20050126 Wednesday January 26, 2005

Open Solaris, my hat is off to you!

My hat is off to you!

In celebration of Sun's milestone announcement of Open Solaris, a dozen folks got together at a local watering hole in Mountain View, CA, close to the heart of Silicon Valley. Several of the Open Solaris pilot members, such as Ben Rockwood, Bob Palowoda, Dave Lampe, Eric Bruno, John Martinez, and Lloyd Staley were in attendance to hoist some cold ones over a job well done by the Open Solaris group inside Sun. Several Sun folks were in attendance as well, such as Chris Baker who was kind enough to bring us hats and DVD/CD cases with Solaris logos on them. We also celebrated that Ben Rockwood has built the Open Solaris sources that have been released to pilot members. Don't tell anyone, but Lloyd Staley and others have built the sources also![wink]

So, let's take a step back to understand why I call this a milestone announcement. It wasn't too long ago that Sun had announced that they would indefinitely delay Solaris 9 on x86. What Sun didn't understand is that there was a community of Solaris on x86 users that were just not ready to let go of the product. Between Jan. 8, 2002 and Feb. 11, 2002, this community would put together a group of representatives to meet with Sun Executives responsible for Solaris at the time. They would find out that people were not only passionate about this product, but would fight dearly to bring it back.

I am not clear why it took a pack of passionate community representatives to get Sun to understand they had a product worth keeping, rather than throwing away, as there were many Sun Engineers that were already pretty passionate about Solaris on x86 already. In fact, while most of the credit is all too often given to the 6 community representatives, much of the credit must go to the Sun Engineers that continued to build and fix bugs for Solaris on x86 after being told they must stop development on it.

Part of the community presentation was that the community could take over the x86 development if Sun would open source the x86 pieces. This was without having much knowledge about the sources, or even understanding how closely tied together the SPARC and x86 sources were (they are one and the same;-). So, after that Feb. 11 meeting I had a phone conversation with my current manager, Neal Pollack, at Sun (yes, I was hired by Sun during 2003) and I'll be honest with you...After I got off the phone with him, I was pretty convinced that it was near impossible for the community to do the proper work for Solaris on x86, let alone fix some of the problems with it. But I decided that we should just keep negotiating with Sun.

As luck would have it, after I joined Sun to work in Solaris x86 Engineering, Neal Pollack, long time advocate of Solaris on x86 would not only take over the management of the core components for x86, but start to have his group make the needed changes that would be key for Solaris on x86 to survive. And at the same time, Solaris was being ported to support the AMD64 line of processors, including not just Opteron and Athlon64 processors, but Intel's EM64T processors that added 64-bit extentions to the 32-bit Xeon processor family. In an open world, there are no prisoners!;-)

There's a lot of skepticism floating around the internet on Sun's move to open up the sources for Solaris. Much of it is coming from folks that laughed and said it would never happen, and that Sun would never be able to free themself of the copyrights within Solaris. And while it not complete yet, the progress made in this effort is really quite amazing.

So from my perspective this is a true milestone. Sun is not only fully committed to Solaris on both x86/AMD64 and SPARC platforms, but I believe Sun has completely turned around from where they were back on Jan. 8th, 2002. This is a new Sun, shining on a new day! Don't listen to Sun's competition who appears to really be grasping for straws. If you believe that propoganda, have I got a bridge to sell you! Please form your own opinion. And while you're at it, have a gander at what our competition is running on their web site. They can always join the Open Solaris pilot if they need an industrial web host.;-)

At our little celebration at The Tied House in Mountain View, we all received a Solaris hat. In honor of all the hard work the Open Solaris team has done so far, my hat is off to you!;-)

( Jan 26 2005, 02:44:14 AM PST ) Permalink Comments [5]

Trackback URL: http://blogs.sun.com/aland/entry/open_solaris_my_hat_is
Comments:

That's a bit unfair finding a random server on their domain running windows when they use HP-UX pretty much across the board. IBM could probably start poking fun at OpenSolaris.org [http://uptime.netcraft.com/up/graph/?host=opensolaris.org] too

Posted by Ché Kristo on January 26, 2005 at 05:20 AM PST #

Che,

You might find it unfair, but what I find unfair is the propoganda that HP is spreading about Solaris on their website. It's not my fault their page that has that propoganda is running Windows 2000, I have no idea what they use on their site across the board. The page I was pointed to by a community member looks to be running Windows 2000, go figure...

Posted by Alan DuBoff on January 26, 2005 at 05:33 AM PST #

Alan I have read in these blogs about your great efforts behind Solaris x86 and the open sourcing of Solaris. Hats off! I had seen the HP article a while back. I wonder if a rebuttal is in order. I mean, it makes several points. So a) are those points valid and/or worth consideration? and b) if they have some merit, how are we addressing them? For a company whose open source stance has long been suspect in the world at large, where almost nothing we do can be good enough or is only playing catch-up with the rest who 'adopted' Linux and the open source community 'before' us, let us prove them wrong by stating just the facts. So someone in the know about Solaris x86 as you can dispel all myths... Thanks

Posted by Umang on January 26, 2005 at 09:03 AM PST #

Umang,

I'll give it a shot, but give me a couple days.

Posted by Alan DuBoff on January 26, 2005 at 10:43 AM PST #

Che, I'm agnostic about the fairness of this. I just wanted to make sure there was no confusion on one point: I can tell you with 100% certainty that the www.opensolaris.org site is running Solaris. As you point out, Netcraft is often wrong; a different view of opensolaris.org can be seen here: http://uptime.netcraft.com/up/graph?site=http://www.opensolaris.org&probe=1 The 'probe=1' seems to cause a "reprobe." I do believe the linux sighting is probably an artifact of the redirector we have sitting in front of the site (spreading traffic across two Solaris boxes).

Posted by Dan Price on January 26, 2005 at 10:35 PM PST #

Post a Comment:

Name:
E-Mail:
URL:

Your Comment:

HTML Syntax: NOT allowed