi Document Interface) ability.

    Friday, May 06, 2011 2:44 AM
  • My primary wish for Silverlight 6?  Existence.  So far, I see no evidence of that ever coming true, though.  :(

    Monday, June 27, 2011 6:32 PM
  • Hey Jack, if SL is for the phone does that mean WPF is for the IPad?

    No, actually it looks like WPF is slated(pun intended) for MaxiPads(I'm just kidding).

    Tuesday, June 28, 2011 3:06 AM
  • Oh I see people here writing almost part of the huge 20 millions + views thread :)

    Tuesday, July 12, 2011 7:34 PM
  • My wish is to continue developing C#, Xaml, SL, for PC and for the next Tablet that will run it !
    I know it'll happen.

     

    Tuesday, July 12, 2011 7:35 PM
  • - Event tunneling

    - Selectable text mode in controls

    - Better install/uninstall experience for the average user (maybe a background desktop process that does the job)

    - Full Linux support (Sorry MS, Adobe didn't leave its platform to a third-party)

    - In-browser WebBrowser for browsers other than just IE

    - p2p support for chat and video calling

    - bundle Silverlight with Windows so it's easier to reach most computers

    - Silverlight on tablets

    Tuesday, July 12, 2011 11:06 PM
  • - Compile Silverlight to native HTML5, CSS3 and JavaScript for In-Browser-Apps. :-)

    That would be so cooool! :)

    Thursday, July 21, 2011 8:52 AM
  • Look at the growth of iOS and Android in Tablet market, just in one Q.


    http://www.sys-con.com/node/1916201

    but we cant loose all the power and beauty of LINQ.

    XAML with MVVM

    Ria Services with Lazy loading with commands from the Client Project.

    the power of Entity Framework

    People like us that make a living developing Business Applications are not happy with the uncertainty that Microsoft is causing.

    Ok, these guys were all hired as SL evangelists. Now look at their blogs and see if find any SL actions?

    http://jesseliberty.com/

    http://timheuer.com/blog/

    http://johnpapa.net/

    http://10rem.net/

    Thursday, July 21, 2011 11:36 AM
  • SILVERLIGHT 6 WISH: STAY ALIVE.

    Friday, July 22, 2011 9:08 PM
  • Just read some financials on web today that Apples cash reserves are larger that Intel and MSFT combined.  Looks like the elephant got caught sleeping doesn't it?

    Friday, July 22, 2011 11:24 PM
  • Add Strike Out (Sample Strike out) feature to all TextBox, TextBlock and RichTextBox controls which should have multiline support too.

    Monday, July 25, 2011 2:44 AM
  • Add the following properties to MultiScaleImage as Ajax Seadragon has them as well.

    • blendTime - The amount of time in seconds that new tiles take to blend from transparent to opaque. Default is 0.5.
    • alwaysBlend - Whether tiles should always blend in and out, not just when they're first loaded. Default is false.
    • immediateRender - Whether the most appropriate tiles should always be rendered first, before any lower-res tiles are rendered. This loses the "sharpening" effect and instead creates a very visible "tiling" effect. Default is false.
    • imageLoaderLimit - The maximum number of concurrent image downloads that can be performed by each viewer. Default is 2.

    Wednesday, July 27, 2011 8:43 AM
  • Just had to share this.  This captures everything with absolute perfection (wondering if jackbond made this ;)): http://www.youtube.com/watch?v=RRFiu0xfQzw

    Tuesday, August 09, 2011 9:45 PM
  • Michael;

    That is a GREAT post! Thanks for sharing that.

    Jp

    Wednesday, August 10, 2011 12:14 PM
  • Just yesterday it was reported that APPLE is now larger than Exxon!  Wow!

    Wednesday, August 10, 2011 12:16 PM
  • “Gentlemen, this is a football.”

    These five famous words by Vince Lombardi communicated that his team would focus on the basics,
    the fundamentals. This is what Silverlight needs to do.
    • Fix bugs!!! I come across a new bug most every day.  Silverlight, VS (especially XAML editing), and
      Expression Blend have enough bugs to fill an ant farm the size of Seattle. Bugs are the top time-waster.
    • Complete the help system and Intellisense, so we can actually get meaningful information every time.
      Also when working in Silverlight, F1 help should always be about Silverlight.
    • Fix the error handling and debugger so there are always meaningful messages, and all errors are
      trapped by Application_UnhandledException in App.xaml.
    Saturday, September 10, 2011 1:32 PM
  • Looks like I won't be getting my wish (that is, "existence").  IE 10 doesn't even support Silverlight anymore:  http://news.cnet.com/8301-30685_3-20105870-264/microsoft-joins-the-anti-flash-crowd-with-ie10/?tag=topStories

    Oh, but you can use it to build Windows apps!  What?!?!  Who cares???  Why on earth would you build Windows apps with Silverlight when you have the entire WPF library at your disposal?  Silverlight was supposed to be in the browser.  IN THE BROWSER!

    Oh, well, SL is dead.  On to HTML 5 and mobile apps, I guess...

     

    Thursday, September 15, 2011 1:46 PM
  •   IE 10 doesn't even support Silverlight anymore: 

    This is incorrect.  Silverlight is supported in the IE10 experience on the classic desktop of Windows 8.

    You would use Silverlight to build desktop applications that you or your client could deploy using the web.

    Thursday, September 15, 2011 4:40 PM
  • I've been absorbing the keynotes throughout the week, and to me, I thought the message was pretty clear: If you were a Silverlight developer, you're now a WinRT developer.  If you were a WPF developer, you're now a WinRT developer.

    *Everyone* is now a WinRT -- or Windows -- developer.

    Silverlight just got upgraded and consolidated into the same platform as WPF.  This is the right move.

    Consider: 400M+ copies of Windows7 have been sold since its 2009 release.  This is compared with 10-20 million non-Windows devices (iOS/Droid)that are out there (based on my very brief research).  Even if you double that, it ends up being 10% of the market reach of Windows.  This has been consistent since the beginning of time, even with the Netscape browser wars; it has always been Microsoft at ~90% vs. the world at ~10%.

    That means the "True" marketplace is on the way, and WinRT is a juiced-up Silverlight to make the applications for it.  No more hacky browser or "can do this in WPF but not in SL, vice versa" business.

    Sure, we will have to wait a while, but the long-term vision is what this is about.  Just plan your business accordingly. ;)

    I'm happy with this.  We still get to leverage all the goodness we've been learning from Xaml.NET and move forward into a brave new world.

    I could be off my rocker -- wouldn't be the first time.  But, really, I think this is good stuff.  Can't wait for the beta tools (no more "preview" spelunking for me ;)).

    Thursday, September 15, 2011 10:01 PM
  • I've only used SL (5 Beta) for 3 months so these might be a bit narrow to my short experience

    - We need a GridLengthAnimation

    - Observable ActualHeight and ActualWidth, in other words make it possible to bind to ActualHeight and ActualWidth

    - Make grids autogenerate rows/columns based on the attributes on the children

    - Make a native WebBrowser control that doesn't want to take over the world, instead of the current COM one

    - Fix the binding order on ComboBox, so it would bind the ItemsSource first then the selectedItem, instead of the other way around (I think this was fixed in 5RC)

    These are just the beginning.

    Friday, September 16, 2011 12:20 AM
  • Fix the binding order on ComboBox, so it would bind the ItemsSource first then the selectedItem, instead of the other way around

    I actually think you can control the order yourself simply by changing the order in XAML, i.e.

    <ComboBox ItemSource="{Binding Items}" SelectedItem="{Binding SelItem}" />

    is not the same as...

    <ComboBox SelectedItem="{Binding SelItem}"  ItemSource="{Binding Items}" />

    Friday, September 16, 2011 1:00 AM
  • nk u can control the order yourself simply by changing the order in XAML, i.e.

    <ComboBox ItemSource="{Binding Items}" SelectedItem="{Binding SelItem}" />

    is not the same as...

    <ComboBox SelectedItem="{Binding SelItem}"  ItemSource="{Binding Items}" />

    Actually that isn't true, I've tried both with breakpoints at xaml bindings and selecteditem is consistently called first. Keep in mind this was a combobox inside an itemtemplate in an items control, not just a standalone combobox.

    Friday, September 16, 2011 1:44 AM
  • Observable ActualHeight and ActualWidth, in other words make it possible to bind to ActualHeight and ActualWidth

    <my:SomeControl Width="{Binding ActualWidth, ElementName=SomeContainer}" />

    ?

    Make grids autogenerate rows/columns based on the attributes on the children

    http://whydoidoit.com/2010/10/06/automatic-grid-layout-for-silverlight/
    http://blogs.planetsoftware.com.au/paul/archive/2010/04/30/autogrid-ndash-part-1.aspx
    http://www.scottlogic.co.uk/blog/colin/2010/12/a-simplified-grid-markup-for-silverlight-and-wpf/

    Fix the binding order on ComboBox, so it would bind the ItemsSource first then the selectedItem, instead of the other way around (I think this was fixed in 5RC)

    The problem is actually more that mostly the ItemsSource will be loaded async. It's best to set the SelectedItem in the ViewModel after you know that the bound collection has loaded.

    Friday, September 16, 2011 5:03 AM
  • Ok so do I understand the PDC summary correctly:

    The current split was in Desktop vs. Internet development (WPF and Silverlight).  But now all of Silverlight migrates to WPF, and HTML5 becomes the new split?  Does HTML5 then become only Internet develolpment or does it now cross-over into desktop.

    Does HTML5 bring all this dumb (in my opinion) Security and half-baked baggage that Silverlight had? (Even with Trusted root certificates?)

