So XenSource and Citrix got hitched today. A marriage made in heaven?

Well, from one perspective, absolutely. Citrix has a big presence in a certain type of desktop virtualization -- display presentation. And, of course, Citrix and Microsoft have a close, even symbiotic relationship. Finally, XenSource and Microsoft have cooperated closely on Microsoft's upcoming Windows Server 2008 virtualization technology (aka Viridian). So, lots of points for synergy, reinforcing existing commercial relationships, and, of course, Citrix needed a way to respond to VMware's desktop initiatives, as Doug Dineley noted in this InfoWorld blog posting. From this perspective, the acquisition is a tight coupling of complementary technologies and companies.

From anther perspective, though, the marriage, as they say, "has issues," primarily revolving around XenSource's existing business model.

XenSource is the commercial sponsor of the open source Xen virtualization project, using Xen as a foundation for its own XenServer products. Xen the technology has many contributors beyond XenSource itself: Intel, AMD, IBM, Red Hat, and others. Most of the other contributors to Xen view the technology as key to the battles of their various markets: Red Hat against Windows, Intel and AMD against each other (coalitions make for strange partnership arrangements), IBM against HP, and so on. Most of these participants are most interested in Xen in its Linux form, unlike XenSource, which, as noted above, has been frank about its Microsoft focus.

What made (and makes) this coalition possible is the fact that it is based on open source, which seems to be able to convince competitors to collaborate in a common effort.

XenSource itself took the role of (if I may mix a metaphor) of Switzerland, neutrally shepherding Xen. Of course, it differs from Switzerland in one very important respect: it does not hold a strictly neutral role. It actively markets its own Xen-based product called XenServer, which undoubtedly formed a large part of the reason Citrix was interested in the acquisition.

As you might expect, a coalition in which a vital member both shepherds a key component and competes with a number of the coalition members carries some tension, as in any coalition in which members' interests are aligned but not identical. The Citrix acquisition is likely to, if not increase the tension, at least add confusion to the tension, as additional interests and initiatives get put into the mix.

Certainly a critical element in keeping the Xen coalition moving forward will be a deft hand in managing open source relationships, particularly given the complex situation described in the previous paragraph.

An important question is how Citrix will handle XenSource's open source activities. It's one thing to assert a commitment to participating in an open source community; the reality of actually working in the fluid, jumbled world of open source community can be quite challenging. I worked with one company that had the software it wrote for its own products rewritten by the person in charge of the project! Believe me, that kind of thing can take some getting used to, particularly if you're coming from a world in which you control the entire product offering. So I expect a learning curve and a period of adjustment, as there is in any new marriage. This will be especially true in terms of keeping XenSource's (and Xen's) Linux focus going, given that Citrix is primarily Microsoft Windows-oriented.

Overall, I believe this is a plus for Xen virtualization. With the resources of a much larger company behind it, XenSource can put more oomph into its Xen efforts. There are many, many exciting developments to come in virtualization, and the bigger and (hopefully) better XenSource will be able to take advantage of them. Critical to making the acquisition work will be to retain the current XenSource players who are involved in the open source activities of the company and allow/enable them to strongly continue their current work. For sure, though, I think the acquisition answers the perennial question "how do you make money with open source?"