rand($thoughts);

The move away from commodity boxes running Linux continues. It’s nice to see Red Hat getting in front of this market trend. Very cool news about RHEL being offered through Amazon’s EC2.

“The combination of Red Hat Enterprise Linux and Amazon EC2 changes the economics of computing by allowing customers to pay only for the infrastructure software services and capacity that they actually use.

Base prices are $19 per month, per user and $0.21, $0.53 or $0.94 for every compute hour used on Amazon’s EC2 service, depending on whether customers choose a small, large or extra-large compute instance size, plus bandwidth and storage fees.”

It’s in limited private beta testing, but you can ask to join the beta here.

Matt Asay & Larry Dignan question whether Sun can grow. Matt believes that Sun needs a stronger software story in order to grow. Sun grew 1% YTY in the most recent quarter. Yay for open sourcing everything?!?

Sun has great hardware products. Their main revenue generating software successes have come from software that is tightly linked to driving more hardware sales. Solaris and Sun’s storage software are examples of this point. Like Sun, HP has great hardware products. Sure, HP does some software, but at the end of the day, the software (like OpenView) is tightly linked to driving more hardware sales. Sun is more like HP than its willing to accept just yet.

Sun’s revenues are over $3B a quarter (~$14B for the year). Sun would need an additional $300M/quarter of software business to show 10% growth. Even if Sun swallowed Red Hat tomorrow, Red Hat’s ~$500M in revenue would result in 3.4% revenue growth for Sun. One could argue that additional software could drive additional server revenue for Sun. It could, but customers have decoupled their software and server purchase decisions years ago, so there is no guarantee.

The future for Sun isn’t in a stronger software story. Sun has never had a strong software story. While open sourcing everything makes for great PR, OSS hasn’t and won’t drive significant (near or long term) revenue for Sun. Do the math…sorry, it had to be said.

The best advice for Sun is to focus on what they’re good at. Servers and storage.

The platform (dubbed Android) will be Linux & Java based. Don’t expect to see phones that implement the platform until 2H08 (which means late December 2008). The Android platform (software stack) will be licensed under the Apache license 2.0.

A preview of the SDK will be available on Nov. 12, 2008. Surprisingly, the NYT article states:

“A week from today, the alliance plans to make available tools for third-party programmers, called a software developers’ kit, Rubin said. But the group’s core technology itself will not be made available under an open-source license until it is commercially ready sometime next year, Rubin said.”

Compare the latter statement to this statement from the developer section of the Open Handset Alliance website:

“We view Android as a “living” platform and look forward to working with the developer community to continuously enhance and enrich the platform.”

Huh? I’ll post any clarification on when the project truly becomes open for outside contributions (Nov. 12th or sometime in 2008) once I find out.

An interesting question and answer from the Android FAQ:

“Q]If the Open Handset Alliance is giving it all away for free, how will the platform be differentiated?

A] Because the Apache license does not have a copyleft clause, industry players can add proprietary functionality to their products based on Android without needing to contribute anything back to the platform. As the entire platform is open, companies can remove functionality if they choose. Applications are not set in stone, and differentiation is always possible. For example, if you want to include Hotmail instead of Gmail, it will not be an issue”

Using the Apache license is a very pragmatic decision. It’s hard for me to believe that all these vendors would agree to participate if the code was going to be GPL’d.

Tom Krazit at News.com is reporting that Google will formally announce a mobile platform dubbed “Android”.

Tom writes:

“A software development kit for what’s being called “a complete mobile-phone software stack” is believed to be in the works and will be released relatively soon thereafter, the sources said. It’s not exactly clear what kind of software will come as part of that stack, but it’s said to include everything you need to run a phone.”

Google is also expected to announce the formation of the Open Handset Alliance, which includes vendors such as Qualcomm, Intel, Motorola, Intel, Spring NTT DoCoMo, and others.

It’s great to see Google herding cats towards a vision of “open handsets”. The apparent lack of Nokia and RIM in the Open Handset Alliance is troubling. It’s difficult to see any mobile phone alliance succeeding without involvement from Nokia (#1 overall) and RIM (leader in smart phone market & expanding into consumer space). But I guess you shouldn’t bet against Google…especially when Google sees future revenue coming increasingly form mobile scenarios.

A while ago Shaun Connolly at JBoss posted an entry titled “What’s in a Subscription“. Shaun states:

“Put simply, a Subscription is comprised of:
1. Software bits
2. Patches and updates to the bits
3. Support in the use of the bits
4. Legal assurance”

I am beginning to wonder if, in the long term, the only truly scalable business model for OSS is one that incorporates #1 & #2 on Shaun’s list. When you think about it objectively, purchasing a ’subscription’ to get “Software bits” and “Patches and updates to the bits” is the model the software industry has been using for decades. The only difference with OSS is that 99% of your users are doing so without paying for the product.

Heresy, yes I know.

But look at the successful OSS vendors & their products (RHEL, SLES, SugarCRM, JBoss AS, and MySQL Enterprise) in the market today. Isn’t financial success driven by access and updates to (specialized) software bits? Maybe other things were important to these vendors in the past, but at this point in their maturity, the major revenue driver is (simplified) access to (specialized) products baby!

If “Support in the use of the bits” was so critical to customers then a larger % of production users would be paying customers. OSS vendors wouldn’t need to direct customers to a paid subscription as a means by which to get (a specialized version) of #1 and #2.

I’m beginning to think that Support *was* the key method to get customers to the table. Now that customers are more comfortable with OSS, Support by itself is not reason enough to purchase (or renew) a Subscription.

As I discussed in a prior post, things like copyrights & patents are vendor issues and should not become customer issues.

Anywho, my (draft) thesis is that, in the long term, the only truly scalable OSS business model starts with gated access to binaries/code and the patches/upgrades to said binaries/code.

Is the enterprise software business model the future of the OSS business model?

First Sanjiva challenged Dave’s views (repeated by yours truly) that OSS vendors should focus on competing against the enterprise gorilla in the market vs. competing with each other.

Next, BEA decided to make his prior post on “whether BEA is screwed” redundant by trying to paint itself as a Web 2.0 firm.

BTW Dave, the press release is from Oct 8th…dude, that’s over a month ago, and it’s just a press release. Who reads those anyway?! ;-)

Anywho, all the best with your new blog at CNet. We’ll be reading with interest!  It’s been fun working with you at InfoWorld and look forward to bad mouthing you now that you’re gone.