    Does anyone know where a PDC 2011 Summary site is?

    Friday, September 16, 2011 9:27 AM
  • I've been watching the keynotes from here: http://buildwindows.com

    HTML5 is a presentation component of WinRT, just like Xaml.  The boys at Microsoft "got cute" IMO and made all of their presentations (including the app store in Windows8) in HTML/JS -- however it's a great evangelist strategy to bring the newbs into the fold. 

    It's also worth noting that this is not the ghetto JS of jQuery, and that's the key.

    WinRT is the new WPF/Silverlight, and supports C#/Xaml *AND* HTML/JS -- you pick your preferred language.  The base libraries are available in C# and JavaScript, if that explains it better.

    In any case, I for one am no longer thinking in websites, but marketplaces.  400+ million clients to support your application on any device is around 10 times bigger than any marketplace available today (eat it, Apple).  That's the vision two years out from next April (when Windows8 is rumored to RTM).

    You build your application once and it will be available to 90% of devices in the marketplace. Last I checked in school, 90% was considered an A. 

    That's the message I got (or at least I believe they should be preaching :)).

    Friday, September 16, 2011 9:54 AM
  • WinRT is the new WPF/Silverlight, and supports C#/Xaml *AND* HTML/JS -- you pick your preferred language.  The base libraries are available in C# and JavaScript, if that explains it better.

    There's just one problem: WPF / SL run on XP / Vista / 7. WinRT doesn't.

    For that reason alone, it can't replace those two until Win 8 and up are as widespread as XP / Vista / 7 are now, so basically for the next 5-10 years or so.

    In any case, I for one am no longer thinking in websites, but marketplaces.  400+ million clients to support your application on any device is around 10 times bigger than any marketplace available today (eat it, Apple). 

    That's really nice  - if you're a consumer app developer. For enterprise apps, this won't be the case for a long time if ever.

    That's the vision two years out from next April (when Windows8 is rumored to RTM).

    Where does this rumor come from? Judging from the demo version, I'd be surprised if Win 8 will be out before October 2012 and I wouldn't be shocked if it slipped all the way to March or even June 2013.

    Friday, September 16, 2011 10:11 AM
  • <my:SomeControl Width="{Binding ActualWidth, ElementName=SomeContainer}" />

    That only works once. I want the binding to update when the ActualHeight and ActualWidth change. ActualHeight and ActualWidth properties do not fire a propertychanged event. Right now the workaround is to put a sizechanged event in the codebehind and update the bindings myself, not cool.

    The examples you gave for the AutoGrid are the workaround that people are having to write. It would be nice if it was in the vanilla Grid control. This is a wishlist afterall.

    The problem is actually more that mostly the ItemsSource will be loaded async. It's best to set the SelectedItem in the ViewModel after you know that the bound collection has loaded.

    Not quite. Both values are coming back in the same async call from the model. I went in as much detail as firing the SetropertyChanged on the ItemsSource as early as possible and the SetPropertyChanged on the SelectedItem as late as possible. Didn't help, Dispatcher or no Dispatcher. This is a bug and it has been fixed in RC, so I'm scratching that off my list :)

     

     

     

    Friday, September 16, 2011 2:41 PM
  • That's really nice  - if you're a consumer app developer. For enterprise apps, this won't be the case for a long time if ever.

    I think if you are doing enterprise apps you can safely ignore Metro for quite a while.  Even the early adopters of Windows 8 will be able to run your enterprise apps in desktop mode, unless they are running on an ARM processor.  And what enterprise shop is going to buy a bunch of ARM processor tablets that won't run ANY of their existing stuff? 

    I think WinRT is a pure consumer play for the time being. 

    Of course, good luck getting anyone to let you use Silverlight moving forward.  With Microsoft basically messaging that they hate Silverlight, why would any company choose it as a technology going forward, even if it is the best choice?

    Friday, September 16, 2011 3:32 PM
  • I think if you are doing enterprise apps you can safely ignore Metro for quite a while. 

    Yes.

    Even the early adopters of Windows 8 will be able to run your enterprise apps in desktop mode, unless they are running on an ARM processor.

    But for Win8 users, SL-Apps will only run in the "ghetto"-mode (the "other" IE, not the shiny metro IE with the plugin-free browsing experience). And for ARM, they just won't run at all. Yay! We'll be able to build second-class solutions.

    I think WinRT is a pure consumer play for the time being. 

    That' what I think, too.

    Of course, good luck getting anyone to let you use Silverlight moving forward.  With Microsoft basically messaging that they hate Silverlight, why would any company choose it as a technology going forward, even if it is the best choice?

    And that's exactly the problem.

    SL has been killed, everything that's not WinRT has been marked as legacy & deprecated. And WinRT is, exactly as you said, purely a consumer play atm. For an OS that probably won't even be done before a 12-15 months from now.
    Which means: the old stuff has essentially been killed, the new stuff is not really usable yet (and  won't be in the enterprise for at least the next half decades), and good luck with your projects in the next few years.

    Friday, September 16, 2011 5:13 PM
  • SL has been killed, everything that's not WinRT has been marked as legacy & deprecated. And WinRT is, exactly as you said, purely a consumer play atm. For an OS that probably won't even be done before a 12-15 months from now.
    Which means: the old stuff has essentially been killed, the new stuff is not really usable yet (and  won't be in the enterprise for at least the next half decades), and good luck with your projects in the next few years.

    This is my problem exactly. Starting a new enterprise web-platform product (6 months development at least + 5-10 years maintenance), and silverlight was the best match thanks to its virtualized nature + solid communication foundation. Microsoft is deprecating that with no clear alternative. 

    Is Microsoft's platform cadence simply too fast for enterprise these days?

    Friday, September 16, 2011 5:23 PM
  • Matt's right in that we're left in limbo when it comes to deciding on a platform right now.  Here's an interesting decision tree for you, though: 

    http://www.telerikwatch.com/2011/09/how-to-pick-your-platform-silverlight.html

    Problem with this is, I would ask the 2nd question first, that is, "Does it need to work on all platforms"?  So far, for every single client I have, for the last 6-12 months the answer has been a resounding "yes"!  I so wanted Silverlight to be the answer to that question, but now that's totally gone by the wayside.  So we're stuck going back in time to the HTML days of yore.  Sure, it's got a "5" after it now instead of a "4", but it's still stinkin' HTML with all of it's cross-browser quirks and all!

    So down the drain go the hundreds of thousands of dollars of Silverlight research and investment that my company has put in in the last 3-4 years.  Now we're painfully transforming into an HTML 5 / mobile developer.  Which means at least we can use Silverlight again at some point whenever the Windows 8 tablets come out, but in the mean time, we're investing in a lot of new JavaScript books.

    Oh, wait, what's this about Google developing Dart as a replacement to JavaScript???  Son of a b...

     

    Friday, September 16, 2011 5:33 PM
  • Matt's right in that we're left in limbo when it comes to deciding on a platform right now.  Here's an interesting decision tree for you, though: 

    http://www.telerikwatch.com/2011/09/how-to-pick-your-platform-silverlight.html

    I love Telerik. But the diagram is a tad optimistic for my taste (9 months? Really?).

    And this kind of decision help is something that should really have come from Microsoft. 

    Friday, September 16, 2011 5:56 PM
  • And this kind of decision help is something that should really have come from Microsoft. 

    You got that right, Matt.  And Telerik is low-rent.  ComponentOne FTW. ;)

    Also, thank you for your great point about the Legacy/Vista/XP crowd.  Yesterday morning before you wrote that, I was all-in on ClickOnce/WPF as the technology of choice for LoB.  But after poking around -- most notably discovering that RIA Services will not be available for WPF -- it's obvious that Silverlight5 is the choice for these scenarios.

    So, after much thought and consideration, this is how the technologies break down in my world:

    • WinRT: Consumer
    • Siverlight: LoB
    • ASP.NET: Marketing/Mobile/LowBand version of WinRT/LoB -- essentially "everything else" (or the "10%")

    (Much respect and recognition goes to ColinBlair for his post on the RIA Services Blog.  He nailed it.)

    If I were to assemble a LoB development team, I would have as the core:

    • 1 Silverlight5 Developer
    • 1 ASP.NET Developer (if mobile use is a requirement).

    If I were to assemble a consumer application development team, I would have as the core:

    • 1 WinRT Developer (for the "90%")
    • 1 ASP.NET Developer (for the "10%")

    Finally, as far as Win8 RTM Rumors: http://www.zdnet.com/blog/microsoft/not-so-crazy-microsoft-rumors-windows-8-to-rtm-in-april-2012/9823

    Saturday, September 17, 2011 11:37 AM
  • Finally, as far as Win8 RTM Rumors: http://www.zdnet.com/blog/microsoft/not-so-crazy-microsoft-rumors-windows-8-to-rtm-in-april-2012/9823

    Yeah, but according to that rumor (which is from June), we should have gotten the one and only beta release "around BUILD". Which hasn't happened. And looking at the Developer Preview, I certainly don't think they'll have a RC by January.

    Also, didn't Angiulo basically say they're still not decided about which parts to port to ARM? 

    Saturday, September 17, 2011 2:14 PM
  • Which hasn't happened. And looking at the Developer Preview, I certainly don't think they'll have a RC by January.

    Yes, you are right.  Even during demos there were all sorts of buggy things happening.  I guess that's what you get when you develop with HTML5/JS. ;)

    Moral of the story: plan carefully.  Looks like there's lots of time ahead of us (which only lengthens the value of SL5).

