none
The future of Lightswitch less clear than ever

    General discussion

  • A simple look at the sessions of Microsoft Build 2011 tell the tale. Silverlight and Lightswitch are nowhere to be found. HTML5 and javascript everywhere. That is the biggest f*ck you that Microsoft is throwing to all of us developers that once were sold on the Silverlight marvels. Yes, It's not dead (yet), but we all can see where this is heading. Sorry, but I think is time to look for greener pastures. You can lock this thread now, like all the others that talk about Lightswitch future.
    Thursday, September 15, 2011 1:33 PM

All replies

  • Yeah, I'm sure Microsoft spent the last two+ years developing a product that was going to die as soon as they released it.

    LS is built using MVVM principles, so the "client" is just one small part of the whole. The Silverlight client can be "swapped out" for a different technology, at any time they feel like it. I wouldn't be surprised at all if the "next" floavor of client isn't already under development right now.

    Please, I know that MS makes some really dumb decisions sometimes, but LS has actually been very well thought-out, & fairly well (if not extremely) "future-proofed".

    And I don't see a question anywhere, so this should probably be a "discussion", not a question.

     


    Yann

    (plus ça change, plus c'est la même chose!)

    If you found this post helpful, please "Vote as Helpful". If it actually answered your question, remember to "Mark as Answer".

    This will help people find the answers that they're looking for more quickly.

    Thursday, September 15, 2011 1:56 PM
  • Yeah, I'm sure Microsoft spent the last two+ years developing a product that was going to die as soon as they released it.

    Just tell that to the Microsoft Kin, after over 1 billion invested and two years of development it was discontinued after 48 days on the market due to poor sales.

    So, don't rule it out.

    http://en.wikipedia.org/wiki/Microsoft_Kin

    Thursday, September 15, 2011 2:10 PM
  • I am still sore from investment in MS Accounting that was cancelled out of the blue.  I suspect the whole thing was some kind of leverage against Quicken and product was never really intended for long term.  The funny part is, LS UI model looks similar to MS Accounting... One of the reasons I liked it and thought it would take off.


    Thursday, September 15, 2011 4:05 PM
  • Another thing I hope they concider is a typical LS app may need multiple Tiles as entry points.  For example, you may want an Invoices tile, a vendors tile, a Notifications tile, etc.  The respective tile will have relative context and take you to relative context.  A tile may also be a great widget to be able to send in an email or add to web pages as contextual nuggets.
    Thursday, September 15, 2011 4:38 PM
  • Forget about SL for a sec. How about no more .Net?  Seems they are boiling the ocean again with WinRT.  I about fell off my chair. Why did Bill G not call me on this?  I guess this is why Scott Gu has moved over to Azure.

    http://channel9.msdn.com/Events/BUILD/BUILD2011/BPS-1005


    Thursday, September 15, 2011 5:01 PM
  • One cool thing not immediately obvious is the end of browser wars and browsers all together.  Windows itself becomes the browser and runs html/java code or c# or c++.  So in theory, mac or linux just needs a WinRT layer and all browsers issues are mute. Interesting solution to a big problem to work same way across all platforms.

    Thursday, September 15, 2011 6:17 PM
  • Any idea how well lightswitch is selling?
    Thursday, September 15, 2011 6:26 PM
  • I have a bad feeling that LS will die or simply microsoft abandons him because now it gives all the thrust to metrostyle
    Thursday, September 15, 2011 10:45 PM
  • Bad news to start my day!  

    I just shifted my gear from Visual Foxpro to Visual Studio LightSwitch,  and while accelarating with it, MS has put a big bump along the way.   I spent my entire 8 months sitting infront of my desktop just to learn .net and LS. What is this,  Playing games?

    Thursday, September 15, 2011 11:35 PM
  • WOAH EVERYBODY!

    Please remember that Silverlight WILL work in Windows 8.

    LighstSwitch will work in Windows 8 (I tested this myself).

    We are good for a minimum of 4-5 years even if the LightSwitch team does not do ANYTHING. However, they WILL do something to keep current.

    LightSwitch IS the safe bet if you need a Line-Of-Business application (as opposed to the Twitter clients that you see in the Windows 8 Metro demos)


    Make Them Ask: That's a LightSwitch App?

    http://LightSwitchHelpWebsite.com


    Thursday, September 15, 2011 11:41 PM
  • Like Michael says, "if you don't use LS, what would you use"? 

     

    To me, it doesn't matter MS gonna burry SL or not.  LS is one of the best thing happened in my dev(!) career. :)  Even if SL not gonna make it v6, I hope MS will still support (or at least hold) SL5 for many more years. 

    And also, I mainly use LS to create our Enterprise intranet apps. And we don't jump on to new technology like public do.  So we should be good for next 5/6 years.

     

    After that??? I'm trying to be optimistic here!! :)


    -Bala
    Thursday, September 15, 2011 11:42 PM
  • Well thanks for pacifying us Mike at least for the moment.  We'll be anticipating words coming from MS themselves.
    Thursday, September 15, 2011 11:59 PM
  • Also, most of LS is the stuff you will still need in a metro style app. So all the code stuff would be mostly same.  The controls and some apis will change, but mostly we don't touch the native controls anyway, unless you have a need to reach in and change specific property.  I suspect this is why they build the Shell extensability in from the start. A new shell, for example, could be metro. I am intrigued by the death of the message pump.
    Friday, September 16, 2011 12:06 AM
  • Well thanks for pacifying us Mike at least for the moment.  We'll be anticipating words coming from MS themselves.


    Believe me I understand.

    I mean I am a Microsoft Silverlight MVP and I am hearing that "Metro" will not allow any plug-ins including Silverlight.

    However, I am never the "last guy left to turn out the lights".

    I knew this change was coming. Microsoft was clear that their "strategy has shifted"

    When LightSwitch came out and I heard that it was going to use the MVVM binding that I knew Microsoft was committed to I gave it a serious look. I realized that it was designed to work with the changes that were being planned. That’s why there is no screen designer in LightSwitch just a “object tree”.

