Solaris, The Road Ahead
By marchamilton on Apr 07, 2007
First of all, I'm going to listen, listen, and listen some more. I'll listen to my new staff, many of them have been doing Solaris marketing for years and know a lot more about it than I do. I'll also be spending a lot of time listening to Solaris engineering and sales staff. I'll then be listening to my old staff in Sun's systems practice working on HPC, webtier, and datacenter solutions. No, I have not forget what they are doing already, but I'll be listening to what they need from Solaris with with a keen new ear. Finally, and most importantly, I'll be listening to Solaris and OpenSolaris users, via the many OpenSolaris discussion groups, blogs, and other forums. While you can reach many more people online, I'll also continue to go out and meet customers face to face, and my travel schedule this month includes visiting Kisti, Korea's national HPC center to meet with several hundred top Korean researchers as well as going to the China Education and Research Conference hosted by the Chinese Ministry of Education and Sun where I will have a chance to meet with over 500 of the leading educators in China. One of the things I learned from Ed Zander, before he was president of Sun, when he was still running our software organization, was never to stop listening.
Of course I do get paid to do more than listen, so next I'll be thinking about how to evolve the Solaris strategy. At the highest level, the Solaris strategy was set long ago: develop the best OS on the planet and open source it so any developer in the world can help make it better. But the devil is always in the details. I'm sure there are a few people at Sun and a few Solaris users that are not sure what the difference is between Solaris Express Developer Edition, Solaris 10, and OpenSolaris, not to mention the many other OpenSolaris distributions such as BeleniX, marTux, NexentaOS, and SchilliX. It should be clear to even casual readers of my blog that Sun is focused on making the enhancements to Solaris necessary to make it even better for horizontally scaled HPC and webtier users. As Ian will be quick to point out, this isn't just about the kernel, it is about the whole environment and ease of use, things like making Rocks available on Solaris. As the new guy, I have the opportunity to come in and look at all that we are doing with Solaris and OpenSolaris, and work with the team to set clear priorities for moving forward. As the saying goes, no sacred cows.
Speaking of Rocks, did you know the Rocks team even has their own show on iTunes, just search for "RocksOn" in iTunes. Beyond tools like Rocks, do we need to get iTunes and Skype ported to OpenSolaris? Or do we need to get Ed Zander to use OpenSolaris on his next phone to compete with the iPhone? The Solaris OEM agenda is one that is on my plate and anyone who makes anything that connects to the internet should be thinking about what their open source OS strategy is. The Linux crowd has been for years. Even though I'll be leading up Solaris marketing, I'll work with Ian to make sure our strategy is not anti-Linux. We have a lot to learn from Linux and if Apple can take some of the best features of OpenSolaris like dtrace and include it in MacOS then I am pretty sure there are some Linux developers who have the skills to do the same. Of course if you are a Linux developer today and don't want to wait for dtrace to come to your favorite distro, you can try this. You don't even need a PhD.
So stay tuned, I'll be talking a lot more here about Sun's Solaris strategy in the months ahead. Of course the best strategy is of no good if you don't execute against it. Jeff Jackson assures me he has the right team to execute on the priorities he gets from marketing, and I'll be making sure marketing executes on their priorities. Some of my new staff is gasping right now and thinking, oh no, not a micro-manager. Don't worry, everyone has until 8:32 am Monday to get their first status report in :). Solaris is Sun's biggest software product. Companies much larger than Sun that spend much more on OS development (oh wait, that would narrow it down to one other company) sometimes miss a beat in execution, what makes me think Sun won't do the same? Well, for starters, with OpenSolaris we have way more developers working on our OS than any other company, and we actually use Solaris to develop Solaris. The last OS I developed was when I was an undergrad at UCLA, but come on, who wouldn't want to have dtrace to use to help debug their code?
Finally, the people on my team have always been one of my top priorities. You are only as good as the people who work for you, so you had better work to develop your people. There have been countless people over the years who have been promoted out of the various organizations I have led. A good number of them ended up coming back to work for me again later in different roles, sometimes repeating the pattern two or three times. There are so many exciting and rewarding opportunities at Sun, not all of them in the Solaris organization.
So there you have it, my 90 day plan. Listen, develop the strategy, execute on the strategy, and take care of your people. Solaris has a very exciting road ahead.