Jim Grisanzio

I spent a few days in China with the OpenSolaris community at Sun’s Tech Days Beijing Conference. This conference tour is really growing in diversity these days. NetBeans participation is huge, and OpenSolaris is all over the place as well. I think we are all helping to build on an already successful conference series with the inclusion of these open source projects. NetBeans got involved in last year’s tour, and OpenSolaris is getting involved this year.

In Beijing this week, about a dozen Solaris executives, engineers, and managers from the U.S. and China participated in multiple events at the conference — the Solaris track on day two, the OpenSolaris Day on day three, the Beijing OpenSolaris User Group meeting at Tsinghua University, the new Beijing Solaris Product User Group, and the University World Tour at the China Academy of Science (CAS) Graduate School of Engineering. That’s a lot. But the demand is clearly there. Very impressive.

Interest in Solaris and OpenSolaris is running very high in China. The sessions were well attended, and they all offered solid technical content. Even I participated, though I don’t talk about technology. I presented my OpenSolaris community story twice — once to the students at CAS and once at the OpenSolaris Day at Tech Days. The CAS event went on for hours and hours on Thursday. It was a very hot day, and the room had no air conditioning. Yet a total of 650 students attended, and many were standing and sitting on the floor all over the place. Amazing. The OpenSolaris Day ran all of Friday, and there were about 165 developers or so. At the conference generally, well over a thousand people were exposed to various Solaris and OpenSolaris conversations during three days of events.

I took some shots from as many sessions as I could, but I clearly missed a lot of stuff. There was a professional photographer floating around with a real cannon for a camera, so I hope the Tech Days team posts his pictures someplace. He was shooting some nice stuff. I see Joey has some images, and so does John as well. There will probably be more in the coming weeks.

Thanks to the Tech Days team for having me around. And thanks to the Sun China guys for taking care of me, so I didn’t get cluelessly lost all week with everything going on. It’s so impressive the scale that these guys are dealing with. I’m looking forward to investing a great deal of time in China with you guys.

A call for open community participation. If you are doing interesting things with OpenSolaris and want to get involved with the Tech Days conference tour, let us know. Let me a comment or email me (jim dot grisanzio at sun dot com). We are looking for non-Sun speakers. Check the schedule for a venue near you. OpenSolaris Day is free, so please feel free to stop by.

There are some really nice quotes in this article about OpenSolaris — Google testing Sun’s OpenSolaris, sources say. There is a lot of activity around Solaris, Solaris Express, and OpenSolaris now. And there’s lots of Solaris out there for people to choose from based on what the need is. Very nice. And a very nice article, too. Well balanced.

A note on statistics: There’s a part in this article where the writer compares OpenSolaris to Linux and OpenOffice — the implication being that we’re not as successful in building a developer community around OpenSolaris. Ah. Let’s see. Linux has been open for how long? And OpenOffice how long? And OpenSolaris? Ok. But the larger point is expressed by Stephen Hahn: “Why do all open source communities have to look the same?” Hahn said. “I would rather have 1,000 developers working together than 12 different Solaris distributions.” I agree. And the stats we release are not intended to be compared to anyone. In fact, whenever anyone asks me who I’m comparing OpenSolaris to I always respond that I’m comparing OpenSolaris to OpenSolaris. A friend of mine in PR laughed at me for that, but I’m serious. I wouldn’t even know how to compare OpenSolaris to any other community. Would you? Considering all the variables, I mean. Everyone’s so different. Different licenses. Different business models. Different technology. Different development methodologies. Different leadership frameworks. Different time periods. Different geographies. Now, I look to a variety of other communities to learn about community building, sure, but I could care less about comparing stats and then drawing conclusions. But even more practically, I can hardly even keep up with OpenSolaris let alone try to figure out who has more of this or that. That’s probably the real reason.

I attended the 18th Nihon Sun User Group Symposium in Tokyo yesterday. It was very nice meeting everyone from the community and hearing about their experiences in this market. I have so much to learn, my goodness. It’s very exciting, though. I was also honored to present my OpenSolaris Story to everyone, and I appreciate the live translation, too. I took about 100 images from the all day event.

Japanese auto makers Honda and Toyota have some pretty aggressive plans for expansion in the U.S. They are clearly on a roll at this point. Sales in Japan are expected to be flat, but there seems to be significant opportunity for growth in Asian markets outside Japan.

More good news — Huge oil find in Gulf of Mexico. Potentially, this new stash of oil “could boost US reserves by more than 50%,” according to the article. Cool. Start drilling. Of course, I don’t think there has been a new refinery built in the U.S. in something like 30 years (maybe one, I’m not sure), which is probably just as important as getting our hands on all that extra oil in the Gulf. But I’ll take it. It will be expensive and controversial and will not matter much for the world supply in the long term, but I don’t care. Every little bit helps for the U.S. supply, and this find is potentially much more than a little bit in U.S. terms. Drill it.