    LightSwitch has not painted themselves into a corner at all. Even Silverlight Custom Controls are easily portable to Metro because they abstract away everything when they force you to use MVVM for all binding.

    I follow the winners and the LightSwitch team is full of all stars that haven’t made a misstep yet.  

    For example, take printing, The printing in Metro will have to be different. LightSwitch sticks to the parts that it must handle and allows you the freedom to use components for  the other parts. Components that will probably be outdated over time. The LightSwitch core stays solid and usable for years.


    Make Them Ask: That's a LightSwitch App?

    http://LightSwitchHelpWebsite.com

    Friday, September 16, 2011 12:10 AM
  • Now you realy pacifies me, I am calm now. LOL

     

    Rkage

    Friday, September 16, 2011 12:27 AM
  • A company that we are developing a new vertical LOB app had one of their major clients walk out on them a month ago due to choosing Silverlight as the client side technology. The app was promoted as Silverlight, rather than Lightswitch, as non-technical people may not necessarily understand something being marketed as being developed in Lightswitch.

    After much deliberation, the client was pulled back and they came on board. Mostly as they had no alternative in my opinion, rather than being OK with Silverlight. One of the points that helped IMO was the fact the application was built using a custom RIA domain service sitting in front of the backend hosted business layer, etc and LS was only used as a UI layer, so in theory the UI could in theory easily be replaced or augmented by an HTML5 layer should the need arise.

    The Windows Build event, although interesting, unfortunately does not do a lot to ease the perception that Silverlight is something of the past. To the contrary, it has probably firmed up that perception. To that extent (and as already experienced first hand) I do not think we will be able to start another major development using LS on top of Silverlight. This is not because we don't want to, but because customers will not allow us to do so.

    We can speculate all we want here, but I hope that MS really lays out the path of LS going forward in the very near future.

    In addition, we need to let all the dust from this week's events settle down so that people can start analyzing in detail what all this means.


    Xander
    • Edited by novascape Friday, September 16, 2011 2:11 AM
    Friday, September 16, 2011 1:46 AM
  • If but I am in the clouds as the question is LS lasted 4 years LS team already thought about the changes coming in win 8 and to change the customers with html5 or simply using metro
    Friday, September 16, 2011 1:55 AM
  • A company that we are developing a new vertical LOB app had one of their major clients walk out on them a month ago due to choosing Silverlight as the client side technology. The app was promoted as Silverlight, rather than Lightswitch,


    Xander

    In the end it will come down to money. Right now a huge web app costs the same as the next huge web app. A LightSwitch app can cost 90% less.

     


    Make Them Ask: That's a LightSwitch App?

    http://LightSwitchHelpWebsite.com

    Friday, September 16, 2011 4:30 AM
  • It would be nice is MS comes out and say that LS V2 will support all of the following UI outputs:

    1) Silverlight (done)
    2) Metro XAML
    3) Metro HTML/CSS
    4) Standard HTML/HTML5

    Happy to choose one of the above at the start of a new project as it might be more difficult to support all 3 simultaneously for a single project...

    Crossing fingers :)


    Xander

    • Edited by novascape Friday, September 16, 2011 4:52 AM
    Friday, September 16, 2011 4:47 AM
  • It would be nice is MS comes out and say that LS V2 will support all of the following UI outputs:

    1) Silverlight
    2) Metro XAML
    3) HTML and/or HTML5

     


    Xander


    That would be world domination :)

    I expect:

    1) Now

    2) 2012

    3) 2013


    Make Them Ask: That's a LightSwitch App?

    http://LightSwitchHelpWebsite.com

    Friday, September 16, 2011 4:50 AM
  • This stuff not even in beta yet.  Talking years away.  They want to wait around a few years?  That would be a circular problem as everything changes after 6 months in the business.  Clients need to stop looking at the shinny bouncy balls and get back to work :-)

     

    Friday, September 16, 2011 6:43 AM
  • Just curious-

    Given Michael's speculation, would this timeline be unreasonable? Neither confirming nor denying anything, but how fast do you expect the uptake of Win 8 will be for businesses, especially small to medium businesses?

    Regards,


    Steve Hoag Microsoft aka the Lights Witch (IEnumerable of Newt)
    Friday, September 16, 2011 6:46 AM
  • I want to throw some additional thoughts in this discussion/speculation.

    I would also be happy if LightSwitch would have been mentionend at //build but I think //build is more about the technological foundation for the future. Talking about LS is in my opinion primarily not really talking about a certain technology but about having a best practices framework for building data driven business apps.


    I think it's rather clever from the LS-Team to hide as most internals as possible, so that it can be replaced in the future. And that's what I'm really expecting. If HTMl5 etc. will be the future, then I'm hoping/expecting that LS will adjust LS accordingly.

    And to be honest I think LightSwitch is not the first class citizen for the hardcore developer who want to tweak every detail or for huge companies with world wide teams. I think the //build conference and the shows are made for them to get them on the train for the future technologies.
    I probably as most LS developer, are more business and end user focused and for me it's important that Microsoft takes care about the right technology, and we will take care to bring solutions to the customers and that's where Microsoft makes money, because the end user are connected to Microsoft with the help of our products and then the user will also have windows, use Office etc.

    If Microsoft would abandon LS they would fail in my opinion with creating a best practices framework for small to medium size business apps with their on toolset.

    I think it's very true what Michael Washington had said in one of his blog posts. I think it was about 3rd party controls but it's also true if you wouldn't have a framework like LightSwitch and have to create such stuff on your own (which is by the way normally impossible because of time constraints)

    He wrote: "In the end, we end up with choice. “choice” is an odd thing, you only want it if you need it, otherwise “choice” causes you to have to do too much work… you have to make a choice. It’s not that you don’t want to do the work, it is just that you are always afraid of making the wrong choice, and that is stressful."