    Saturday, September 17, 2011 3:18 PM
  • - Compile Silverlight to native HTML5, CSS3 and JavaScript for In-Browser-Apps. :-)

    I think this is very likely, they did it with ASP, they did it with ASP.net.  Both of those technologies had a long history of success.

    Saturday, September 17, 2011 9:53 PM
  • I think this is very likely, they did it with ASP, they did it with ASP.net.  Both of those technologies had a long history of success.

    Those are very bad examples.

    Both ASP and ASP.NET were built from the ground up to map to Html/JS. 

    Any additional client side code has to be JS. And the server side VB/C# is compiled and executed as normal .Net MSIL. There is NO cross compiling involved.

    Also, there's just one way these environments work: the way they work when they're outputting Html / JS. A Silverlight cross compiler would have to make sure that the behavior of the output in a very different environment (Html/JS) would be exactly the same as in Silverlight. And it would have to work that way in all kinds of under all kinds of different OSs. Because if only small things behave differently, your app might start doing funky things. And you'd have no way to control this behavior from your Silverlight code (because you have no way to influence the Html/JS output directly). That's never ever going to work.

    It's probably a better bet to wait for the Miguel de Icaza to provide a fully SL5 compatible Moonlight than to wait for the magical SL 5 --> Html 5 cross compiler.

    Saturday, September 17, 2011 11:48 PM
  • I think most of us agree we just don't like Javascript, but perhaps we better just bite the bullet and start using it.  Why?  Because even in my line of business (traditionally only ASP.NET and Winforms...slowly moving to WPF and maybe SL (more likely Flash)) I see directives like this "We have to be able to serve phones).  So if that's the case why target a framework? 

    Remember Java's mantra write once run anywhere?  It worked very well for Java and it is still strong today.  There's entire shops that are dedicated to Java only.  So in my mind if we adopt this strategy which is pick the most universal framework, then we don't have to target anything. 

    I think any experienced programmer knows one thing very clearly...  If you have multiple solutions on mulitple platforms you are headed for lots and lots of work.  In fact, I've seen entire multi million dollar projects (like a baby) thrown out with the bathwater because upper managment just doesn't understand the power of refactoring and coding to interfaces.  MSFT has done this to us consistently e.g. You cannot embed Word or Excel (even legal copies) into your own application, rather you have to use interop and or tricks to make (as in the case of ASP.NET) the application think it's Word or Excel.  If you honestly analyze those interfaces you find out that 1) It doesn't work just like the real thing and 2) You have to manually add function that would otherwise be handled in the application normally.

    Look at this glaring gap... The RichTextBox both in Asp.net, SL and even in WPF.  There's no native support (really) for dropping an image from the clipboard.  Yes you can do it but you have to wrap it in a blockui element when in fact the Rich Text protocol supports it natively. 

    I think it's time for us to give up on MSFT's gap and do what we all hate to do, embrace HTML5 and JAVASCRIPT all because MSFT interest in protecting themselves transcends the ability for us programmers to work around the landmines.  Now that they are opening up the OS to HTML5, it's an opportunity for us to jump.  I would have NEVER picked HTML5 and Javascript to be the center of the universe but am left with little options.  (WPF lover that I am)...

    So perhaps we need to ask MSFT simple things like... Can you please allow us to set breakpoints for Javascript in the IDE debugger instead of having to jump to run time debugger via F12 in the browser?  Arghhhhh!   We love VS2010 and have to use the browser to debug javascript the now universal client side language....What a mess.

    Monday, September 19, 2011 9:21 AM
  • BTW How can MSFT drop RIA that's ridiculous.  In fact I was waiting for WPF to adopt it.

    Monday, September 19, 2011 9:24 AM
  • Totally agreed, v2kea60.  There have been so many things we've invested in in the last 3 years -- Silverlight, RIA, Entity Framework, WPF -- that we're just dropping altogether now.  I don't want to.  I love that stuff and it's been very good to us.  But reality has set in and the future is barrelling towards us with mobile as the center of everyone's world.

    Every client we have, and I mean EVERY client, from the 10-man boutique shops to the $4 billion corporations (our clients run the gamut), has said that they want to support mobile and/or tablet in everything they are doing and will be doing.  That means mobile Safari, the Android web browser, or native apps on iPhone/iPad/Android(/even WP7).  So we just can't do Silverlight anymore.  When Microsoft pretty much said Silverlight is no longer a browser-based technology (or not going to be for very long), that did it.  If there's no hope of ever getting Silverlight in the mobile space, then we have to be done with it or we'll die along with it (the web version of SL, that is -- I know, it will live on as WinRT).

    So, I agree with v2kea60 in that we need to have better tooling for Javascript and HTML 5.  That's where we have to be and I, for one, just cleaned out the local Barnes & Noble of every Javascript and HTML 5 book I could find that looked relevant and was published in the last 6-12 months.  (Literally last week -- this Build conference has completely solidified our company direction for me!)

    Hey, I've missed Javascript -- we haven't done it seriously as a company for many years -- but 1 year ago, I really thought we may never have to use it again, if only SL would show up on mobile browsers.  But we gotta move on, and as a Microsoft partner, this seems to be the way they're pushing us, oddly.  Code on, brave adventurers!

    Monday, September 19, 2011 9:38 AM
  • I'm doing with the SL4 COM support now.  Basically I'm using the ADO COM library and connecting to SQL CE files.  However with a different connection string it work to SQL Server as well.  Now ADO COM is certainly not ADO.net, so its a little rough.  It does work for my purposes very well. 

    Wednesday, November 09, 2011 9:53 AM
  • Looks like I may have been right about SL6's non-existence:

    http://www.zdnet.com/blog/microsoft/will-there-be-a-silverlight-6-and-does-it-matter/11180

    And SL5 may not even be supported on Firefox and Chrome, so SL4 may be the last cross-browser version!  Wow, back to HTML, I guess.

     

    Wednesday, November 09, 2011 9:57 AM
  • FWIW, I just tested and SL5 works with FireFox and Chrome.

    Thursday, December 22, 2011 8:23 AM
  • Most need the support of WinRT and 3D with XNA (ARGB all) - format compatibility.

    P.S. from Russia with love Wink

    Saturday, December 24, 2011 12:24 PM
  • I voted 1 vote but huh, at least it's one more :)

    However my wish would just to continue developing SLX and having the customer's apps running on a Tablet and phone as well,  I don't care not having an Android or iPhone for them,  I just want to have our customers the possibility to run their apps on their Windows Phone, their Windows Tablet and everhwere if possible.

    In the browser of the phone or the browser of the tablet at least....

    Happy new year

     

    Wednesday, December 28, 2011 1:08 PM
  • JavaScript + WCF Web Services + asynchronous, stateless AJAX calls&#160;= cross-platform joy. If so, what left with C#, xaml, wcf, blend and vs with the ease of use and the speed and beauty of the apps you can do ? Will someone at Microsoft understand that the web world needs Silverlight ? No HTML, no java, no ajax, all that crap is not for me.
    Wednesday, December 28, 2011 5:50 PM
  • IS there a Silverlight 6 planned?

    What is the FUTURE of Silverlight and related technologies?

    Thursday, December 29, 2011 2:51 AM
  • Go to http://www.microsoft.com/presspass/events/build/ you&amp;#39;ll get a few answers there. Sorry, unable to make a link out of it with the iPad on this forum nor any RETURN works. And I just saw that the apostrophe changed for the html character once I posted the message.
    Thursday, December 29, 2011 5:09 AM
  • Have you investigated Script#?

    See the video here - Script#: Compiling C# to JavaScript using Visual Studio

     

    The video doesn't play.  The control loads and I'm using SL 5 runtime but the video just don't play.
    Oh, strangely it works with Chrome ?????

    Thursday, December 29, 2011 5:35 AM
  • Doesn't play right for me either.  Although I was able to hit the "format" button and choose another one that mostly worked.  Looked like a SL version, but not sure.  And when it wasn't MAXIMIZED, the mouse position wasn't properly interpretted so I had to move the mouse an inch or two above where it was supposed to be in order to click on things.

    Nice. Not!

    Thursday, December 29, 2011 8:20 AM
  • adding complete library of WPF such as Interop and enabling screen capture may be a good idea for silverlight 6.0. We currently use FJCore for that, but this feature doesn't completely satisfy us.

    Friday, January 06, 2012 5:29 PM
  • Window is one of best features in sl5, but something shoould be added to that. for example currently we can't hide control box of a new window, or making the window fixed in size or position is not possible. So, adding some features for new Window object can be a good idea. Thank you Microsoft for Silverlight...

    Friday, January 06, 2012 5:34 PM
  • ...

    or making the window fixed in size or position is not possible.

    ...

    What do you mean?  You can size and position a ChildWindow.

    Saturday, January 07, 2012 7:47 AM
    1. Fix bugs.
    2. Get rid of all the different prefixes, like "sdk:" and "tk:"
    Wednesday, February 01, 2012 10:40 AM
  • I speak, about capabilities of a new window (not a child window) and need to make its position fixed, so that user can't resize or move it.

    Sunday, February 05, 2012 9:32 AM
  • Whish that silverlight 6 does come out.

    Sunday, February 12, 2012 10:49 PM
  • Yes, I just come to Silverlight, but all say it will die soon.

    As a Enterprise System Developer.

    I had been developing in java for several years, then asp.net.

    finally I came to silverlight, and I think it's very very good, we don't care if the client has to install a plugin, 

    I and my team all think silverlight is very good for our enterprise system developing, it solves many B/S based system's problems, it's great!!!, it's truly amazing.

    But we all are sad about microsoft's indication.