Here are the top 20 countries visiting opensolaris.org since we opened on June 14, 2005:

I’m not at all surprised by the distribution of countries, but the percentage spread is still somewhat surprising more than a year later. I would have though that the countries in Eastern Europe, Asia, and South America would make up a bigger percentage of the traffic by now. But, in fact, the traffic from the U.S. and Western Europe has actually increased slightly faster in 2006 when compared to 2005. Looking back on how we opened the project during the first year, I think that perhaps this makes sense. Also, to me it still represents a gigantic opportunity for growth for the OpenSolaris community. Several things are happening now that will probably change these numbers during the last quarter of 2006 and the first half of 2007. Language is absolutely an issue, and we’ll be addressing that as well as engaging more of these non-western regions directly.

It’s easy to track this stuff over time, so we’ll see what happens when the percentage of traffic from the U.S. begins to drop as activity increases in other areas. The U.S. will provide the vast majority of traffic for the near future, but I’d like to see these numbers more evenly distributed from across the world as the community grows and diversifies. Again. It’s pure opportunity for growth out there …

Here are the top 20 countries visiting opensolaris.org since we opened on June 14, 2005:

I’m not at all surprised by the distribution of countries, but the percentage spread is still somewhat surprising more than a year later. I would have though that the countries in Eastern Europe, Asia, and South America would make up a bigger percentage of the traffic by now. But, in fact, the traffic from the U.S. and Western Europe has actually increased slightly faster in 2006 when compared to 2005. Looking back on how we opened the project during the first year, I think that perhaps this makes sense. Also, to me it still represents a gigantic opportunity for growth for the OpenSolaris community. Several things are happening now that will probably change these numbers during the last quarter of 2006 and the first half of 2007. Language is absolutely an issue, and we’ll be addressing that as well as engaging more of these non-western regions directly.

It’s easy to track this stuff over time, so we’ll see what happens when the percentage of traffic from the U.S. begins to drop as activity increases in other areas. The U.S. will provide the vast majority of traffic for the near future, but I’d like to see these numbers more evenly distributed from across the world as the community grows and diversifies. Again. It’s pure opportunity for growth out there …

Bonnie Corwin updated the OpenSolaris Program Team just now with a gorgeous list of everything we’ve opened so far on OpenSolaris. It’s quite a list, I must say, and it demonstrates quite clearly that the opening of the Solaris code has been an ongoing process from even before we officially launched in June of 2005. Very nice. I think the OpenSolaris team opening all this code has a lot to be proud of. For upcoming releases, see the roadmap. To get the code, see the downloads page.

Here’s Bonnie’s list …

• 01/25/05: DTrace Source Code
• 06/14/05: OS/Networking Consolidation Source Code
• 10/28/05: JDS Consolidation Source Code
• 11/10/05: DevPro Consolidation: SCCS/make Binaries
• 11/15/05: OpenGrok Source Browser Source Code
• 11/16/05: ZFS Project integrated into ON build 27
• 01/27/06: Network Storage Consolidation Source Code
• 02/22/06: DevPro Consolidation: libm/libmvec Source Code
• 02/28/06: DevPro Consolidation: libmtsk Binaries
• 03/06/06: Install Consolidation: Packaging Tools Source Code
• 03/29/06: SFW Consolidation Source Code
• 03/31/06: Documentation Consolidation: Source for
• ZFS Administration Guide
• Device Driver Tutorial

• 03/31/06: X Window System Consolidation Source Code
• 05/10/06: Globalization Consolidation: Source for OS/Net Consolidation Message Files
• 05/31/06: Documentation Consolidation: Source for
• Solaris Dynamic Tracing Guide
• System Administration Guide: Solaris Containers — Resource Management and Solaris Zones

• 06/12/06: DevPro Consolidation: medalib Source Code
• 06/26/06: Companion CD Source Code
• 06/30/06: Documentation Consolidation: Source for OpenSolaris Developer’s Reference
• Solaris Containers: Resource Management and Solaris Zones Developer’s Guide

• 07/28/06: Documentation Consolidation: Source for
• Solaris Volume Manager System Administration Guide
• Solaris Express Installation Guide: Basic Installations

• 08/31/06: Documentation Consolidation: Source for
• Solaris Trusted Extensions Installation and Configuration Guide
• Solaris Trusted Extensions Label Administration
• Solaris Trusted Extensions User’s Guide
• Solaris Trusted Extensions Transition Guide
• Solaris Trusted Extensions Developer’s Guide
• Solaris Express Installation Guide: Solaris Flash Archives (Creation and Installation)
• System Administration Guide: Basic Administration
• System Administration Guide: Advanced Administration

• 09/11/06: BrandZ Project integrated into ON build 49

Not bad, eh?