    To come to an end I think //build is not the right conference for expecting LightSwitch stuff and if Microsoft would have decided to have a few sessions then people would say, look there are much more sessions for xy, so this could mean LS is on the skids.

    LS is the tool I'm absolutly waited for and if lots of people creating solutions with it, then it's the best thing we could do to keep it developing in the future. In the end it's all about money.

    Concerning the time frame Michael speculated I could perfectly live with. I can live with the current LS solution quite a while and the only wish I would have is to see a strong commitment to LS from Microsoft from time to time, because it's not just all about money it's also about psychology.

    Cheers

    Uwe

    Friday, September 16, 2011 8:58 AM
  • Just curious-

    Given Michael's speculation, would this timeline be unreasonable? Neither confirming nor denying anything, but how fast do you expect the uptake of Win 8 will be for businesses, especially small to medium businesses?

    Regards,


    Steve Hoag Microsoft aka the Lights Witch (IEnumerable of Newt)


    Steve,

    In my opinion it is more a "fear issue" rather than a "I actually need this" issue. What I am trying to say is that while the real truth is that Silverlight will work fine for everyone for the next 5-10 years, people are afraid Microsoft is about to announce some new tool and basically say:

    "Hey stupid, you 're now suppose to use this new HTML/JavaScipt ‘tool’ to create business applications"

    • I am an idiot
    • My boss and my co workers will think of me as an idiot
    • I should have just used HTML and JavaScript


    Now, Microsoft has been clear from the very beginning that Silverlight was a "sweet spot" for Line of Business applications. It was just not what people wanted to hear (I admit including me). All that happened at //build was more information, but it remains consistent. This is why I believe that LightSwitch WILL remain and will continue to use Silverlight.

    ----

    What people want is for the LightSwitch team to say:

    "Yes we can output 'other things' like Metro, and we are working on it"

    When this does happen, people will try Metro output for certain things, but realize that Silverlight provides the best performance (and also works on the MAC (Metro does not work on the MAC).

    Now I know that the team already said that! (“Yeah, the scenario is important and we want to support in the next version”) , but that was before //build and we assumed it was about HTML. But now I realize that it was probably about Metro so please make a Blog post and say it again :)

    ----

    HTML output, however, is "world domination", plan and simple :)
    I NOW understand why LightSwitch won't do HTML output it now:

    1. Metro output must come first. Metro is FAR from done. It is where we were with Silverlight 2. Metro has to get to the abilities of at least Silverlight 4 before LightSwitch can have what it needs to work right.
    2. HTML5 will not have most of the browser incompatibilities sorted out for 2-3 years. LightSwitch must has a "stable experience".

    Further reading:
    http://social.msdn.microsoft.com/Forums/en-US/lightswitchgeneral/thread/5cbc655d-0f43-43f4-8a7d-959f44929c3d
    http://social.msdn.microsoft.com/Forums/en-US/lightswitchgeneral/thread/0cd8c86e-a727-472f-ad66-433c8f8e70e9/#113f942e-1cb1-47ef-8ed0-0ad2d1920daf


    Make Them Ask: That's a LightSwitch App?

    http://LightSwitchHelpWebsite.com

    Friday, September 16, 2011 1:33 PM
  • I think that there is no future even my already happened with visual fox, silverlight and now that it happened with Ls is fear forever Microsoft abandons us on the road
    Friday, September 16, 2011 1:45 PM
  • Of course Lightswitch, nor Silverlight are dead (yet). I have big VB6 apps running even today.

    But Build was the official statement on the Microsoft development world's affairs. Lightswitch, Silverlight and even .NET framework are second rate, niche tools that have started to go the way of the dodo. Of course they still work. Of course LS is the best enhancement to productivity in LOB apps since VB. But they are not the future. So, who's going to tell your boss or customer you have chosen VB6 for your next work?

    The new paradigm, metro and winRT, is LS-free, Silverlight-free and even .NET-free. Is a paradigm to compete in the smartphone, tablet and even desktop world. But condems us LS types to a legacy niche. Maybe Microsoft provides us with a Metroswitch, where the UI is HTML5/js, but are you going to bet your income on that maybe?

     

    I still think that LS is a great product. I just don't think it has great potential anymore.

    Friday, September 16, 2011 2:22 PM
  • "Lightswitch, Silverlight and even .NET framework are second rate, niche tools that have started to go the way of the dodo. Of course they still work. Of course LS is the best enhancement to productivity in LOB apps since VB. But they are not the future. So, who's going to tell your boss or customer you have chosen VB6 for your next work?"

    Hi Kangurill.  Not sure about some of that.

    1) .Net has not and is not a second rate tool. It has been *thee best framework in computer industry for 10 years. I has been better then Duracell. That includes docs.

    2) AFAICT, WinRT is just a native version of .Net framework, but shipped with OS. Interface callable by both native and managed code. There are a few reasons for that including better perf and OS cross-dependencies. The OS shell folks could not figure out a way to build OS stuff on .Net because of dependencies as it was an out-of-band and optional library.  With this bold step, they are saying .Net is great enouph to be first class ontop of the OS - not just a side deal. The extra speed of the .Net/WinRT runtime will not hurt anyone either. ;)

    3) LS is three tiers. The server tier and middle tier have nothing to do with WinRT as such. The client tier is really two layers. The UI and client side biz logic including RIA proxy.  The UI part would only need to change. The other stuff should just work (at least in theory). There could be some namespace changes and some api changes I would expect, but mostly same. Point is, if they created a whole new LS for WinRT, that would have duplicate 90% of existing logic.  So I would look at it more like a vNext thing, instead of a forklift.  LS will evolve, not be replaced. This will happen till the end of time.

     

    Friday, September 16, 2011 8:37 PM
  •  if they created a whole new LS for WinRT, that would have duplicate 90% of existing logic.  So I would look at it more like a vNext thing, instead of a forklift.  LS will evolve, not be replaced. This will happen till the end of time.