    We hope there will be silverlight 6 or if Microsoft doen't want silverlight any more, can Microsoft give it to some open source community like Moonlight or Apache?

    We want to use silverlight, we can't use HTML5 in enterprise for a very long time.

    silverlight is a great enterprise solution, please don't give it up, and make sure windows phone and tablet (just windows)

    can use it, so we can purchase windows phone and tablet just for that reason(support our silverlight system).

    Tuesday, February 14, 2012 7:30 AM
  • Silverlight will be supported (probably in its current form) until 2021.

    I also think it is THE best platform currrently available and that HTML5 is a long way out AT BEST.

    Although Silverlight might not continue to get improvements (and only needs a few), the new Windows 8 with WinRT and xaml support SHOULD mean that the knowledge you gain in Silverlight will be directly usable in Win8.  Your xaml and MVVM patterns should be able to move easily to Win8.  That SHOULD allow running on Win8 PCs, tables, and phones.

    But we give up cross platform.

    You should also be able to run Silverlight in the "legacy desktop" mode of Win8 as is.

    So continue on your Silverlight path is my recommendation.

    Tuesday, February 14, 2012 8:30 AM
  • If HTML5 is all that why are the SQL Azure tools done using Silverlight?

    Wednesday, February 22, 2012 11:45 AM
  • Make it

    Thursday, February 23, 2012 3:31 PM
  • If HTML5 is all that why are the SQL Azure tools done using Silverlight?

    Perhaps, like us, they started on Silverlight before the HTML5 edict came out and they haven't re-done them yet.

    Or perhaps, like us, they might like SL so much that they aren't going to re-do them.  But somehow I doubt that since their target audience is not necesarily a Windows client.

    Saturday, February 25, 2012 9:15 AM
  • Just a little update on this thread, since it has been the semi-official thread of the post-Mix'11 heartbreak meltdown (not to mention one of my favorite threads of all time :)).

    I have been exploring Silverlight 5 and it is really quite beautiful.  I've got a lot of magic happening now and a lot less hacking to get things working the way I expect.  The markupextensions are especially a game-changer, IMO.  In fact, from a technology solutions perspective, choices are easier now as I think there really isn't a reason anymore to use WPF.  If you're doing desktop/browser-based applications, Silverlight5 all the way baby!

    Anyways, with the advent of Windows Runtime, I will be porting over to that as soon as I can (read: as soon as ReSharper gets out a EAP for VS11, LOL!).

    So, all is not lost for Xaml development.  Xaml-driven development is still the superior way of developing applications and I look forward to porting to and embracing Windows Runtime.  In the meantime, you can see what I've put together for Silverlight 5 here:

    http://framework.dragonspark.us/

    Lots of cool application stuff in there.  Stuff I have been trying to get implemented for years now, starting way back in ASP.NET.  Oh my humble roots...  Anyways, onward towards Windows Runtime!

    Saturday, March 03, 2012 3:15 PM
  • Wake up people, Silverlight's gone bye bye.

    It looks like there ain't gonna be no Silvelight 6.

    Friday, April 20, 2012 10:29 AM
  • I'm over it.  Moving on to mobile development for "rich" applications and HTML 5 for web.  Sad to see SL go, but I guess it was inevitable.  Excited to build some Windows 8 Metro apps and sell them on the Windows Marketplace, though!  Onward and upward!

    Friday, April 20, 2012 12:59 PM
  • * Understand that not all programs for all purposes work well to run as fullscreen application, some users requires access to 4-6 applications and windows at the same time and release a SL update.

    *  Update BCL (and other core libraries) in order to match .NET 4.5 (incl async, MEF2, WeakEvent<T> etc)

    Friday, April 20, 2012 2:09 PM
  • Double bufferring feature with best performance for 2d game programming

    Sunday, May 13, 2012 1:03 AM
  • Hope there will be a mvvm project template in new visual studio.

    Tuesday, May 15, 2012 5:15 PM
  • It is pretty easy to make one in VS.  I think there is Save as template or some such.  I have made a Silverlight MVVM template for Embarcadero Prism.

    But you are right.  It would be nice to have more out of the box support for MVVM.

    Or course, it would be even nicer if we ever get ANY support for silverlight after 5 release.

    Tuesday, May 15, 2012 5:36 PM
  • That there will be a Silverlight 6?

    1.  Microsoft to accept that not all applications are baby applications that must live in a Sandbox ... that some of us develop real applications to do real work and don't want or care about a Sandbox -- yes, we assume the responsibility cause we're grown ups.

    2.  Sooo...how about providing a "conversion/migration" process to go from SL5 to HTML5 including all the code behind?  Needs to really work, not the usual 70% and we're done approach that Microsoft embrace.  Hit a migration button and that's it, no fuse, no R&D, no endless google searches for answers to problems, no filing connect bug reports ... yes, that means Microsoft would have to actually test what they release -- I know, a dirty word, "TEST" ... oooh, that's reserved for us developers right?

    3.  Direct SQL access would be nice so we can do away with having to use VERY SLOW and data size restricted Web Services that does nothing more than increases ROI (debugging time and deployment time) for no other reason that someone's belief that "only secure applications are coded this way" -- how many times do we have to prove that Web Services are far easy to hack than a well secured SQL server??

    4.  Stop making assumptions about how Applications should work and how developers should code ... to even suggest that Microsoft know ALL the possible real world applications requirements is foolish at best and almost insulting.  Stop attempting to control everything and provide what we really need, flexibility!!

    Rob

    Tuesday, July 03, 2012 1:38 PM
  • That there will be a Silverlight 6?

    1.  Microsoft to accept that not all applications are baby applications that must live in a Sandbox ... that some of us develop real applications to do real work and don't want or care about a Sandbox -- yes, we assume the responsibility cause we're grown ups.

    2.  Sooo...how about providing a "conversion/migration" process to go from SL5 to HTML5 including all the code behind?  Needs to really work, not the usual 70% and we're done approach that Microsoft embrace.  Hit a migration button and that's it, no fuse, no R&D, no endless google searches for answers to problems, no filing connect bug reports ... yes, that means Microsoft would have to actually test what they release -- I know, a dirty word, "TEST" ... oooh, that's reserved for us developers right?

    3.  Direct SQL access would be nice so we can do away with having to use VERY SLOW and data size restricted Web Services that does nothing more than increases ROI (debugging time and deployment time) for no other reason that someone's belief that "only secure applications are coded this way" -- how many times do we have to prove that Web Services are far easy to hack than a well secured SQL server??

    4.  Stop making assumptions about how Applications should work and how developers should code ... to even suggest that Microsoft know ALL the possible real world applications requirements is foolish at best and almost insulting.  Stop attempting to control everything and provide what we really need, flexibility!!

    Rob

     

    Then make everything yourself and do not rely on Microsoft. See Linux for example.

    If you're not interesting by what Microsoft gives to their developers then I strongly suggest you to see how Open source developer works and work with them.

    Good luck.

    Friday, July 06, 2012 7:57 PM
  • Hi Guys

    I can't understand what really microsoft is going to do, what will happent to us as developers and the companies as investors.

    The silverligh looked like a magic tools for prgramming in web , by programming tools I mean a real tools for developing LOB software data centric full of business logic needs spead and usabilit like insurance software, not just a toy to bulid a graphical site.

    Unfortuantely HTML 5 really destroeyd my dream, again javascript ...

    It seems Microsoft at least until now has not got any public solution for programmin in web and other smart tools and platform(like android) so It seems we have to use, yes, a toy dynamic language for making an enterprise software,

    I am not sure It is possible but It seems we again have to migrate from all good things of .Net C#(strong types, good IDE, really good debuging tools, compiletime errors, managable Dlls,  Intellisense, Linq,WPF,Windows Forms,   annoymous, .....) and do object oriented programming in javascript. In my opnion It would be like a real nightmare 

    Can u imagin 5000 page of javascript source code, who can manage It, how can find bugs in it? how can develop tools in it? how can learn bussines developer to programm in it? I really do not know:-)
    1- I think at least Microsoft should introduce a tools like GWT for converting c# to html javascript 

    2- Improve .net IDE to fully support javascript I mean things like compiling javascript...

    3- Maybe continue Silverlight in other way

    But who knows about future in IT espicsially in web :-)

    Friday, August 17, 2012 12:18 PM
  • Here's another one who doesn't understand.

    1) Continue what you are doing.  Why would you stop now ?
    2) Windows 8 UI is in evolution.  Right now it targets multiple device, tablet, phone, pc and who knows what.

    You need to make a choice for which audience you have to do an application.

    WEB = Silverlight, HTML/JAVA, AJAX, ASP.NET and whatever you want.  This is far from dead
    Windows 8 UI = HTML/JAVA or XAML/C#, VB (Store to make $ for you whatever kind of app you do)
    LOB = Whatever you want (WinForms, SL, HTML, ACCESS Cry and who knows what else)
    LOB + WEB + SL KIND OF APP + ALL DEVICES = NON EXISTANT YET

    So in a sense just make sure you use what you need for the audience that will use your app.
    don't try to cry, it's no use.  Make sure you do the best with what you have.

    Friday, August 17, 2012 12:40 PM
  • Here's another one who doesn't understand.

    1) Continue what you are doing.  Why would you stop now ?
    2) Windows 8 UI is in evolution.  Right now it targets multiple device, tablet, phone, pc and who knows what.

    You need to make a choice for which audience you have to do an application.

