Is it time to fork Xen?
An interesting article by Matt Asay with a rebuttal in the comment section by Simon Crosby, Xensource’s CTO.
Red Hat and some of the other distro’s should (continue to) fund the developers that work on (a forked version of) Xen. Or perhaps focus on KVM. It is obvious that Microsoft has no interest whatsoever in keeping the development of Xen on Linux going. Why fund the competition? Simon’s pitch about the religion thing is creative. This is not about religion and not about what customers want. It is about *how* customers *will get* what they want.
Applications are already (e.g. SugarCRM, Google) or soon will be happily running on cheap LAMP setups (Linux/Apache/MySQL/PHP or variations thereof). Nicely humming away in a virtualized setup. This is a threat to Microsoft. Suddenly it’s all about the applications and no longer about the underlying Operating System. Simply because it just requires a browser. No more, no less. Have you ever heard anyone say: “wow I was really glad that I ran Vista when I logged into Gmail to do my email.”?
On the Desktop side Microsoft’s business model is showing its age. And on the server side cracks are starting to show as well. Xen and KVM on Linux offer a great opportunity for businesses. If employees only need a browser to access business applications why would they buy Vista and the rest of the Microsoft menu for top dollar if the IT dept. can deploy thin clients and virtualized Linux servers offering those applications to the browser via a LAMP setup. Even if employees need a fatter client than the IT dept. can deploy a flexible cost-efficient desktop solution from Red Hat or any of the other vendors out there.
Now what about Groupware functionality (Outlook & Exchange)? There is no denying that this is a tough nut to crack. Scalix was bought by Xandros so only time will tell if Scalix will (is allowed to?) continue to Open Source its Exchange compatible Groupware solution. If not then there are several alternatives out there like Open-Xchange. If necessary I would assume that major players like Red Hat will step up and fund the development of an alternative. As a final resort one can always install 2003 with Exchange and OWA on top of a Linux/Xen/KVM setup until an alternative is available. Hey let’s be practical and not religious ok :)
So all this talk about the engine is irrelevant from a market point of view. It’s about how business will access their information. Obviously Microsoft wants the underlying technology to come from Redmond. Others prefer that not to be the case. Basic rules of competition make it very hard to accept that Microsoft through Citrix or Citrix itself will continue to fund Xen’s development on Linux while Xen and Citrix are betting on and siding with Microsoft. Time will tell.
Update: Matt Asay has written a mea culpa reaction to Simon Crosby’s comments titled Open-sourcing my error on XenSource.