VMware's application vision is ahead of its time...which might be perfect

Brian's article on VMware's Project Horizon got turned over quite a bit in the comments. Some people, Brian included, think it's awesome.

Brian's article on VMware's Project Horizon got turned over quite a bit in the comments. Some people, Brian included, think it's awesome. Others just yawn. For me, right now in September 2010, I think that Horizon is more of a necessity than some whiz-bang technology that's supposed to knock your socks off. If you're going to be in the business of application delivery (not just desktops), you have to be able to manage and provide access to those applications. Now, if VMware is right about the vision that was presented at VMworld, I think Horizon will be huge.

VMware is about apps, just not Windows apps. This was made pretty clear during the opening keynote. There was so much focus on any app, any device, anywhere that I half expected the Moody Blues to start playing while Mark Templeton charged the stage WWF wrestler-style. Providing end users access to applications is important, for sure, but that message is only new coming from VMware, not the rest of the industry.

What's different about VMware's approach is that Windows Apps are barely in the picture. The slide below, shown during the opening keynote at VMworld, depicts a Windows app and simply describes it as an "Existing App," as opposed to New Enterprise Applications and SaaS Applications. These two types of applications make up what VMware CEO Paul Maritz called the "New Stack", and deal not only with how applications are deployed, but where they come from to begin with. VMware's put a lot of work into apps in the last few years, acquiring SpringSource and Zimbra (not to mention the work they've put into Origami/Horizon). It was summed up in this slide, after which Maritz said, "One thing history teaches us is that when you have a wholesale shift of stacks, there are winners and there are losers.":

It's right about this time that I tweeted "Maritz isn't just saying Windows APPs are dead. I think he's hinting that WINDOWS is dead!" to which I received a few replies about how Windows is everywhere and it will never go away. I want to go on record saying that I agree that Windows won't go away soon (or quietly), but applications are changing, like it or not. The fact that you can use them without Windows doesn't mean Windows is dead or dying, it just means that it's not the only app platform out there anymore. I didn't throw out my Windows computer because I got an iPad. There's things I can do better on Windows than I can on the iPad. Same for Mac vs. Windows (Hell, I run Parallels AND Fusion. Call me crazy). So fear not, Windows people! As long as there are Windows apps, there will be Windows!

After a few sleepless nights trying to figure out how the vision laid out at VMworld fits with our world of desktop virtualization and application delivery (not to mention why the hell we still don't have a client hypervisor product from them), I finally settled on an explanation: It's ahead of its time.

How much ahead remains to be seen, but I don't see a mass migration to this Hybrid Cloud, New Application Platform, New Infrastructure (I think all that means is shoving a VMware layer under your existing systems) in the next year or two. However, if you look two, three, or four years down the road, after the technology has congealed a bit and the market decides which way it wants to go, I think VMware will be in position to say "we've been doing this for years, and we can get you there."

Is it a gamble? I think so. If the enterprises resist and push that time frame out to 5-7 years, that could be tough on VMware financially. They could find themselves sitting on top of a really kickass solution with nobody to sell it to. Still, they payoff is potentially enormous. If the timing is right, and if Windows really does turn into the XP mode of the future, VMware is in a really good spot with the app framework, infrastructure, and products to make it all happen.

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.


not new... Citrix add that (with Dazzle and NetScaler), Siteminder had that, WebShpere, BEA...

reality : cloud is still vaporous for a long time, windows is still there for a long time (think about TN3270 age!), saas is already there (think abut salesforce.com)...

the only good point is VMWare have now a "user point of view" and is going to get high in the stack... not sure they have the right DNA!


@Gabe I think you have to ask how do they intend to make money? Let me explain. Today VMW makes money by selling ESX and locking you in. That profit center is slowly being eroded by others, with MS the major looming threat in the enterprise and Citrix in the Cloud with buddies like Amazon etc. I could bucket RedHat with Citrix also, but they are far behind according to  the cloud vendors that I speak with. In the future VMW will have to move up the stack. They will try to make money out of management and automation. There is nothing wrong with that IMO, however there problem is, it's too locked in. It's management and automation of their stack, their vClould, on their hypervisor. If they control that they will also try to lock you in with their mgmt stack and try to get app developers to write apps that lock you in further. Hmm sounds a bit like a MS strategy, where did Paul come from.....

So I am sure Horizon will claim all sorts of things, but the question will be where is the money, for VMW it won't be in an open stack which is where VMW will struggle more and more in the coming years. If they were smart they would build a more open mgmt stack so people don't feel locked into their stack only.

I'd agree with Kata, that a lot of this is not new. All VMW have done is painted a picture of a big computer some day in the cloud and that mgmt and automation will change and oh, yeah Windows is not that relevant moving forward etc. It's all vision with no product or execution. I am getting tired with all this Horizon gaga and the lack of focus on what is real for the next few years. I don't think Cisco was too excited by Horizon.......

Let's not assume Windows will not evolve to become more SaaS like. Let's not assume MS will not get their developer ARMY to change to embrace their evolving windows platform. Let's not assume that MS will not have their own Horizon type thing that is the evolution of Systems Center. Let's also not assume that MS will execute fast, so that will leave plenty of room for others with assets to solve this problem to do a lot and who already are doing user based things and stuff at the network layer.

So it is before it's time. SaaS for sure will become more important, but not as fast as people think in the real world, just too much legacy stuff that will also evolve. The reality is that desktop vendors need to solve many of today's problems and not simply paint pictures of vision that are delusion until they proove themselves with real execution.


Are they listening to customer feedback and demand when developing Horizon? Much like when they were listening to the demand of type-2 hypervisors vs. type-1?

I think this was a strategic position to lead where Citrix/MS don't. Year after year proved that the desktop was a lost cause for them.


I can't even get developers to move away from Oracle 6, PowerBuilder 9, .NET 1.x and IE6.  Windows dead in the near future??? Ummm yeah good luck with that.