    WEB = Silverlight, HTML/JAVA, AJAX, ASP.NET and whatever you want.  This is far from dead
    Windows 8 UI = HTML/JAVA or XAML/C#, VB (Store to make $ for you whatever kind of app you do)
    LOB = Whatever you want (WinForms, SL, HTML, ACCESS Cry and who knows what else)
    LOB + WEB + SL KIND OF APP + ALL DEVICES = NON EXISTANT YET

    So in a sense just make sure you use what you need for the audience that will use your app.
    don't try to cry, it's no use.  Make sure you do the best with what you have.

    Thank you,

    yes man I know and I  exactily  mean LOB + WEB + SL KIND OF APP + ALL DEVICES = NON EXISTANT YET ,

    but there is one tools just HTML5/Javascript but is like a joke 
    so we have to just watch will happen 


    Saturday, August 18, 2012 2:36 PM
  • I'm full of hope too but not a dreamer.

     

    Saturday, August 18, 2012 5:12 PM
  • I don't know why I have the feeling that all this Silverlight stuff is a ploy to get developers onto Windows 8, which just so happens to be XAML based; I hope I’m wrong.  All I can say is if Silverlight becomes a "Microsoft-Device-OS-Only Cross-Device Platform" [say that 5 times fast], then it will confirm my suspicion. I can only hope that Moonlight will start supporting multiple OS’s and devices also in time, and perhaps that may be a saving grace (and perhaps Microsoft should help it do so – it would only work in their favor), but I wouldn’t hold my breath.  Honestly, the whole thought of having to resort back to HTML/JavaScript makes me feel a bit feral.  I have yet to see companies cooperate with standards; even in healthcare HL7 interface communications (I’ve done a lot of work in software development for Healthcare).  The only true way to build something to run across multiple systems, devices, etc., and have it look/run the same everywhere is NOT by standards, but to either 1. Have a single company create and maintain the development platform (tainted sometimes by money and politics), or 2. Built it open source by the community (which will probably lack somewhat in quality and coherency (due to volunteers), and good development tools).

    Of course, the other “SMART” thing MS could do (and may be doing [see: http://bit.ly/NLo0jm]) is to continue to have SL as a means to run software cross-platform, while also supporting XAML/C# apps it in all their own platforms.  This might attract many developers to the Microsoft platforms, knowing that they can also easily have versions to run on other non-MS platforms as well (which I assumed SL was going all along a few years back).

    In conclusion, if SL doesn’t work out, then I’m creating my own, and HTML/JS can [insert vulgar statement here] … anyone else on board!? LOL ;)

    [edit] I suppose also, in time, if SL5 gets deployed with Windows 8 - along with the increase in XAML/C# apps, and the market place - and since SL5 I'm sure will be around for quite some time, there's a good chance there will be a future time when SL5 (at the very least) will still be a great platform to use, since it may be on the majority of computers around the world (http://bit.ly/8l4Y9Q). Honestly, I'm still optimistic in any case. ;)

    Saturday, August 18, 2012 11:35 PM
  • As jamesnw said,

     

    It really seems as a ploy to get developers on de WinRT train. I have already tried WinRT, is fast, but not that fast! It also seems to have about the same memory footprint as Silverlight. WinRT seems a lot like Silverlight 3 on Steroids, Silverlight 5 is far more evolved.

     

    My wish at this time now is that microsoft would make Silverlight Open Souce.

     

    And as jamesnw said also, I'm getting on his boat, gladly would gladly join any silverlight like team, because HTML 5 + JS can take a Hike!!!

    Sunday, August 19, 2012 1:04 PM
  • I think I hit a nerve, my post was deleted. ;)

    Anyhow, it's here also now: http://bit.ly/Na5mG2

    [edit] Ok, they restored it. LOL ;)

    Sunday, August 19, 2012 1:17 PM
  • Indeed, it seems it was deleted. I couldn't find it to quote it.

    Sunday, August 19, 2012 3:33 PM
  • I agree that SL5 is a great platform for LOB apps.  I thought moonlight was basically dead.  The site hasn't been updated since 2009 (am I wrong?)  The big advantage to SL is that it is a much more mature platform and it works on older versions of windows, XP, Vista, 7, 8.  However, WinRT apps only work on Win8.  So building a LOB for WinRT is a bit silly unless you can get your customer to upgrade everything to Win8.  Feature wise SL5 is very complete, not sure what else could be added.  I've been complaining to the SSRS team for years to create a silverlight report viewer.  They don't seem to care about silverlight users.  They just point to the ASP.net control.  It works but it doesn't integrate naturally and feels like a huge hack.

    Monday, August 20, 2012 9:25 AM
  • Oh, didn't realize that Moonlight was dead, thanks.  Just confirmed it in this post I found: http://bit.ly/Azb27I (also talks about Silverlight, and why the Moonlight momentum "lost air")

    Monday, August 20, 2012 9:48 AM
  • Feature wise SL5 is very complete, not sure what else could be added.

    For example, Silverlight Rich Text supports only basic formatting. And it is almost impossible to create fast custom implementation of formatted text because you can't draw text directly on the screen. The only way to implement this is by placing many TextBlock objects on canvas but it's too slow for most real text-processing scenarios.

    I've been complaining to the SSRS team for years to create a silverlight report viewer.

    By the way, we developed SSRS viewer for silverlight if you still need it.

    Monday, August 20, 2012 11:01 AM
  • I second that. I had to take the RichTextBox control and basically wrap it and build my own editor (which still doesn't work as well as I'd like - I still may have to build my own from scratch). Undecided  As well, I found out the resulting markup is extremely big (very "bloated") for the simplest of formatting.  Would be nice of it supported basic HTML formatting instead, rather than a quasi-half attempt at mimicking it.  For that matter, even a "proper" RichTextBox" that can ACTUALLY LOAD A .RTF FILE would help be keep things like TOS, Disclaimers, etc., in separate files, then load them for display.  It was a pain in the a-- to transfer formatted stuff from Word (or even HTML, which is a shame) into anything Silverlight based.  People already know HTML, why force XAML formatting for RTB!?  It would only take me one day to implement [very] basic HTML formatting in a custom editor control. Yell (and yes, of course, there are 3rd-part controls that can "pick up the slack")

    For me, Silverlight is bitter sweet: I totally love it over all - but sometimes it feels unnatural, and I proceed putting forehead to desk in repetitive motion. Laughing

    Monday, August 20, 2012 1:39 PM
  • I wouldn't say Silverlight is complete at all.

    There are many WPF features missing in SL, here are some:

    • Rich text controls
    • Documents
    • Shared size group
    • Reporting controls
    • Several controls and layout techniques of WPF that doesn't exist in SL
    • Some types of Bindings
    • More

    So saying it's complete is really bullshit.

    Besides I think whatever we talk here is bullshit, cuz Microsoft really has decided to f••• SL off. There will never be another SL. If MSFT would have been serious about SL there woulda been at least an Android and iOS version.

    The ultimate solution I would wanna see for SL is a SL+C# to HTML+JS compiler, meaning I write whatever I write in SL, and instead of a plugin at the client, HTML5+CSS3+JS is emitted! cuz really with hard work, I believe this can be done with HTML5.

    Lemme know what you think, even we all know this is dreams, SL is over. MSFT has already decided to replace C# with efin JS.

    Monday, August 20, 2012 5:51 PM
  • I guess if complete means 100% WPF compatibility then you are right it is not complete.  To me, it just means all the pieces are there to build line of business apps.  The only thing on your list that is a pain for me is the missing reporting controls.  I blame the SQL Server team for that one, not the Silverlight team.  There are 3rd party controls but ugh.. I don't want to deal with that, not to mention the price. 

    I like the idea of a magic HTML5+Javascript converter but I'm not sure how this compiler would work.  Wouldn't it require to bring over everything in the Silverlight .NET libraries to javascript libraries somehow?  Also, it would be very strange to have a silverlight app running on iOS or android using microsoft silverlight UI widgets instead of the native OS widgets.  I've done a fair amount of development with Xamarin's monotouch and monodroid.  You can reuse a lot of the business logic and data access but the way the UI technologies and application life cycles work make it impossible to share that code.

    Here is what I think would be cool.  I want MS to give us compatiblity libraries for metro style apps in SL5.  I realize certain parts won't port (like the store).  I wan't to be able to have a single code base that will look/work sort of like those fancy new Win8 apps but in silverlight 5.  That way we can run them on older versions of windows.  Eventually we could abandon the SL versions once everyone is on Win8 or newer (that's like 5+ years away!)  Hmmm.  I wonder if I could build part of this without MS...

    Tuesday, August 21, 2012 1:52 PM
  • Ok.  Javascript.  Ugh.

    Wednesday, August 22, 2012 9:45 AM
  • Each time I see the word JavaScript, I go to the toillette.  Sorry I'm back,

    However when I see the word Silverlight, which by the way not only the word but everything I do and everywhere I go I need it and I see it and I'm using it, I just feel happy.

    It's crazy how this thing changed the way I do apps every day for a living.  How easy and fascinating it is to work with the tools and see the result that should have been the platform that entirely replace HTML/JavaScript ..... arrggg sorry I have to go to the toilette...

     

    Wednesday, August 22, 2012 9:53 AM
  • He, he!  I know, I know.  It's more geared towards mobile development, so what you can definitely take to the bank is that it is better than Objective-C and/or Java, but I know that's not saying much!

    That being said, I am struggling a lot with decisions on how to build new Enterprise-level applications!  Clients still want web-based, but they want it to work on iPads, Android tablets, phones, etc.  So Silverlight is out.  But HTML + JavaScript + CSS (what we dub here as "the mess") is a complete nightmare!  C# + XAML is spectacular!

