My thesis is that Taiwan must shift to embedded software development and OpenSolaris provides the perfect opportunity.

First time, I blog bi-lingually.

1. Focused applications: It takes years, no matter how determined and well-funded an organization is, to develop a mature, enterprise-grade application software. China should allocate its resources smartly by focusing on what the country needs most. An industry forum, led by the government, that manages the resources for optimal productivity will speed up the software maturity.
2. Standardize on APIs: Everything becomes obsolete faster than predicted. To protect the infra-structure, system components must adhere to standards that are mature, secure, and stable. If so, these components can be replaced with ease. The whole system can ride the waves of innovation without experiencing disruptions -- getting better all the time.
   China government should establish component-level standards, such as system APIs, and encourage adherence.
3. SW/HW Integration at platform level: We know China has a great foundation for computer platforms and its components. But the country is missing a big opportunity. If these platform and component industries can be more integrated with system software or middleware, their effectiveness, performance, features, and cost structure can be greatly enhanced.

No, I did not download Solaris source. (I am in an airport. Wasting 10 minutes or more of the purchased bandwidth does not seem right.) I browsed with the wicked fast tool.

Man. Am I proud of being part of this organization.

I never materially contributed to OpenSolaris. But I sat next to those people who made it happen. Did I say I am so proud of you guys? Yes, yes. I am getting old. Repeating is now a habit.

What did I browse? I search for PAM module. I explore the AMD hat layer. And for fun, I look for Tim's login name. After 30 minutes, I realized that I will never be bored again.

Can someone make me an engineer again? It would be so much more exciting. Or, sorry, Claire. Fanatically devoted (cult-like).

Technorati Tag:OpenSolaris
Technorati Tag:Solaris

WTO membership is a national goal. When this country sets a national goal, all 1.2 billions people march toward it. Like the rest of the world, the predominant desktop for government and industries is Windows/Office. Unlike the rest of the world, the number of licenses and the number of installations do not quite match. This gigantic gap does not jive with WTO.

What to do? "Legitimize all those unlicensed installations," says Microsoft.

But China has options. With the amount of money they must pay to legitimize and the talents they have. They can develop their own solutions, replacing Microsoft. After all, this is a country that 1.2 billion people will really march toward the same goal set by the government. Also, since this is about a quarter of world population, they don't really care for the existing market share. Whatever current market share, China's decision will change that. However, no amount of money and talents can cheat time. Large software developments take years, if not decades. China needs something faster. Enter open source solutions.

GPL gets in the way. Even with strong government support, companies must be able to make money. GPL requires you to donate the fruits of your labor back to the community. For years, China lives with the limitation of GPL as the only option. CDDL now gives innovators choices: donate if you wish, profiting from it will be just fine too. CDDL makes software truly free. China now has a true choice to rival Microsoft. Sweet.

Cheaper, Faster, Better, More Secure

Linux has the installation base and the religion factor. Switching costs are real and religion discussions are non-engineering.

• Solaris is cheaper than Linux. Unlike the real thing, commercially distributed Linuxes are quite expensive. The commercially distributed Solaris costs exactly zero. Can't go lower than that.
• Solaris is faster. Just bring on the benchmarks. We beat Linux.
• Solaris is better. Solaris can run years without rebooting. Solaris shines brighter with the latest CPU technologies - CMT, 64-bit, dual-core. Linux does not offer DTrace, Container, Least Privilege, FMA, SMF, and another thousand advanced functionalities.
• Solaris is secure. It is so much harder to hack Solaris that most hackers do not bother. You can conduct business with a peace of mind if the OS is Solaris. After all, IRS runs Solaris. DoD runs Solaris. Wall Street runs Solaris.

Everyone points out the faces they know. "Josie is here, Darrin is there, look at Chris." What I see is a smiling group that are comfortable with themselves. I can feel the room where the picture was shot. It is hard to remember the rain ouside that day.

I won't unwrap this box. It will be kept on my shelf wherever I go.

Never have I thought a simple command can stir up so much passion. The more I dig, the less hopeful I am that any change will be acceptable.

We are our worst enemy

First is the revered taxonomy document. There shall be only 3 components in a release number: major, minor, and micro. All must be numbers whose numerical values can be compared. The value also must be ever increasing, but not necessarily sequential.

Then the fact that Solaris has not used the "micro" component for about 10 years now. The (re)introduction of this concept is to be taken seriously and extremely cautiously. People have written scripts and programs that assume no such component. Once present, these scripts and programs will cease working and hell will break loose.

Philosophically, what exactly is a "micro" release? This thing that we call "updates" are not -- nor are Solaris Expresses. Is it right that their uname -r yield exactly the previous mini-release number?

Application compatibility? Commitments made to customers (implicitly or explicitly)? Tradition? What behavior are we trying to encourage? Rules and policies?

Sigh, again.