Gas is going down, down, down. Fascinating. This is welcome news to anyone who buys gas for absolutely anything. Unfortunately, the price may be going too low to really encourage serious conservation in the U.S., which is badly needed with gas being so cheap over there now. But it was cheap at $3 a gallon, too. This article mentions that “[f]rom peak to trough, oil has fallen more than $16, the steepest drop since the first Gulf War in 1991. But deep $15 corrections in the autumns of 2004 and 2005 were followed by new peaks within about half a year.” Interesting. But I wonder how much this recent price drop is resulting from the lack of hurricanes slamming into the oil-rich Gulf Coast this summer and fall. I’m surprised the Gulf region is even there at all based on what the forecasting pundants were predicting for this year based on the last couple of years. Last year especially, of course. I guess they were wrong, which is good news for a badly hit region of the country.

Well, the economic gloom and doomers out there were disappointed again. Cool. The consumer price index moderated in August in the United States:

Inflation seems well in check now. But was it ever out of check recently? I doubt it. Regardless, it will be difficult for the Fed to argue that we need an interest rate hike when they meet on Wednesday. But we’ll see. And gas prices are coming down, too. What’s going on? I thought everything was coming apart? The economy may be slowing, true, but that’s far from all the sky-is-falling garbage we’ve been hearing lately about inflation and employment. Housing is still an issue, but that market has been softening for months now. Many expect even more pain — much more pain, actually — to come for housing speculators. And of course certain industries are hurting, like auto, but that’s not really related to the entire U.S. economy so much as it’s related to a few companies getting beat in a rapidly changing market. But overall the economy seems to be experiencing what they call a “soft landing” in this cycle. It will be interesting to see how the market reacts to the upcoming mid-term election in the U.S., an election which promises to be filled with more party extremism than ever.

John Clingan asks a fascinating series of questions in How should a corporation deal with leakers?

Well, I think that companies need realize a few things right at the top:

1. most leaks are pretty meaningless, but the over-reaction to leaks can be a bigger problem;
2. many reporters have sources in most organizations, and there’s no way to stop this;
3. all organizations leak because all organizations are made up of people and people talk.

In other words, get used to it. I find that very few companies accept these three items as reality, though. So, leaks will happen, no question. But there are some very practical steps to reduce the number of leaks and decrease the influence of what’s leaked:

1. open up as much as possible so that leaks become even more meaningless than they already are;
2. stop over-reacting to those leaks because it only creates an internal environment of fear;
3. seriously limit the number of people who have access to business-critical information, so dangerous leaks are less likely to occur; 
4. reduce the amount of confidential information so what is confidential is easier to manage and better respected.

In other words, break the monopoly and exclusivity over substantially more of the information in the company. That will help for sure. For truly business-critical information, though, that’s a serious matter and that information is best kept confidential. I have no problem recommending a totally closed approach to that part. For example, although I work pretty much in the open these days, there’s no need for me to see the financial of the company until the announcement. An extreme example, I realize, but I’m simply arguing that companies keep a much smaller amount of truly sensitive information bouncing around their networks, so that classified information is easier to manage and less likely to leak.

John mentions launches, too. I believe “launches” are hopelessly broken in the technology business. I’ve done many, and I can’t think of a way to fix them, so I try to avoid them if possible. I also don’t do them very well, so maybe that’s my problem. But fortunately, launches don’t have the power they once had. I view them as parties, basically. They can be great fun if you don’t have to plan them, but they have little significance in the long run. Are customers really making multi-million dollar buying decisions based on a company’s well synchronized product launch that didn’t leak? I’ve never seen any data suggesting that’s the case. Are launches the only way to bring new products to market? I doubt it. Not anymore, anyway. However, launches are a great source for leaks because literally hundreds of people work on them for months, and the level of activity reaches insane craziness just before launch day. That’s right around the time when many companies go out and brief press and analysts under non-disclosure, so now there is information inside and outside the firewall all timed to go off the minute you “launch” whatever it is you are launching. This make no sense to me. And it’s a very leaky process, too. So, John asks, when something does leak, what does the company do? Well, if the leaker was obvious and the intent was malicious, sure, punish the employee. I don’t have a problem with that. But internal investigations worry me because many times leaks are inadvertent, and it’s easy for well-meaning and talented people can get caught up in something unfortunate. Mistakes do happen in this game. So, in that case, I don’t think a company should go after the employee. In that case, it’s just a part of doing business. It has to be an obvious case, and the leak has to be at the level of affecting the entire business. That’s the key. Product launches are simply not big enough in my opinion. I’ve never seen a product launch that was harmed by a leak, and I can assure you many, many, many of the products I worked on have been leaked.