    Then there's the possibility of native Windows apps, but then do we choose WPF or WinRT / Windows 8 Store-type apps?  Since Windows 8 isn't even out yet, we have to go with WPF for now, but that's not exactly "future proof" and you're leaving out Macs and every mobile device in existence!  So we're left with Titanium, which currently builds for iOS, Android and Web and promises to add Windows Phone 8 and possibly Windows 8 native output as well.  Yeah, it's JavaScript, but it's either that or the JS+HTML+CSS MESS!!!

    We've gone from a clear, direct, cross-platform utopia of Silverlight+Everywhere back to the stone ages of cross-platform night terrors!  No idea which way to go on any given project, so we are leaning towards Appcelerator for lack of better options.

    Wednesday, August 22, 2012 9:54 AM
  • We are going back to the caves and broke! I totally agree with AvatarJohn. I've checked Titanium, really not interested in getting into the JS world, but ponting this out is mostly appreciated.

    Any one has any other ideas on how to build apps without "the mess", or the consumer oriented and platform locked WinRT?

    Best Regards.

    Wednesday, August 22, 2012 10:22 AM
  • right...but i don't think there will be silverlight 6...i wish it will.

    Plz remove permission for 3D programming in silverlight with xna...or make it user friendly...like a pop up, or dialog box.

    Wednesday, August 22, 2012 10:54 AM
  • I can't believe this thread hasn't been locked already. ;)

    We've gone from a clear, direct, cross-platform utopia of Silverlight+Everywhere back to the stone ages of cross-platform night terrors!  No idea which way to go on any given project, so we are leaning towards Appcelerator for lack of better options.

    FWIW, I've been checking out Visual Studio LightSwitch.  It supports both Silverlight and HTML5 ("soon" at least).  That way, you can dev Silverlight clients for your administration/desktop/Windows environments, and HTML5 clients for everything else.  I've been checking out the videos and material behind LightSwitch and have been impressed with their philosophy.

    It's mostly meant for LoB applications, not gaming/3D.  But for me, that's OK.  And since it takes care of all the dirty work of application plumbing, I am spending more time on my application.  Which is the reason why I created a framework in the first place.

    Anyways, LightSwitch seems promising, and it keeps Silverlight alive while embracing the filth of HTML5 (while hiding all of its guts from you as a developer). 

    Win. :)

    Wednesday, August 22, 2012 6:20 PM
  • Huh.  Now there's a great suggestion!  I haven't touched Lightswitch in months and just about forgot about it.  I see it's built into Visual Studio 2012, guess I'll have to check it out!  Thanks!

    Wednesday, August 22, 2012 6:24 PM
  • Then make everything yourself and do not rely on Microsoft. See Linux for example.

    If you're not interesting by what Microsoft gives to their developers then I strongly suggest you to see how Open source developer works and work with them.

    Good luck.

    So because we don't have tools we need, you're suggesting I write them myself via Open Source ... are you encouraging developers to move away from Microsoft then?  Sorry, but I thought this thread was SL6 Wishlist ... as in people make suggestions on how to improve SL -- since future development of SL technology IS DEAD, the thread got extended to Getting Microsoft to actually produce something developers want rather than something Microsoft can use to leverage "their" desktop and Mobile OS grand plan.

    What about developers that want central deployment, browser independence, OS independence, direct SQL access, aren't funded by advertising companies and want to live OUTSIDE of the Sandbox??  As has been pointed out -- this doesn't exist.  It doesn't exist because Microsoft don't want it to exist, they had the time, money, and opportunity make this "dream" a reality ... it's sad that you think this is or should be a dream objective -- it should have been a reality 5-7 years ago.

    How long as Microsoft been creating web technologies now?  You really think they couldn't have made that dream a reality by now if they wanted to?

    Why would I move to LightSwitch or for that matter any other MS technology that does more of the same and never really progresses?  SL was finally showing some signs of intelligence at Redmond, only to be junked ... SSDD.

    It boggles my mind to see so many developers happily move on to the next technology that doesn't really solve anything.

    Thursday, August 30, 2012 9:52 PM
  • ... are you encouraging developers to move away from Microsoft then?

    Absolutely not.  What I'm encouraging is to be happy with what we have right now which is working very good and enhence our skill without having Microsoft making it for us.  The key here is that we all want Silverlight to live, YES we want Silverlight 6, 7, 8, 9, whatever.  However What is more important ? What we're waiting for from Microsoft ? Or what we can do today with what we know and have ?

    Silverlight is far from dead.  Now let's get back to what this thread is all about.  Wishlist for next version.

    Here one I ask for today : I wish SL 6 could run on Windows Phone 8 browser.

    Friday, August 31, 2012 6:30 AM
  • Silverlight is far from dead. 

    Have you seen the video (http://www.neowin.net/news/former-microsoft-pm-silverlight-is-dead) post from the ex-PM (Project Manager) of Silverlight?  He (Scott Barnes) quit Microsoft because in his own words "Silverlight is f_cking dead" - the SL team was disbanded and re-assign to other teams.  Now I suppose it's possible Microsoft could get a new PM and put together a new team but there is absolutely NO indication that is happening.

    If you think Scott Barnes was just some disgruntled MS employee, think again ... his background is impressive and he had successfully predicted many of the "challenges" (kinder word than "failures")  Microsoft put themselves into.

    SL has another 10 years of support then it's gone -- of course it doesn't run on Win 8 Metro and only runs on Win 8 when in "desktop" mode.  I've already noticed that SL activity had died down to a trickle ... I guess everyone's busy writing HTML 5 books for VS 2012. ;)

    Friday, August 31, 2012 11:55 AM
  • You're probably right but I'm too. Silverlight isn't dead because one man left. Also I'll be doing SL even when this thread will be completely gone and you'll probably give up writing here before I stop making SL apps.
    Friday, August 31, 2012 12:11 PM
  • You're probably right but I'm too. Silverlight isn't dead because one man left. Also I'll be doing SL even when this thread will be completely gone and you'll probably give up writing here before I stop making SL apps.

    NO, SL is surely dead.

    Many ppl left including me and the entire company I work for, as soon as we realized SL has no future.

    Abandoning SL and investing all the effort in efin HTML+JS is a horrible mistake by MS.

    Cuz I don't give a shit about Win8, which is for consumer apps only.

    Silverlight was a perfect solution for LoB apps, and it IS dead. There won't be a SL6, wake up ppl, whatever MS tell you SL isn't dead, it is dying out slowly. Isn't dead means there is constant development and support for it. This technology, although being mature enough, still lacks some key features as mentioned above, and as soon as MS announsed to ppl to invest in HTML+JS instead of SL+C#, this means "we &ucked you, you all got busted".

    So, I'm now getting into more permanent long-living (BTW - open-sourced + multi-plat as well) Unix-based technologies, I still think C#/VB.NET are the best languages in the world, but you will admit MSFT is ^ucking its developers again and again, and my @$$ hurts already, so I'm out of this business. bye.

    Saturday, September 01, 2012 1:31 PM
  • As you wish :D
    Saturday, September 01, 2012 2:09 PM
  • Wish: Open Source Silverlight.

    Tuesday, September 25, 2012 12:14 AM
  • When doing LOB for corporates I use Silverlight and will for a long time
    When doing Web applications for the entire world I use ASP.NET, AJAX

    Good luck to all for the future of Web development...

    HTML X and JAVA X = NO GO !

     

    Tuesday, September 25, 2012 9:12 AM
  • Now with Sinofsky's departure (horray), I have to wonder if we're heading back to better days again. :)

    I've heard of a Silverlight6 rumor, but haven't seen anything yet officially materialize.

    I've been continuing to investigate LightSwitch, but I'm starting to feel that simply ditching it and going into Xaml-based Windows Runtime development is the way to go now.

    That is, Xaml-based Windows Runtime development is the "new" Silverlight. :)

    With all the cool form-factors now with Windows 8 Phone and Surface, you can develop your Windows Runtime application for these target forms, and then have a Droid/iOS developer/team port it over to the "other" platforms.

    Monday, November 26, 2012 2:00 PM
  • I also think it's cool (and telling) that C# Windows Runtime development is leading over HTML5 development 3-to-1 over in the Windows Store Application forums:

    http://social.msdn.microsoft.com/Forums/en-US/category/windowsapps

    Monday, November 26, 2012 2:08 PM
  • From what I know of Sinofsky, he came in and gave us Windows 7 (out of the depth's of Vista hell) ... the most successful OS since WinXP ... and Win7 has finally surpassed WinXP (only just).  Julie Larson-Green was responsible for the UI of Windows 8 and "program management".  Given that Windows 8 is a horrible failure (worse than Vista) and Surface is "Disappointing" (in Ballmer's own words) was it Sinofsky conflict with Julie or the overall direction Windows 8 went or was it Sinofsky that crushed all other Microsoft efforts (including silverlight)?  And if Sinofsky was so disruptive, why did Ballmer give him so much free reign?  I'm getting a little over Ballmer just sitting back and do nothing, not make final decisions ... either manage Microsoft or just get out.

    Most of the people I've talked to (technical and non-technical) that got exposed to Windows 8 don't like it (and it wasn't just a matter of getting "used to it") ... they call it "confusing".  Confusing is MOST definitely a UI problem, and that problem falls squarely on Julie Larson-Green.

    I don't know if Sinofsky leaving as a good thing, it might just be the opposite?  Was he the last person standing ground while Ballmer praised Julie with "Her unique product and innovation perspective and proven ability to effectively collaborate and drive a cross company agenda will serve us well as she takes on this new leadership role,"

    Sinofsky leaves Microsoft

    and another article on the same story.

