CloudFoundry, VMware And “Evil Plans”

Picture Credit: Gapingvoid.com

In my post about CloudFoundry (previous CloudAve coverage) last week, I talked about concerns some people have on VMware’s intentions and how I think it is a non-issue. I thought I will do a post explaining what these concerns are and outline the reasons that made me feel confident about CloudFoundry’s future. I was bullish on OpenStack during the early days and I am equally bullish on CloudFoundry now. I see both these projects as a necessity to avoid any future consolidation towards monopoly in their respective market segments. Having made the intentions for my support to these projects clear, I will go ahead and explain my thought process on VMware’s role in CloudFoundry.

First, what is the problem here?

After VMware, a company with proprietary roots, announced the open source PaaS project, some pundits and practitioners were skeptical about VMware’s intentions in the long term. They saw CloudFoundry as a hailmary paas, oops, pass to get traction in the cloud computing space. They expected VMware to move CloudFoundry to a proprietary license after the project gets enough developer traction. In short, some people refused to believe that VMware is even serious about their openness mantra. Their skepticism cannot be dismissed as FUD outright. VMware was a proprietary company that refused to play nice in the interoperability game in the past and they did try hard to get into cloud infrastructure space under their own terms and didn’t get the expected traction. All their competitors, including Amazon and Microsoft, had a great story to tell and VMware was in a desperate situation from the cloud standpoint. Yes, CloudFoundry was their hailmary pass but it was a well executed pass. In spite of all this, I still think VMware doesn’t have any “evil intentions” with the CloudFoundry project.

Why not?

Let me highlight some of the points that made me arrive at this conclusion. Even though I am arguing from the standpoint that VMware has no “evil intentions”, my ultimate point of view is “even if they have one, it doesn’t matter”.

• As I did in the past, when similar concerns were raised about Rackspace’s intentions with OpenStack project, I want to first highlight the license part of the story. CloudFoundry comes with Apache 2 license. Even if you don’t trust VMware, you can trust Apache license to protect not just the users of the software but the software itself from any evil plans by not just VMware but also from any of the players in the ecosystem. If it was GPL, I might have some concerns about what VMware can or cannot do. With Apache, VMware has essentially told the world that they can embrace the project with complete confidence.
• In my opinion, VMware’s monetization attempts with CloudFoundry will pale in comparison to what they get in the infrastructure side. Even though I am not being 100% accurate here, I would even claim that VMware has no great plans to monetize big from the CloudFoundry but, rather, they want to use CloudFoundry as a way to stop their enterprise customers from looking outside for their cloud infrastructure and platform needs. At least in the foreseeable future, infrastructure is where VMware plans to monetize. Eventually, they could potentially monetize on the platform side but since they had already commoditized the space by releasing CloudFoundry as open source code, I seriously doubt if they can do it. Their plans to focus on their infrastructure offerings is evident from the fact that CloudFoundry platform doesn’t have the scalability and some of the automation features built in. VMware wants CloudFoundry enterprise customers to rely on VMware’s infrastructure tools for scalability and automation needs. Any monetization from the CloudFoundry side will be minuscule for them and, as I mentioned in this paragraph earlier, they will be more than happy as long as CloudFoundry helps them to retain their existing enterprise infrastructure marketshare.
• With companies like HP, Joyent, ActiveState and Appfog investing their resources into CloudFoundry project, any attempts by VMware to change the direction of the project will backfire big time and the very threat of forking will keep them sincere. Wait for more service providers and enterprises in the OpenStack ecosystem embrace CloudFoundry, the pressures of such large community will make it even more difficult to execute any “evil plans”. On a side note, I hear murmurs about OpenStack community wanting to build their own PaaS layer. I would advise them to instead work with CloudFoundry and integrate deeply with OpenStack platform. Such a cooperation will augur well for both communities and end users.

These are some of the reasons why I strongly feel that CloudFoundry is safe till the next disruption which will completely alter the landscape. I would love to hear from others on this topic.

Related articles

• imabonehead: Man Survives Steve Ballmer’s Flying Chair To Build ’21st Century Linux’ | Wired Enterprise | Wired.com (wired.com)
• VMware’s Cloud Foundry Ranked Top Developer Platform (informationweek.com)
• Joyent is Community Lead for Node.js on Cloud Foundry (joyeur.com)
• Exclusive: HP Runs VMware’s Open Source ‘Cloud OS’ (wired.com)
• Cloud Foundry meets the enterprise with Stackato (gigaom.com)
• VMware’s Cloud Foundry Tops List of Best Cloud Development Platforms (devx.com)
• AppFog launches first Cloud Foundry service with paid plans (appfog.com)