Windows apps to non-Windows devices is still a hole in VMware's end user computing strategy

While business users will need Windows apps for years to come, they very soon won't need a Windows desktop.

On Monday VMware's CTO of end user computing (née "desktop") Scott Davis blogged about VMware's end user computing strategy with regards to the cloud.

Late last year I wrote that Terminal Server would be huge again in a future non-WIndows web app-loving world for the simple reason that while business users will need Windows apps for years to come, they very soon won't need a Windows desktop. The iPad/Android/Mac/Local OS/web/whatever is a fine device manager, and really users only need a simple way to access the increasingly "old school" Windows apps. (Fun related article: Is Windows in danger of becoming the "XP Mode" of the cloud?)

This is where a Terminal Server delivering seamless Windows apps to any client is perfect. Citrix knows that. Microsoft knows that. Quest knows that. It seems like everyone knows that except for VMware.

Of course calls for VMware to enter the Terminal Server space are not new. I blogged about that way back in 2007 (before they bought Thinstall). But while that article was more about VMware competing against Citrix, today's reason for digging up this old argument is centered around VMware's strategy for extending their end user computing into the cloud.

In his blog post, Scott wrote about " the new world of heterogeneous devices, their disparate operating systems and application platforms. It’s clearly no longer a Windows-only world."

He talked about how individual SaaS applications are islands and something like VMware Horizon is needed to stitch them all together (which I agree with -- Horizon is sweet!) "side by side with the universe of Windows Client/Server applications that will be critical for many years to come even as they gradually recede in importance." He also talks about how VMware will be expanding the types of devices they support and gave examples of VMware MVP and how it can support side-by-side instances of Android.

I agree 100% with what Scott is saying in terms of where things are going. But what I don't get is how VMware's products actually tie back to this?

Scott talks about how View and ThinApp deliver Windows and Windows apps as a service. But they both have major limitations.

The problem with VMware View is it delivers the entire Windows desktop, complete with its own task bar, Start button, desktop wallpaper, desktop icons, and Alt-Tab list of apps. If you're using that from a thin client where you're "replacing" your local desktop with the remote desktop, fine. But for your iPad/Android/whatever, I mean come on. That's the Windows desktop as a service. But those kinds of users don't need a desktop -- they already have one! They just need their apps. (And in fact the "second" desktop from View is actually more confusing and less convenient.) Why wouldn't VMware add a solution that just delivers the single applications in a well-integrated way into all these non-Windows desktop clients?

Of course some would say (Scott included) that ThinApp is the "Windows apps as a service" offering. True. Except ThinApp applications require a Windows desktop OS to run. So you can't use ThinApp to just pop an application into your iPad or MVP Android. Yeah, you can offer them to non-Windows clients by combining your ThinApps into a View VDI desktop, but then you're back to the same problem where using that View desktop is an entirely new and separate desktop than your local device's desktop. So until VMware can deliver ThinApps to non-Windows platforms, this is another thing about VMware's end user computing strategy that doesn't make sense? (We tried this four years ago with limited success, although startup Deskstream seems to have gotten it working.)

So where's that leave us? I think for VMware it's more of the same. A lot of talk, but products that don't exactly align to the vision that they're talking about. Unless I'm missing something?

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

Wow. This is a good article. Keep 'em coming!


Great analysis about the VMware's blog entry.

Ever since the existance of VMware, they have been desperately trying to get beneath MS in terms of the platform, then offer superior services and push them out of revelance.

IMO, they entered the arena with VDI because developing/selling a TS solution at that time meant filling MS pockets as well with TS CALs. Soon they realized MS would capitalize on VDI with VECD (VDA) much like they capitalized on TS years earlier.

Now VMware is stuck. They don't want to advance View because of VDA, and they don't want to develop a TS solution because of TS CALs so what do they do? They have to deliver the apps somehow.