     

    I don't think they would ever make LightSwitch run in Metro (WinRT), just output to Metro (WinRT).

    Make Them Ask: That's a LightSwitch App?

    http://LightSwitchHelpWebsite.com


    Friday, September 16, 2011 8:40 PM
  • I see no reason they would not. If they believe there own dog food, then it should not take much changes to make it a "native" WinRT citizen as they showed converting a Scott Gu sample app.  There are a few reasons why they would really want to. Such as integrating with the metro ecosystem of common controls, look and feel, multi-touch, html5 support,  UI subsystem, and others.  I for one will love to see the day I can dev my LS app once and target Cell, Tablet, and PC and things just work.  In theory, they could port WinRT layer to Mac and Droid and things would also just run.  Kind of reminds me of a new HAL for Web apps, but on top of the OS instead of at the bottom.  I suppose Posix kinda attempted that on NT but never got anywhere.  Also side steps all the 50,000 issues with supporting 4-5 different browsers and nuance.   
    Friday, September 16, 2011 8:58 PM
  • That's what I'm talking about. That is the commitment I want to see from Microsoft. The Lisghtswitch with Metro UI ("Metroswitch" as I called before). Because I see W8 Tablets as clients, and in that land there are no .NET, Silverlight clients. There are only fast, responsive UI, on native xaml that runs on winRT. And winRT is no .net in the sense that is not managed, so don't tell me they are the same. If you have not followed Build you will discover that the new magic language is....... C++!!  unmanaged!  I almost fell off the chair. This is not serious, and that is the reason MS is doing so poorly in the enterprise, in terms of development. You don't leave your investors and developers in the clear. You care for them. You tell them there is some way them and their skills can stay relevant. My impression is that MS wants to conquer the smartphone and tablet space even at the cost of the desktop space.
    Friday, September 16, 2011 9:12 PM
  • Hmm well I guess when Visual Studio runs as a native WinRT app then LightSwitch will run as a WinRT app.

    I have heard no plans of this. The Visual Studio that is in the Windows 8 is running in desktop mode.


    Make Them Ask: That's a LightSwitch App?

    http://LightSwitchHelpWebsite.com

    Friday, September 16, 2011 9:16 PM
  •  

    I still think that LS is a great product. I just don't think it has great potential anymore.


    Remember LightSwitch IS Visual Studio.

    You don't believe that Microsoft is going to abandon Visual Studio?

    See Server and Tools below:

    http://www.tannerhelland.com/2962/where-does-microsoft-make-its-money-2010/

     


    Make Them Ask: That's a LightSwitch App?

    http://LightSwitchHelpWebsite.com

    Friday, September 16, 2011 11:18 PM
  • There's a slight difference between a "me too" phone product that nobody wanted, & a revolutionary, almost nothing in the market like it software product that nearly everyone that sees it wants it, don't you think?

    Yann

    (plus ça change, plus c'est la même chose!)

    If you found this post helpful, please "Vote as Helpful". If it actually answered your question, remember to "Mark as Answer".

    This will help people find the answers that they're looking for more quickly.

    Friday, September 16, 2011 11:35 PM
  • "And winRT is no .net in the sense that is not managed, so don't tell me they are the same. If you have not followed Build you will discover that the new magic language is....... C++!! "

    1) Your talking like WinRT and W8 is a product for a year.  This stuff is still vabor right now.  Close your eyes for a year.  I don't chase the shiny balls any longer until after Beta2.  You can get a crap load of work done between now and then so I would not factor it in yet.

    2) I watched all Build shows.  Not sure you read that 100% right.  The library is Native, but you can still run c# and vb ontop. The fact that the internals are native c++ makes no difference (except being faster).  All of .Net eventually makes native calls today. It will still have a runtime. No diff to your c# view of the world.

    Friday, September 16, 2011 11:43 PM
  • "Hmm well I guess when Visual Studio runs as a native WinRT app then LightSwitch will run as a WinRT app."

    Not sure VS actually needs to run on WinRT to do this. It needs to Produce the right exe's and dll's that can run on WinRT. It is probably a new compiler and libraries.  They gave away 5000 of these WinRT developer tablets with VS and dev tools. So I assume they have the general workings fleshed out. I would completely expect a WinRT native version of VS in 5 years. It is way too early for them to announce any key app on winrt as the platform is still vabor ware.  I suspect selective and gradual migrations. For example, not sure how apps like Word would even benefit.

    Friday, September 16, 2011 11:58 PM
  • I think the whole issue at stake here is one of perception. Perception from the client, not from the development community. We understand that it is status quo for the foreseeable future, but the marketing machines out there are hyping a different story.

    I do not know enough about Metro style apps just yet, but my suspicion is that business applications (what we target with LS) will continue to run as Desktop Apps and will not be running as Metro style Apps (for a long time anyway).


    Xander
    Saturday, September 17, 2011 12:03 AM
  •  For example, not sure how apps like Word would even benefit.


    For Office they may just adapt the Office that they put on the phone. It works fine but of course on the phone the screen is too small.

    I also suspect that the XBox will get this Metro/WinRT thing and I can imagine with a wireless keyboard I can use Microsoft Word to write a letter using my XBox on my TV.

    So WHEN I am able to output a Metro/WinRT app from LightSwitch and host it on Azure, people can hit the site on their Windows tablets or their XBox and use my app.

    So we STILL need Visual Studio and LightSwitch to create these apps.


    Make Them Ask: That's a LightSwitch App?

    http://LightSwitchHelpWebsite.com

    Saturday, September 17, 2011 1:30 AM
  • So WHEN I am able to output a Metro/WinRT app from LightSwitch and host it on Azure, people can hit the site on their Windows tablets or their XBox and use my app.


