I think Jim Zemlin, executive director of the Linux Foundation (LF), is doing a fantastic job. His ability to reach beyond the vendor community to attract participants like the Chicago Mercantile Exchange (CME) to the LF is useful, important, and game-changing.
His smacking Solaris around, however, is not. In a recent InfoWorld article Zemlin takes on Solaris and argues that it's irrelevant and wasteful, and that Sun should simply join together with the Linux community.
While I'm sympathetic to this view-- Linux plus Sun engineering would be even stronger--I'm not sure this is the way to go about it. I doubt he spent much time browbeating CME into joining the Linux Foundation, or that such a strategy would have worked. The same is true with Sun. Carrots, not sticks, might work, but it will take time.
On the other hand, it's instructive to consider what Sun would lose in joining up with Linux. Sun is master of its own destiny on Solaris. With Linux? Well, let's just say that it's unlikely to displace Red Hat's dominant position anytime soon.
Some might cite Novell and NetWare as an illustrative example for Sun to follow. I don't think Sun's Solaris is yet analogous to NetWare, though of course there's always the chance of Solaris becoming a performance legend rather than a performance leader. In the meantime, Sun is probably right to keep forging ahead with its restructuring plan around open source, one that is at times distracting from industry open-source standards, but one that also keeps the open-source product market competitive.
By joining forces with Linux, then, Sun would not only be losing its Solaris performance strength and control, but also its marketing differentiation. Yes, Linux would likely benefit, but not Sun. Given that Sun sells systems, not operating systems, is it any wonder that it's not champing at the bit to abandon Solaris for Linux?