Buy SpringSource and declare that Web Apps are the future and Windows Apps are legacy. Don't worry though, they will allow integration of Citrix XenApp for the legacy windows apps. In their eye's they are thankfull that someone (Citrix) will enable access to legacy apps while they are waiting to be transitioned into Web Apps.

But VMware's platform strategy is as monothilic and locked in as MS where we would be just trading one dictator for another.

I agree that Web Apps are the future, but that is a very long ways away, expecially when you take in account the diverse ecosystem of windows apps/developers in todays world.


I also wanted to add the VMware MVP is as flawed as VMware View Client Local Mode and I am glad VMware recognises it.

"This technology is directly analogous to View Client Local Mode"

If you want to know details about MVP, check out this blog entry from back in Nov of last year from OKL. (I am sure they are biased, but they are also very knowledgable and point out a lot of facts)

A few other great blog articles are here too:

Years in the making, OKL will be coming out with " SecureIT Mobile Enterprise" shortly to counter VMware MVP, so this will be a great sight to see.

OKL are doing it right, and they just released the first formally verified OS the world has seen.

Well, at least VMware MVP will come to market hopefully. I prefer the competition rather than seeing another flop like VMware CVP.


Well, how difficult is it for VMware to come up with a something like Xenapp and provide thinapp as a part of that TSE suite, considering they already have PCOIP and ain't it better than ICA ( :P), and I am pretty sure MS would not complain, and I am really not sure if anyone even finds a method to run windows applications without windows, MS will even support it and as with many enterprises as soon as MS says we won't it, unless you are really big to flex the muscle  with MS most enterprises will not invest in that solution.

Today everything you see is VDI, Even with Citrix all they talk about these days is Xendesktop, if delivering apps is all we care about why isn't Citrix focusing on their core strengths and why are they making it feel like eventually Xenapp will just become part of xendesktop as opposed to separate product as it's been sold and used by enterprises for so many years.

So I guess it all really boils down to what MS decides and does, rest all the things we talk about and discuss eventually change and If you want to run windows apps, you have to run on Windows OS.


Should not be as trivial as it looks like... Numberous company have tried to enter the SBC market and failed... Remeber Newmoon, Tarentella...


That's all wishful thiking.  TS and VDI will be confined to internal/private clouds.  The best TS and VDI can wish for in the age of the cloud is improved, Cloud-like management and automation.  

The future of the corporate network is a mishmash of private/public Clouds.  For the most part, TS and VDI will remain inside the corporate data center.  Windows apps are to the Cloud what the DOS prompt is to Windows.


Well there are some companies able to bring windows applications seamless to non windows OS' like <href a="> which uses Java to bring the apps clientless to Mac OS and Linux


Device toys are not replacing Windows desktops anytime soon. They will add new use cases.


VMWare should buy the GraphOn Technology and offer a published applications solution....



I'm glad to see your comment...

GraphOn has a VERY interesting technology. we have not only Windows, but also UNIX and LINUX - all with the same home-grown protocol.

We cook some other things which are even more exciting in the near future.

As Andrew Wood wrote last year on our older model: "GO-Global is unique of all the Presentation Virtualisation solutions discussed in this paper in that it does not require Microsoft Windows Terminal Services/RDS. GraphOn has been to deliver simple access to applications rather than a full desktop."

For the ones that are less familiar - check this out for out current windows solution:

and this one for the overall product line:

Share your thoughts please :)


Completely agree with the article.

The other aspect is that the density of users you can achieve on full desktop virtualisation against 'traditional' application are an order of magnitude (or two!) apart.

Providing choice App Virtualisation plus desktop virtualisation where a business can choose what cost/benefit they want is essential and something all Server Based Computing providers must offer.

Additionally, one commonly mentioned reason for moving to SBC is the green/environmental benefits of this approach. In Europe there was an independent study done by the Fraunhofer Institute that looked at

I also found that it was recently updated to include an Application Virtualisation V's Desktop Virtualisation comparison:

They make for long but very interesting reading. It shows that choice is even more important on how to access virtualised apps or desktops is more important than ever.