    That, for sure, will be cool.  And maybe someday on the iphone/ipad.
    Saturday, September 17, 2011 4:34 AM
  •  
    dream lot Wyliam  xd
    Saturday, September 17, 2011 4:36 AM
  • I tend to agree. We develop some very large commercial enterprise LOB applications and for many – many years we were in VFP. This was a strength because we could easily surpass our competition due to the speed and dynamic nature we were able to develop and pump out functionality and new solutions. Over the past 4 years we have been migrating from VFP (desktop) to HTML (VFP backend) and now have almost completely replaced the VFP backend with ASP.net. No easy task when you’re dealing with millions of lines of code (even though the business logic and rich client UI are seperated). Although we can do quite a bit with HTML, JacaScript and AJAX, with 20 years in the business, I am still not convinced one can achieve the same dynamic, high performance UI for a data centric, LOB applications using HTML as opposed to a rich client app. And if this is doable, it won’t be easy unless you are very well grounded in HTML5, Javascript, AJAX, .Net, perhaps flash or Silverlight (just in case) have a great LOB framework to build upon and are well grounded in Internet communication protocols and concepts in asynchronous programming, IIS and server or cloud deployment processes. This is the reality of what we have to contend with when developing commercial grade, data centric LOB globally deployed apps.

    I have  been following LightSwitch for a few months and this is the first product I have seen (since VFP) to actually bring productivity back to data centric LOB app development. Once you get .Net under your belt, I actually believe a developer may be more productive with LS then VFP because much of the deployment headaches are taken care of. I also believe a LS developer can be far more productive in the creation of a data centric LOB application with LS then without (using .Net technology). I am very excited about LightSwitch and I am sure it will be around for (at least) a few years. After having tools like VFP for 15+ years, I think we tend to get a bit spoiled and believe that unless we have some confirmation that a new tool or platform will be around long after we are gone, it’s not worth investing our time and development resources in. Unfortunately, this is not the reality and we have to live with what’s available and be able to change when needed. I think with a properly developed application (even with one based on LS), at worst, you would be looking at changing out the front end in a few years – unless Microsoft takes care of this with some iteration of LS.

    I also believe LS is not being given the credit it deserves from many in the development community. It seems extensible enough and when I ask others in the community why they feel LS not worth their time, I usually hear something like – it’s not flexible, extensible or scalable enough. I have a running bet with a few friends to prove this and to date I see no evidence to the contrary. Perhaps many developers feel that if they did not create the framework themselves, it’s not good enough. Coming from a development background, I know the feeling but after doing this for 20 years, eventually you realize that you don’t always need to create everything yourself from scratch.

    I really think LS can be used to develop some great extensions and standalone presentation UI’s into many LOB systems (that already exist) and look forward to learning everything I can about this product. I’m also watching the third party extension community very closely since that’s usually a good forecast of things to come.

     

     Benjamin 

     


    • Edited by NGCBSiegel Sunday, September 18, 2011 3:19 AM
    Sunday, September 18, 2011 2:57 AM
  • According to the new chart I think we're going to be ok.

    Tuesday, September 20, 2011 7:19 PM
  • About time we got a LightSwitch roadmap!
    Tuesday, September 20, 2011 7:30 PM
  •  

    Is that it for this thread?

    Can't development make some general statements without revealing something strictly confidential?

    If development can't state something like:

    - There is a commitment for a long term life of this product, with other versions planned

    - There is a commitment to explore creation of apps other than Silverlight, although nothing will be announced at this time

    Then one would have to conclude that the product really Is dead.

    This product fits my organization perfectly - but we Can't build a business dependency on it without some statement of intent that there will be a future for this product - that includes non-Silverlight app creation.

     


    • Edited by FR MSDN Wednesday, September 21, 2011 12:57 PM
    Wednesday, September 21, 2011 11:47 AM
  • One thing is for sure. This is a healthy community with lots of different points of view. I'm most certainly glad to see that. Nothing fartest from my mind than Microsoft abandoning .net or something along those lines. But the marketing machine is full steam to metro and W8 and WinRT. It would be sad that this momentum was harming for Lightswitch. You see, for me this technology never was an ms access replacement. Never. It was obvious from the beguining that lightswitch was a wonderful promise for high productivity in the LOB world.  I would hate to let that go...

    Wednesday, September 21, 2011 11:50 AM
  • Hi,

    LightSwitch is far from dead. Without revealing anything specific I can confirm that the following statements are true:

    - There is a commitment for a long term life of this product, with other versions planned

    - There is a commitment to explore creation of apps other than Silverlight, although nothing will be announced at this time

    Regards,

     


    Steve Hoag Microsoft aka the Lights Witch (IEnumerable of Newt)
    Wednesday, September 21, 2011 7:37 PM
  • Thanks Steve.  This is really good news and anything you can tell us to reassure ourselves and our customers will go a long way.

    A Roadmap would be huge.

     

    Thanks,

    Dave

    Wednesday, September 21, 2011 7:47 PM
  • LOL = Steve's Sig


    Steve Hoag Microsoft aka the Lights Witch (IEnumerable of Newt)

    Wednesday, September 21, 2011 7:49 PM
  • Hi,

    LightSwitch is far from dead. Without revealing anything specific I can confirm that the following statements are true:

    - There is a commitment for a long term life of this product, with other versions planned

    - There is a commitment to explore creation of apps other than Silverlight, although nothing will be announced at this time

    Regards, 


    Steve Hoag Microsoft aka the Lights Witch (IEnumerable of Newt)


    That's all I needed, and I'll bet that's all that most need for now.   Thank you very, very much, Steve.

     

    Wednesday, September 21, 2011 7:52 PM
  • Hi,

    LightSwitch is far from dead. Without revealing anything specific I can confirm that the following statements are true:

    - There is a commitment for a long term life of this product, with other versions planned

    - There is a commitment to explore creation of apps other than Silverlight, although nothing will be announced at this time

    Regards,

     


    Steve Hoag Microsoft aka the Lights Witch (IEnumerable of Newt)


    This works for me :)

    I do understand there are reason you can't say more.