    Another article on Sinofsky leaves Microsoft

    and another one

    Sinofsky leaves


    Monday, November 26, 2012 9:49 PM
  • Does that mean I have hope that Silverlight will make a come back ?
    So we can have Silverlight working on the tablet in IE Modern UI as well as making Windows Store app in Silverlight ?

    My God, asking this is asking someone to give me 7 million dollars !

    Monday, November 26, 2012 10:16 PM
  • Most of the people I've talked to (technical and non-technical) that got exposed to Windows 8 don't like it (and it wasn't just a matter of getting "used to it") ... they call it "confusing".  Confusing is MOST definitely a UI problem, and that problem falls squarely on Julie Larson-Green.

    Consider me one of them.

    I do like the new Windows Explorer 8 features and the new ribbon is really cool, also I like the colors of the windows, it just feels more modern.

    But the new start menu??? this is an awful invention when talking about desktop. It's an unbelievable space and time consumer. I've been using it for 2 months and NO I don't get used to it. It could be good for Surface/tablet/phone, but if you have a mouse and keyboard, it's just an unnecessary space consumer.

    So Win8 is a greater mistake than Vista.

    Now back to SL, I loved SL very much and I think if they want to bring it back to business they have to implement a Silverlight client engine that is pure JS and completely plug-in free, meaning that whatever I develop in Silverlight, it gets emitted to the users browser as HTML5+CSS3+JS.
    Wouldn't that be cool (c'mon, JS is still a f%%ed up language, C# is the best lang in the world, and RIA is awesome)
    Silverlight was the best choice for LOB, but im not gonna develop new projects on a technology that is announced dead.

    Monday, November 26, 2012 11:00 PM

  • Silverlight was the best choice for LOB, but im not gonna develop new projects on a technology that is announced dead.


    That's why I'm going to dive into Xaml-based Windows 8 development.  It's the closest thing to Silverlight.  The drawback obviously is that there isn't the marketplace presence of Windows 7 (500M+ installs of Windows7 vs 10-20M??? Windows8).  However, that is just a problem of time.  No matter what, Windows Runtime will be around for a while (longer than Silverlight???), so it's a better long-term move... in theory. :)
    Tuesday, November 27, 2012 1:06 PM
  • I think silverlight is still the king for LOB, but all the programmers and team leaders in enterprise actually knows that silverlight is a dead tech base on msft's behavior (such as sl not supported on winrt, ie10 can't refresh sl page, blend 5 never release, never talk about sl in dev conference, removal of sl site, removal of sl toolkit samples online, support flash not sl in metro IE.....).

    My team leader tell me not to use silverlight for new projects, use mvc instead, even it's much harder than silverlight.

    I personnally think he might be right, cause I don't want to use a dead tech either, even it's truly great....

    I think windows 8 xaml is only for consunmer apps......I can't imagine that kind of apps can be used in enterprise.....

    So Sum up, I think the current situation for sl in enterprise is:

    many programmers like sl very much and sl is great for enterprise apps.

    but they don't want to use dying tech (not supported by winrt nor surface pad...).

    and windows 8 xaml cannot be used in enterprise..

    So we had to choose html5 + tons of js libs +  chrome for new LOB projects.....OMG....

    Tuesday, November 27, 2012 2:58 PM
  • I prefer saying Silverlight isn't dead as it works perfectly well on Desktop PC.  The biggest whish everybody should have is having Silverlight 6 back on track and integrated into all MS platforms WP8, WRT, W8P Flawlessly that is working in all browsers of those platforms that means not only desktop browser but modern UI too. 

    I don't think one second that having this would make everybody not buying apps on the Windows Store because we would all use apps on the browsers.  Completely false but one thing is sure Silverlight apps are way more beautiful than Windows Store app if not written in DirectX like a game.

    This my boys would bring all corporates into the MS ship !  Lots of sells for the devices.  Many people happy, Santa Claus would be really real !

    Wednesday, November 28, 2012 11:33 AM
  • I think silverlight is still the king for LOB, but all the programmers and team leaders in enterprise actually knows that silverlight is a dead tech base on msft's behavior (such as sl not supported on winrt, ie10 can't refresh sl page, blend 5 never release, never talk about sl in dev conference, removal of sl site, removal of sl toolkit samples online, support flash not sl in metro IE.....).

    My team leader tell me not to use silverlight for new projects, use mvc instead, even it's much harder than silverlight.

    I personnally think he might be right, cause I don't want to use a dead tech either, even it's truly great....

    I think windows 8 xaml is only for consunmer apps......I can't imagine that kind of apps can be used in enterprise.....

    So Sum up, I think the current situation for sl in enterprise is:

    many programmers like sl very much and sl is great for enterprise apps.

    but they don't want to use dying tech (not supported by winrt nor surface pad...).

    and windows 8 xaml cannot be used in enterprise..

    So we had to choose html5 + tons of js libs +  chrome for new LOB projects.....OMG....

    I am in the position of setting our companies development efforts and we have absolutely NO projects aimed at Win8.  For one reason, it's still not "unified", I'd have to produce ARM and Intel specific builds -- this is good how??  But there again, why would anyone want to develop for Win8/Surface when it's clearly not selling well and has a worse adoption rate than Vista? (even Ballmer admitted this).  You can blindly keep faith/hope that eventually Win8/Surface will catch-on but don't be disappointed if you find yourself sitting in an empty room looking for work.

    For my business projects (current and future), if Microsoft continue on their current course they will become less and less relevant.

    In this day and age, Microsoft are NOT in a position of "telling me what I need" (I have too many options to NOT need anything Microsoft), all they succeed in doing is reducing the developer pool as we move on ... less developers = less Window Store applications = less users/consumers = less revenue.

    If you look at this technically SL is an Add-On to a browser, there is no real reason why the SL "add-on" can't be incorporated into IE10 browser code so it's no longer really an "add-on" ... IE10 has to know how to work with HTML5 soooo what's the difference?  There is none, HTML5 is part of the IE10 code, SL is not.  

    The smart choice would be to simply integrate SL into IE10 along with HTML5 support.  The ONLY reason this did NOT happen is because Microsoft want to be like Apple ... control over Applications via a Windows Store - read more revenue.

    I've never experienced a company (Microsoft) as being so confused/lost, so out of touch with reality, and on a path to irrelevance.  And the driving force behind all this isFear of Apple.  Good decisions are never made in fear.  It's like watching the Titanic where Microsoft is the Titanic.

    Wednesday, November 28, 2012 9:41 PM
  • Wow... lots of really great discussion here.

    I personally know of 3 companies that are getting a LOT of Windows 8 work underway.  Coupled with the fact that my application idea is consumer-oriented (and that I still think Windows Runtime is a viable longterm play in Microsoftland), it makes the most sense for me to dive into Windows Runtime development.

    Oh ya, and then there's this: http://www.nbcnews.com/technology/technolog/40-million-windows-8-licenses-sold-first-month-executive-1C7287682


    Wednesday, November 28, 2012 9:47 PM
  • The smart choice would be to simply integrate SL into IE10 along with HTML5 support.  The ONLY reason this did NOT happen is because Microsoft want to be like Apple ... control over Applications via a Windows Store - read more revenue.

    They really should integrate sl into IE 10.

    C# is keep goind down in TIOBE language ratings. 

    I don't know if the number was correct, but I think a lot of .net/sl programmers were leaving msft developer Camp.

    Thursday, November 29, 2012 2:27 AM
  • The smart choice would be to simply integrate SL into IE10 along with HTML5 support.  The ONLY reason this did NOT happen is because Microsoft want to be like Apple ... control over Applications via a Windows Store - read more revenue.

    They really should integrate sl into IE 10.

    C# is keep goind down in TIOBE language ratings. 

    I don't know if the number was correct, but I think a lot of .net/sl programmers were leaving msft developer Camp.

    I think the smart choice would be develop a HTML5+CSS3+JS client that will replace SL client.

    I believe most (and maybe even all) SL can be replaced with JS calls and CSS transitions. So it would be amazing to have LOB apps made in C#, RIA and all the goodies, and the compile engine will emit its JS crap to the client, aint that gonna be awesome!?

    I think that would be the best LOB solution for fluid and responsive UI, actually not only for LOB, as soon as it's gonna work this way it will be completely platform independent.

    Please yall vote for this great feature:

    http://dotnet.uservoice.com/forums/4325-silverlight-feature-suggestions/suggestions/1713161-compile-silverlight-to-html5-css3-and-javascript-

    Thursday, November 29, 2012 2:41 AM
  • Wow... lots of really great discussion here.

    I personally know of 3 companies that are getting a LOT of Windows 8 work underway.  Coupled with the fact that my application idea is consumer-oriented (and that I still think Windows Runtime is a viable longterm play in Microsoftland), it makes the most sense for me to dive into Windows Runtime development.

    Oh ya, and then there's this: http://www.nbcnews.com/technology/technolog/40-million-windows-8-licenses-sold-first-month-executive-1C7287682


    Licenses (which includes OEM pre-sales) is the same smoke and mirrors numbers Microsoft used with Vista sales.  You need to look at usage numbers for a more accurate reality of your potential Windows 8/surface market:

    Bad news for Microsoft

    40 million license not the real picture

    Must read if you are thinking of Windows 8 development

    I (my company) just can NOT take the risk, doing any Windows 8 or Surface development is VERY VERY high risk.  Even if your application is consumer-oriented, you would be wise to look at alternative platforms (Android, iOS) and at least stay diversified.

