Open-sourcing my error on XenSource
What can I say? Even the most opinionated among us are sometimes wrong. I just happen to be wrong more than most. :-)
The unfortunate thing about writing all your thoughts down in a blog is that it makes it very clear just how wrong I can be sometimes. My "code" is online, for everyone to see, analyze, and critique.
And critique you do. :-)
A case in point isand its alleged abandonment of the Xen project. John Vigeant, a friend from my Novell days and XenSource's director of Business Development, kindly swatted me in an e-mail for errors in my post.
Witness my sackcloth and ashes (with John's permission--he must have some perverse pleasure in seeing me don this hairshirt :-):
John pointed out that at least one reason for XenSource's strong emphasis on Windows is an attempt to steer clear of the business models of the other vendors that deliver Xen to market. Namely, the Linux distributors. As John noted, although lots of XenSource customers and users mix Windows and Linux on their platform, the use case that is solely Linux-focused is more suited to (say) Red Hat Enterprise Linux (RHEL) on RHEL or even RHEL on SUSE Linux Enterprise Server (SLES).
In short, since the distributions themselves are trying to develop businesses around Xen, by emphasising a platform model--Switzerland of operating systems, if you will--XenSource positions itself against VMware, serves the VMware customer base, and doesn't go head to head with the distributions. This makes a lot of sense to me. Also, since XenSource's view is that very few Windows users will want to install RHEL just to virtualize Windows, it makes sense to have a completely different use model: one that is modeled on VMWare ESX.
John also made the point that forking Xen is not the reason it is separating oversight for the Xen project from XenSource the company. Although there are major and valued contributors in the community, it's still the case that XenSource contributes a majority of the Xen code. XenSource remains committed to the advanced development and transparency of the Xen development process, John noted, and given the reasonableness of what he told me, I admit defeat. I believe him.
I dislike being wrong, but I dislike perpetuating my errors even more. I appreciate John's candid, constructive response, and apologize for misreading the company's intentions. I guess all this means that I believe that XenSource is sincere when it states that doesn't try to game the community, since there is no competitive advantage to be had from that, anyway.