    Make Them Ask: That's a LightSwitch App?

    http://LightSwitchHelpWebsite.com

    Wednesday, September 21, 2011 8:06 PM
  • Where did this chart come from?
    Thursday, October 06, 2011 12:15 PM
  • We will see light switch to any other UI. XAML, nothing else matters ...
    Petr Antoš
    Thursday, October 13, 2011 6:24 PM
  • If Microsoft can Kill Silverlight (well at least wanting to kill) when SL is used by millions of developers around the world and for hundred of thousands of companies plus  fortune 500 , so far enterprise has not allowed Microsoft to kill SL...

    But I wonder, if LightSwitch not one of those oops projects from the new changing Microsoft

    Microsoft should use LightSwitch and make it the framework that allows .Net developer to build HTML5 + JQuery applications, becasue obviusly that is where the new Microsoft is heading.

    setting emotions and tie invested in LightSwitch aside... will it survive and be next to Windows 8 HTML5 /JS ?

    loot at Prism for example.

    where is Ria services V2 ?

    why all the masterminds from Silverlight, Ria services are pretty much gone from MS.... even Karl from prism is also gone

    Regards

    Thursday, December 01, 2011 6:32 PM
  • Hi,

    LightSwitch is far from dead. Without revealing anything specific I can confirm that the following statements are true:

    - There is a commitment for a long term life of this product, with other versions planned

    - There is a commitment to explore creation of apps other than Silverlight, although nothing will be announced at this time

    Regards,

     


    Steve Hoag Microsoft aka the Lights Witch (IEnumerable of Newt)

    The problem with creating apps other than Silverlight is that in many cases existing user code is tied to System.Windows.Controls.   For example, something as simple as code that sets alternating colors in DataGrid rows now has a dependency on a particular UI.  To prevent that code from breaking, MS would have to come up with an alternate assembly that mirrors System.Windows.Controls and implement the functionality of all those controls in the web client.  What about third party libraries?  It just seems unreasonable.

    If Silverlight 5 is the last version of Silverlight, then most likely, LightSwitch is at the end of its line as well.  As @Dave Vorgang hints at in his edited version of the Win8 platform, the only thing they can effectively do is retool LightSwitch into a new development tool (MetroSwitch ?) which leaves existing LightSwitch applications having to be ported if they want to move forward with platform functionality.

    As someone not yet using LightSwitch or Silverlight, and evaluating the pros and cons, I cannot see any point in investing in a platform that has what appears to be a dead end already defined.  The story would be different if MS would come out and say that SL 6 or perhaps even LS 2 are in the queue, but they have not.  It's too bad because we're looking for a RAD alternative to Lotus Notes.

    Tuesday, December 13, 2011 9:16 PM
  • Hi Rich,

    LS2 is in the queue.  Hey, that was easy :) Welcome aboard!

    Tuesday, December 13, 2011 9:49 PM
  • Hi Rich,

    LS2 is in the queue.  Hey, that was easy :) Welcome aboard!


    And that statement comes from where?
    Tuesday, December 13, 2011 9:56 PM
  • It comes from the Quote embedded in your prevoius post from Steve Hoag. ("- There is a commitment for a long term life of this product, with other versions planned").

     

    Tuesday, December 13, 2011 10:02 PM
  • Hey Guys, Lots of smart people here. Most are worried about the future of this great technology, but i think the tea leaves are decently clear if you just look at what they have said: 1. Lightswitch just builds a model Lightswitch was designed around MVVM and abstracts you away from the details of silverlight. You don't write silverlight, the compiler does. Take a look inside the lsml file. 2. RIA JS provides the plumbing for html5/js clients for ria WCF RIA Services Toolkit (September 2011) Also includes an early preview of the jQuery client for RIA Services (RIA/JS http://www.microsoft.com/download/en/details.aspx?displaylang=en&id=26939 3. At SQL Pass, Microsoft demoed Power View running on an IPAD and Andriod. This is a custom silverlight app that was developed over 2 years by MS. 4 Months post RTM Microsoft has committed to making it work across mobile devices with a reasonable fidelity interactive experience. MS has said there is only one code base as well. The video is up on youtube Roll it all up and it kinda makes sense. Sometime in 1H of 2012 Microsoft gives us the rest of the story. We all know that HTML5/JS is not a good way to write an app but I don't really care what MS outputs things to when we build. I assume you will see a XAML output type for windows apps and html5/js output type for cross platform apps. With the growth of post pc devices the cost of created SL runtimes would sky rocket. Rather then rewrite WPF/E on all these platforms html5 provides a good enough experience while reserving the premium experience for users of Windows. It all makes sense and fits with the things we've seen. Time for them to unveil the plan pretty I hope.
    Thursday, December 22, 2011 3:09 AM
  • Hey many around here still use COBOL and their systems still work.

    All of these technologies reinvent themselves many times over before they settle. And they will only settle once they are open. HTML/CSS are open royalty free standards. It's why there's so many video formats. MPEG4 and H.264 have so many patents by dozens of different companies registered against it that they have a seperate company just to manage royalty collection and distribution. It'll never last, its too complicated to manage. Content providers will find a better way.

    Flash and Silverlight are depreciating technologies because content providers have found a less complicated open way, thanks to W3C.

    Sunday, March 11, 2012 7:34 PM
  • Sidestepping the browser issues is the biggest thing you said.  That is a true time suck for developers, like printer support in the '80s.
    Tuesday, April 24, 2012 1:15 PM
  •  Silverlight and Lightswitch are dead.
    Saturday, December 29, 2012 10:15 AM
  • Sorry to say, that  this  "statement" is not extremely informative. 

    paul van bladel

    Saturday, December 29, 2012 4:37 PM
  • Sorry to say, that  this  "statement" is not extremely informative. 

    paul van bladel

    I agree with Paul.  Silverlight is a fully supported technology at this time.  LightSwitch will continue to evolve with its HTML Client support. 