    Friday, November 30, 2012 5:16 PM
  • The smart choice would be to simply integrate SL into IE10 along with HTML5 support.  The ONLY reason this did NOT happen is because Microsoft want to be like Apple ... control over Applications via a Windows Store - read more revenue.

    They really should integrate sl into IE 10.

    C# is keep goind down in TIOBE language ratings. 

    I don't know if the number was correct, but I think a lot of .net/sl programmers were leaving msft developer Camp.

    I think the smart choice would be develop a HTML5+CSS3+JS client that will replace SL client.

    I believe most (and maybe even all) SL can be replaced with JS calls and CSS transitions. So it would be amazing to have LOB apps made in C#, RIA and all the goodies, and the compile engine will emit its JS crap to the client, aint that gonna be awesome!?

    I think that would be the best LOB solution for fluid and responsive UI, actually not only for LOB, as soon as it's gonna work this way it will be completely platform independent.

    Please yall vote for this great feature:

    http://dotnet.uservoice.com/forums/4325-silverlight-feature-suggestions/suggestions/1713161-compile-silverlight-to-html5-css3-and-javascript-

    I like the idea of "translators" but they rarely work well and often only work with the most simple of simple applications.  And then how would a translator deal with third party products like Telerik RadControls or the use of "interactivity" and "interactions" DLLs (which many use) ... just not viable for any moderately complex project.

    JS is a nightmare to manage/maintain and deal with compared to SL.

    The sad reality is that cross-platform single application is never going to be.  Microsoft's opportunity to make it happen was back when IE was 92% market share ... but that's what happens when a large company failures to understand it's future and how to manage that future in the present.   Ballmer has consistently told the world that the iPhone will never catch on, it's too expensive, yada, yada, yada -- it's that same ignorance that got Microsoft in their current mess of a situation.  And this isn't my "hindsight", I've been beating this same drum since 2000 and the problem was clear to me back then. 

    • .NET framework - should never have been an optional install (MS finally understand that too late)
    • click once applications - stalled (there is so much more than could have happened here that didn't)
    • technology of the day - at one point it looked like MS was racing against itself producing one technology after another with ZERO cross division support and I guess they expected developers to just somehow drop their current projects and switch mid-stream to these new technologies??  Who the heck wants to deal with 50 different project types just to build a single application?  

    and the list goes on and on of good ideas with really bad implementations.

    The ONLY salvation out of this mess is that the more web services and logic that resides on the server, the better.  This isn't optimal at all, because direct SQL access is A LOT faster than Web Services but because of mis-guided security fears public SQL servers are now taboo even though a public web server has a proven record of being less secure ... but hey, with web service security you can just blame the developer right??

    It's a sad situation, but we move on, find yet another syntax to learn and implement and just keep bumping from different technology to different technology wasting a huge amount of time "re-learning" a different way of doing what we already did before.  Software engineering has become lost and inefficient.

    Rob

    Friday, November 30, 2012 5:57 PM
  • Well well... Someone has been busy:

    http://xamarin.com/

    Monday, March 18, 2013 8:05 PM
  • I'm working on something right now similar to that:

    http://igg.me/at/DreamSpace/x/764997


    *** NodeJS .NET/Mono Alternative! http://igg.me/at/DreamSpace/x/764997 ***

    Tuesday, March 26, 2013 8:38 PM
  • Hmmm... I think you're mistaken, James.  Your concept isn't anything like Xamarin.  Xamarin doesn't do anything with JavaScript and compiles assemblies to native OS's.  JavaScript offers a sub-par and incongruent developer experience in Visual Studio (and will continue to do for quite some time).  This is why Xaml-based developers stay away from it.

    I do wish you the best with your framework though. :)  That's awesome that you're trying to put something together.

    Tuesday, March 26, 2013 10:27 PM
  • There's a UserVoice vote created for Silverlight 6.  Please take some time and dedicate some love to this item:

    http://visualstudio.uservoice.com/forums/121579-visual-studio/suggestions/3556619-silverlight-6

    Sunday, April 07, 2013 4:43 PM
  • Not sure how Linux users stand with viewing Silverlight now that further development of the Moonligt plugin has been discontinued from mono. I'm pretty much a begginer wth silverlight but I would like all desktop platforms to be able to view my silverlight apps.

    Kind regards

    Will.


    Will Robinson

    Wednesday, April 24, 2013 8:51 PM
  • Not sure how Linux users stand with viewing Silverlight now that further development of the Moonligt plugin has been discontinued from mono. I'm pretty much a begginer wth silverlight but I would like all desktop platforms to be able to view my silverlight apps.

    Kind regards

    Will.


    Will Robinson

    Stay away from Silverlight, it's a dead product.

    Shimmy

    Wednesday, April 24, 2013 9:53 PM
  • Not sure how Linux users stand with viewing Silverlight now that further development of the Moonligt plugin has been discontinued from mono. I'm pretty much a begginer wth silverlight but I would like all desktop platforms to be able to view my silverlight apps.

    Kind regards

    Will.


    Will Robinson

    Stay away from Silverlight, it's a dead product.

    Shimmy

    Don't listen to him.  He doesn't know what he's talking about.  Silverlight is great at 5 of course it could get even better and I'm sure a new version will come one day.  Right now, focus on what you can do with version 5 because it's very alive !


    • Edited by GearWorld Wednesday, April 24, 2013 10:51 PM
    Wednesday, April 24, 2013 10:50 PM
  • Not sure how Linux users stand with viewing Silverlight now that further development of the Moonligt plugin has been discontinued from mono. I'm pretty much a begginer wth silverlight but I would like all desktop platforms to be able to view my silverlight apps.

    Kind regards

    Will.


    Will Robinson

    Stay away from Silverlight, it's a dead product.

    Shimmy

    Don't listen to him.  He doesn't know what he's talking about.  Silverlight is great at 5 of course it could get even better and I'm sure a new version will come one day.  Right now, focus on what you can do with version 5 because it's very alive !


    My team has developed Silverlight applications and liked it very much (I still think Silverlight is amazing, and is hands down the best technology for LOB applications), but let's face it, the plug-in world is dead. Silverlight is doomed sooner or later it will disappear.

    As I said above, I wish they would have made some sort of Silverlight client side library that translates the XAML into HTML5+CSS3 and interactive JavaScript with jQuery.

    Any way this is not going to happen as it's not on the priorities of Microsoft, so for a new user who wants to learn a new technology, Silverlight is definitely not the right choice. You'd better off learn HTML+Javascript (god knows how much I hate JS, but MSFT didn't leave us any choices) and develop those apps using ASP.NET MVC razor.

    So talking about love, I think that the combination of WCF RIA that Silverlight has is awesome and beats any other technology around. But it IS doomed. And that's because of MSFT. I surely hate them for that.


    Shimmy

    Wednesday, April 24, 2013 11:24 PM
  • Thanks for the  reply's, from a novices point of view: I just wanted to build an application that I could attach to a web site that any one regardless of desktop operating system could view. My problem came about when I joined a particular forum in which I posted a link to the website in which my Silverlight application was attached. I was hit with an onslaught of disgust from Linux users because I had written my app in Silverlight and they could not view it? Why didn't I use flash, Mono, java or flex were just some of the questions being thrown at me. Apparently the Moonlight plugin was always one or two versions behind the current version of Silverlight so no one with Linux could view up to date Silverlight applications anyway. I wanted to ask them if they could view my app with the last plugin made available by moonlight but thought better of it so I've bought a Linux based laptop from eBay to satisfy my own curiosity. I'm hoping my silverlight  app runs on this as it's just a calculation based application only, then at least I'll be able to put a link to the moonlight plugin download alongside the application on the website. I did actually look into ASP.NET right from the start (About a year ago) and joined their forum to ask the question that I have a VB desktop application that I wanted to turn into a web application and would this be a difficult task for me as a complete novice? I was advised to go down the Silverlight route. Is ASP.NET the way forward?

    Kind Regards

    Will.

     

    Will Robinson


    • Edited by willrob68 Thursday, April 25, 2013 7:33 AM Grammer
    Thursday, April 25, 2013 2:59 AM
  • Here's the complete compatibility list.

    BTW, regarding Silverlight's deadness, an Android/iOS version is not planned, I don't remember the source, but I read it in an official blogpost/email.

    Definitely IMHO I prefer ASP.NET even more than PHP.

    ASP.NET MVC Razor is a great engine that no other technology beats. Really.

    Silverlight is an awesome technology if it's going to be internally used by your company, but not as a wide-targeted user solution. I'll repeat again, if you're new to Silverlight, don't waste time on it. MSFT announced it's death at previous build event by not talking about it at all and talking about HTML5-CSS3 and £ucking javascript instead.

    I wish they wouldn't do it. I hate javascript. It's an error-prone hardcoded language. Replacing C# with JS is like replacing blue label with urine. But that's what it is. MSFT gave it up.


    Shimmy

    Thursday, April 25, 2013 3:23 AM
  • Thanks for the Advice Shimmy. I won't give up totally on silverlight but I've learnt my lesson about limiting myself to one technology and I'm definitely going to go back and have a go at ASP.NET MVC Razor if this will solve my problem.

    Hey Shimmy, you sound just like those guys with Linux OS having a go at me about the silverlight app but your case lies with the JS replacing the C#

    Perhaps this is the way forward for MS giving up on good things instead of making them even better. It just leaves all those people who have been using and relying on these particular products to make a living (perhaps) in the lurch. Now they have to struggle to get up to speed to learn a new Technology.

    Will.

     

    Will Robinson

    Thursday, April 25, 2013 8:03 AM
  • Question