Again, the concept here is to reduce the classified information — substantially — and isolate it from the majority of the company. At the same time, open as much of the operations as possible so the majority of your people are working in the open with information distributed pervasively across the organization so everything doesn’t have to squeeze through one tiny little funnel at one end of some process somewhere. Now, many levels of corporations will hate this opening suggestion because many people trade in the business of gatekeeping what they think is super-secret information. And some in the press may also not like this because much of their business is based on gaining access to the highest value information possible — which is classified, of course, and which has to come from leakers. What I’m suggesting undermines the leaking process on both sides.

John also asks, “If a reporter knows the information is being unethically disclosed, is it ethical for the reporter to print it?” Well, I’m not a reporter, but I doubt that this is something for the reporter to judge. I don’t see ethics as being a consideration in this case. If reporters get information and it’s good information, the vast majority of them will go with it. Ethics are not involved, other then to keep the source confidential, of course, and most reporters are well-meaning people and honor this and they need to cultivate sources. But in many cases, you can tell where the leaks come from based on the information being leaked.

There are no easy answers to this. But an open, community-based approach to your operations is the best solution I can think of. Also, I think Sun could be a pretty good model in all this. I think we are doing an interesting job in this area. Many parts of the company are very open these days, and other parts are in the process of opening. Actually, I can’t think of a company that has as many people working in the open as Sun. Can you? And have we had any serious “leaks” around here lately? Not really. And if so, what has the harm been? There’s so much information about Sun and it’s products and strategies out there that it’s hard to tell if a leak has occured because the level of conversation around the company is pretty massive at this point. Back to my point that most leaks are meaningless.

So, it may be anti-intuitive, but I think a culture of openness may be a good way to transcend leaks, while at the same time respecting and protecting the company’s true intellectual property: it’s classified information.

Some buzz out there on The Citizendium Project, which means potentially some competition for Wikipedia. The community model planned for Citizendium will be different from what Wikipedia uses, so it will be really interesting to see how it goes. I don’t think anyone has implemented the definitive community model for the generation of content, so new attempts should be welcomed, right?

Nice conversation here on the biology of the female brain. All science, no idiotic gender politics, and utterly fascinating. Not that science can’t be political at times, of course, but there’s a distinct lack of dumbness in this interview. Give it a listen.

I’m particularly interested in the parts about exposing young girls to math and science at an early age and then encouraging them over time to excel in those subjects (or any subject, really). They do quite well, as you’d expect. I’m also interested in how the brain changes — quite literally — based on all of our experiences, and the distinction between male and female brains due to hormone exposure before we are born.

I have a baby girl, so I’m getting more and more interested in her brain. She’s doing some remarkable things. Learning is taking place at a faster and faster rate. It seems automatic now. Sure, we’re teaching her stuff, but I have the distinct impression that she’s learning much more than we are teaching. She’s easily recognizing words in Japanese and English at this point, and she’s starting to respond to entire sentences. She’s making connections that clearly show awareness and thinking. What the heck could she be thinking about? She dances, which is a real kick. How does she know to dance? And what’s up with the books? What could she possibly be doing with all the damn books she has around this place? I don’t get it. But she loves books. Japanese books. English books. Picture books. Even the dictionary. She sits there and turns the pages or insists that you turn the pages for her while she watches. She helps us put the futons away and picks up the pillows and blankets. Strong little kid. And she knows where everything goes. The funniest thing now is that she takes her shoes off before coming into the house. That’s the one that got me this week. I take her out for a walk each night when I get home, and when we come back she stops cold and absolutely will not enter the apartment without attempting to rip off her shoes. Just like us. Though we’re not so animated about it, and we don’t cause big scene in the process. We didn’t teach her the shoes bit. She just picked it up, I guess. She’s like 15 months or something now, and I feel like I’m being watched.

So, yes, that little brain inside her head is getting interesting indeed …

Thanks to Chris for finally getting Jarod Jensen to blog. This is good news for anyone interested in DTrace. See Jarod’s stuff at DTrace Blog: The Texas Ranger. I collect Jarod links here if you are interested in more background. He’s one of those must-meet guys, trust me.

Since I put so much of the bad news on my blog over the last three years, I’m rather enjoying reading some good press for a change. Aren’t you?

Here’s a nice piece in InfoWorld — Sun is winning in the server market. And here’s my favorite quote: “Sun took the bold step of open sourcing its crown jewel, Solaris, to take the ‘proprietary’ millstone from around its neck. Now Solaris is the only one of the Big Three Unixes that is open source.”

The article also talks briefly about some of the challenges we’ve gone through recently and also touches on OpenSPARC and Opteron. I’ve never worked at a company that has re-made itself so dramatically in just a few years as we have done. Actually, I’ve never worked for a company for more than a few years, period, so this is all very, very new to me.