none
Application Virtualization VS Web based apps

    Question

  • Hello,

    This question is more of a future looking request. We are having some discussions within our company regarding application Virtualization. People are pushing towards developing web applications rather than investing in virtualization infrastructure. Besides the sandboxing argument. Is there any talking points FOR or AGAINST virtualization over developing web based applications?

    My gut is telling me to lean toward web based apps over virtualization.

    Thanks for any thoughts.

    S
    Sean
    Tuesday, January 05, 2010 9:35 PM

Answers

  • Hi,

    I asume that it never will be an either/or question.

    Of course nicely written WebApps would not require Application Virtualiuzation approaches any longer, but - shortly - there some big issues that will arise

    - You still have old Applications. It won't be possible to re-code every legacy application within the the timeframe that you specified with "5 years". Look into organizations and count the apps that are were developed before 2005. So, there will be applications that can't be webyfied and there have to be solutions to deploy them

    - It sometimes is said that "devleopers are the (deployment) admin's natural enemy". If developers would act in a way that Microsoft proposed ages ago, we would not need any Application Virtualization solution, because allmost all of the issues solved by App-V come from applications that where not coded using Microsoft's essential development guidelines. So, don't expect WebApps will act "nicely" on the Desktop.

    - A new runtime environment (call it Ajax, AdobeAir, Silverlight or whatever) will cause its own issues. We all remember the "Java applications are platform agnostic and can run whenever a JRE is there". And today. There are more then a few applications that are virtualized BECAUSE they use Java. Bevcause they need a certain version of Java - and they just can't select the "right" version if more then one versions are installed on a machine. So, AppVirt is going to be used to "bundle" these new runtime environments with certain Applications just to allow the Apps to work.

    - The first company I was working for was a Web Company. Well, luckywise they disappearded and I was forced to move to another company that later then lead me to application virtualization. But I perfectly remeber Presentation Bullte Points saying "in a few years, you can do ANYTHING from within a browser". This was about 2001 and still there are Developers that are coding VisualBasic instead of ASP.NET or any of th # languages. Surely today's technologies allow much richer WebApps, but surely they will not overhelm all the exisiting applications within 5 years, probably they never will.

    I am not saying the App-V is the better approach, because it is not. It only fixes applications issues that developers could have avoided. For me, it can't be a either/or and therefor you should not refuse to deal with AppVirt. Maybe AppVirt is not so relevant anymore in 5 years, but within the coming few years it will solve lots of your application trouble and I am sure that it is valuable even from a commercial point-of-view if topics like "stability" and "flexibility" are included in you commercial scope.


     
    Falko
    Wednesday, January 06, 2010 5:50 PM
    Moderator

All replies

  • Hello,

    Since this is a pretty general question;

    I do believe they are targeting two different things;
    Web-based applications wish to utilize the constant connectivity possibilities
    Application Virtualization strives for application compability

    Now, even though I know that softricity intended for ISVs to deliver virtualized applications - this has obviously not caught on and thus if I were in a developing applications I would probably not aim for it.

    However, as always I would try to understand what I would like to deliver to the customer - as web-based and app-v are just two different distribution models - where the latter is most commonly used by internal ITs to resolve poorly designed applications distribution.

    /Znack
    Tuesday, January 05, 2010 9:56 PM
  • So here is what I am hearing:

    1. Application Virtualization has the ability to deliver functionality once if the clients are not actually connected. (this is a plus for App-v). With HTML 5 standards coming out, the offline browser app will have a higher ability to function while offline, but lets be honest, that is a ways out. So win for App-v.
    2. App-V strives for application compability, I guess you mean between compability between applications. (this is a plus for App-v). however, isn't this the very reason why we build applications via web browser.
    3. You are saying that developing applications for the App-v infrastructure is something that has not yet really caught on, so you would steer away from dev'ing toward App-V (this is a plus for focusing on web-apps)
    4. App-V and rich web-apps ARE competition to each other.

    I guess I do not hear a viable reason to sink time and $$ into a platform that could be obsolete in 5 years.

    I was thinking that App-V might help in: Faster time to market, rolling out apps might be quicker with App-v technology. Perhaps they are more secure, since locking down port 80 can be tough at times as well as the access that the browsers now have to OS, browser hooks can run deep. Also, I thought that the App-v platform might help us with some old applications that are not currently running on the current databases. It sounds like if I have some old tech, App-v might be the way to go, but we should balance the cost of bringing the old applications up to date with robust databases that can feed a web app, with the quick (I am assuming)time to market that App-V provides.

    Are there any more thoughts on this? We have a meeting tomorrow afternoon. I would really like to get as much insight as possible.

    Thanks again.

    S
    Sean
    Wednesday, January 06, 2010 2:16 AM
  • Hi,

    I asume that it never will be an either/or question.

    Of course nicely written WebApps would not require Application Virtualiuzation approaches any longer, but - shortly - there some big issues that will arise

    - You still have old Applications. It won't be possible to re-code every legacy application within the the timeframe that you specified with "5 years". Look into organizations and count the apps that are were developed before 2005. So, there will be applications that can't be webyfied and there have to be solutions to deploy them

    - It sometimes is said that "devleopers are the (deployment) admin's natural enemy". If developers would act in a way that Microsoft proposed ages ago, we would not need any Application Virtualization solution, because allmost all of the issues solved by App-V come from applications that where not coded using Microsoft's essential development guidelines. So, don't expect WebApps will act "nicely" on the Desktop.

    - A new runtime environment (call it Ajax, AdobeAir, Silverlight or whatever) will cause its own issues. We all remember the "Java applications are platform agnostic and can run whenever a JRE is there". And today. There are more then a few applications that are virtualized BECAUSE they use Java. Bevcause they need a certain version of Java - and they just can't select the "right" version if more then one versions are installed on a machine. So, AppVirt is going to be used to "bundle" these new runtime environments with certain Applications just to allow the Apps to work.

    - The first company I was working for was a Web Company. Well, luckywise they disappearded and I was forced to move to another company that later then lead me to application virtualization. But I perfectly remeber Presentation Bullte Points saying "in a few years, you can do ANYTHING from within a browser". This was about 2001 and still there are Developers that are coding VisualBasic instead of ASP.NET or any of th # languages. Surely today's technologies allow much richer WebApps, but surely they will not overhelm all the exisiting applications within 5 years, probably they never will.

    I am not saying the App-V is the better approach, because it is not. It only fixes applications issues that developers could have avoided. For me, it can't be a either/or and therefor you should not refuse to deal with AppVirt. Maybe AppVirt is not so relevant anymore in 5 years, but within the coming few years it will solve lots of your application trouble and I am sure that it is valuable even from a commercial point-of-view if topics like "stability" and "flexibility" are included in you commercial scope.


     
    Falko
    Wednesday, January 06, 2010 5:50 PM
    Moderator
  • Falko,

    Thanks for the reply. From a legecy perspective, I understand that Appv is the way to go... I just do not see the value in it as a platform moving forward vs going web driven.

    Thanks again for the input.

    S
    Sean
    Thursday, January 07, 2010 6:42 AM