    LightSwitch is part of Visual Studio - and I just checked . . . Visual Studio still has a pulse.


    Garth Henderson - Vanguard Business Technology

    Saturday, December 29, 2012 5:31 PM
  • Indeed, due some specific market conditions (the S. Jobs effect), we know already now that in 10 year silverlight will go end of live. 10 years that's, in terms of IT time perspective extremely long. Other technologies come and go and we just accept that once in a while something disappears. Be happy, we have this clear perspective, especially with a technology like lightswitch which is picking up new client technologies (sharepoint apps, jquery mobile, ...) at super speed.

    I have stopped (when it comes to IT) to reason in terms of what's in the pipeline for the next years. I want to build apps now, which mature technology. At this very moment silverlight is the most mature technology for line of business apps (How many Jquery,  knockout, single page apps, asp mvc apps do you know in enterprises currently ?) I'm quite confident that the apps I build today with LightSwitch (and the silverlight client) can be ported to future versions of LightSwitch supporting other UI technologies. 


    paul van bladel

    Saturday, December 29, 2012 5:50 PM
  • very interesting discussion .

    I have used Silverlight (my baby) to develop (in the past) robust Enterprise Solutions and nothing comes even close to it right now. Unfortunately the Iphone, Ipad and Android are becoming crazy popular in the hands of the customers, So we were forced to learn HTML5, Knockout, Breeze, RequireJS, JQgrid, Jquery, and moved to MVC. no because we wanted to , but because the enterprise market is asking for it. JavaScrip is so unorganized and nothing like the first class products we have in .NET.

    SPA is the way to go, definitely; but is yes, there is no good SPA applications at this point. maybe in one more year. Gmail and USAToday are probably the only true SPA.

    Microsoft is very aware of this trend to HTML + JavaScript so (I think) LightSwitch will be the perfect platform to build solid SPA in the future. It makes so much sense.

    on the server : SQL Server 2012, Entity Frameworks 6, Models (POCO clases) ,  Repository (for the business logic) , contracts (interfaces) , etc.

    on the client: Html5 + JavaScript, plus all the JavaScript libraries the team wants to use.

    I think LightSwitch will deliver it, full power, with good Code Practices and best Patterns. Very easy to migrate from Silverlight to HTML5
    . and you can have your application ready now and latter migrate it to pure HTML5 if you need to
    • Edited by Oscar Agreda Saturday, December 29, 2012 8:02 PM
    Saturday, December 29, 2012 7:59 PM
  • > Microsoft is very aware of this trend to HTML + JavaScript so (I think) LightSwitch will be the perfect platform to build solid SPA in the future. It makes so much sense...I think LightSwitch will deliver it, full power, with good Code Practices and best Patterns. Very easy to migrate from Silverlight to HTML5
    . and you can have your application ready now and latter migrate it to pure HTML5 if you need to

    While that may become true, today we don't have a concrete enough statement on what direction the HTML support is Really going in to be able to make a sane business plan based on all of this.

    My shop is at a crossroads with development technologies and we'd love nothing more than to standardize on LS, but we need to know that there is desktop browser Silverlight-less path coming soon.  [as in, inside of a year]  Until I can get a clear statement on that, I'm stuck being unable to make the case for widespread LS use.  Something tells me I'm far from alone.

    Friday, January 04, 2013 3:31 AM
  • My shop is at a crossroads with development technologies and we'd love nothing more than to standardize on LS, but we need to know that there is desktop browser Silverlight-less path coming soon.  [as in, inside of a year]  Until I can get a clear statement on that, I'm stuck being unable to make the case for widespread LS use.  Something tells me I'm far from alone.

    100% agree and support your statement as we are in exactly the same boat. We really do need desktop browser HTML support in LS with at least the same capabilities as the Silverlight version has today.

    You may want to follow/contribute/support the following thread in the HTML preview forum: Will the RTM version of "HTML Client" support Desktop Browsers?

    Regards


    Xander

    • Edited by novascape Friday, January 04, 2013 3:58 AM
    Friday, January 04, 2013 3:43 AM
  • My shop is at a crossroads with development technologies and we'd love nothing more than to standardize on LS, but we need to know that there is desktop browser Silverlight-less path coming soon.  [as in, inside of a year]  Until I can get a clear statement on that, I'm stuck being unable to make the case for widespread LS use.  Something tells me I'm far from alone.

    100% agree and support your statement as we are in exactly the same boat. We really do need desktop browser HTML support in LS with at least the same capabilities as the Silverlight version has today.

    You may want to follow/contribute/support the following thread in the HTML preview forum: Will the RTM version of "HTML Client" support Desktop Browsers?

    Regards


    Xander

    I have.  There's nothing concrete there, either.   

    Friday, January 04, 2013 7:20 AM
  • Guys, I BET that this thing will be baked into LS soon: http://www.typescriptlang.org/

    Petr Antoš

    Tuesday, January 15, 2013 2:56 AM
  • Hmm this thread has answered the question I was about to ask. I have advise my consultancy whether to invest in LS, but my concerns were that it doesn't seem to be supported much by MS.

    I will therefore advise against, at least for now.

    Tuesday, January 15, 2013 10:53 AM
  • "it doesn't seem to be supported much by MS"

    I don't know where you got that idea from, & I certainly don't agree with it.

    LightSwitch V2 is now an integral part of Visual Studio 2012, & with the work that's going on to support an HTML client it's obvious LightSwitch isn't disappearing any time soon.


    Yann - LightSwitch Central - Click here for FREE Themes, Controls, Types and Commands
     
    Please click "Mark as Answer" if a reply answers your question. Please click "Vote as Helpful" , if you find a reply helpful.
     
    By doing this you'll help others to find answers faster.

    Tuesday, January 15, 2013 1:43 PM
  • Hello Yann,

    I can understand PuzzledGeek2 and I can only repeat myself - the problem is the poor information policy of Microsoft - no Roadmap and no information about the future (what's comming next; future of Silverlight;)

    the new HTML 5 is a good extension of LS but HTML 5 is a hype and mobile web apps are not the primarily LOB apps (my opinion)

    so for me after building one app I will stop building new ones until it is clear what's comming next

    robert

    Tuesday, January 15, 2013 2:03 PM
  • the new HTML 5 is a good extension of LS but HTML 5 is a hype and mobile web apps are not the primarily LOB apps (my opinion)

    robert

    Robert, I totally disagree with what you've written here.  HTML5 is the best solution for RAD apps across all devices.

    The roadmap for LS is to use OData web services for RAD managment of Model-ViewModel and having universal HTML View clients that support commercial HTML component suites.

    MS has provided an incredible solution with TypeScript that makes coding with JS bearable.


    Garth Henderson - Vanguard Business Technology

    Tuesday, January 15, 2013 9:43 PM
  • Fair enough Robert. While of course I'd like to know "more" about what's coming, the bottom line for me is that the saving in time & coding effort that using LightSwitch gives me outweighs the "not knowing".

    LightSwitch was written from the very beginning to be "future proof". It was moved from a V1 stand-alone product to an integral part of VS for V2. V3 is currently being actively worked on.

    All of that indicates a fairly high level of "support" from Microsoft as far as I'm concerned (as dumb as I think some of the "marketing" decisions have been from time to time).


    Yann - LightSwitch Central - Click here for FREE Themes, Controls, Types and Commands
     
    Please click "Mark as Answer" if a reply answers your question. Please click "Vote as Helpful" , if you find a reply helpful.
     
    By doing this you'll help others to find answers faster.

    Wednesday, January 16, 2013 1:17 AM
  • You cannot do all with a websolution HTML 5 which you can do with WPF/Winforms - here are a few points:

    • generell      response speed
    • keyboard      only operations (speed)
    • drag      & drop with Office automation
    • mass      processing
    • advanced      reporting (User Designer)

    for the main apps (CRM) my customers cannot work with websolution because of the points above - I have build a lot websolutions for the extranet and they work great but for inhouse apps I'm not a fan of that (and why allways build solutions for all devices??? – for me this is an hype)

    robert

    Wednesday, January 16, 2013 10:16 AM
  • What I see is a great need of a book that explains the intrinsics of this technology; I mean the new LS html client. I see it has so many implications and sources of study and analysis.  An e-book from ADefWebserver would be just great ... Of course I don't mean just samples of the use of the technology, but the theory and main concepts behind it. Maybe a comprehensive book like the one Yann did already about the silverlight client. 

    Bye ... 

    ADefwebserver
    Wednesday, January 16, 2013 1:34 PM
  •  An e-book from ADefWebserver would be just great ... Of course I don't mean just samples of the use of the technology, but the theory and main concepts behind it.

    Everything I have learned so far is posted in these Blog posts:

    http://lightswitchhelpwebsite.com/Blog.aspx

    We have to wait for a full LightSwitch release of the HTML Client before you will see any actual books.


    The Visual Studio LightSwitch Marketplace

    http://LightSwitchHelpWebsite.com

    Wednesday, January 16, 2013 2:04 PM
  • Everybody can see just how prolific you have been on this subject; and the blog post are there free for anybody to read. Keep the good work; many thanks for that ... 

    Bye ... 

    Wednesday, January 16, 2013 2:09 PM
  • Yes, I can't say that I wasn't initially disappointed that there wasn't a WPF client option (although I remember hearing that in the VERY early days LS was based on WPF, but then got switched to Silverlight). With the web-everything, & the push to the cloud that MS is so invested in, I can't see us poor desktop guys getting much desktop-specific love.

    But with the new ability to publish an OData service layer from LightSwitch does open up the possibility of creating a WPF application to consume it. But you would then lose all of the RAD screen design capabilities that way. But I'm not sure how fast it would be, performance-wise, compared to using EF directly.

    Still, that's almost exactly what I was doing before LightSwitch came along (WPF client, EF data layer etc), & I really wouldn't want to go back to doing all the plumbing manually, so I guess it's a trade-off, but one I'm prepared to make.


    Yann - LightSwitch Central - Click here for FREE Themes, Controls, Types and Commands
     
    Please click "Mark as Answer" if a reply answers your question. Please click "Vote as Helpful" , if you find a reply helpful.
     
    By doing this you'll help others to find answers faster.

    Thursday, January 17, 2013 2:38 AM
  • I have investgate on following MS technology.

    Silverlight 1 & 2 about 1 year and build 1 game.

    ASP.NET Dynamic -data and build 1 LOB project.

    So i will have more careful to spend time on further MS  new product.


    My Blog : http://tatmingstudio.blogspot.hk/

    Tuesday, October 22, 2013 5:59 AM
  • the biggest Problem is that there is no real strategy at M$ - we use Vusial Studio since VS 6 and have the problem not to know where going on next:

    - Winforms (is it dead?)
    - WPF (not real investments since years)
    - Silverlight (dead)
    - HTML5 (not usefull for LOB's)

    why does M$ kill all good technologies (VB6, Silverlight) and set all the energy on HTML5 which can only build simple Web/Mobileapps and not real LOB's?

    robert

    Tuesday, October 22, 2013 6:08 AM
  • From my POV, Winforms is not dead, but WPF does.

    There is no reason to create a fancy application by WPF for desktop application.

    Design the form layout on XAML is nice if spend time on how the nest element works.


    My Blog : http://tatmingstudio.blogspot.hk/

    Tuesday, October 22, 2013 7:04 AM
  • This is not a Microsoft thing but rather an industry thing.

    Your career will be in HTML5 no matter who your vendor is.


    bill ross

    Monday, December 23, 2013 8:50 PM
  • Not agree - to build real LOB's HTML5 is not the right Technology (now) - maybe in the future...
    Tuesday, December 24, 2013 1:40 PM