Phase 3 of the Sun Model

Liberty Staircase

I wrote recently about the Sun Model for open source business, my high-level overview of how Sun is working with open source.

To summarise:

  1. remove barriers to software adoption between download and deploy;
  2. encourage a large and cohesive community of software deployers;
  3. deliver, for a fee, the means to create value between deploy and scale, for those who need it.

I've had a number of comments and questions about that third phase. It can include all kinds of value-creation, depending on the product in question. Here are some examples of delivering value for people who have already deployed and are heading towards scale:

  • For Solaris and OpenSolaris, Sun offers subscriptions that include the updates, support and warrantly that allows deployers to get the maximum up-time and performance for the minimum cost. You can get the same results yourself by hiring experts to do the work for you, but the Sun subscriptions save money and time.
  • For MySQL,there is the same sort of deal with the addition of software features needed only by those between deploy and scale, such as MySQL Enterprise Monitor.
  • For Glassfish, again, there is a subscription offering that's perfect for those who have taken the decision to deploy and now want the greatest value with the least fuss.
  • ... and so on, across the portfolio.
Devlievering value can take many forms, and nothing is absolutely forbidden unless is creates a barrier between download and deployment in any way.

...and hardware too

But it would be a mistake to believe Sun's open source strategy is only about software. As has been frequently explained, Sun is a systems company, and the news last week and today underlines that fact by showing two new ways Sun is offering value for those between deploy and scale:

  • Systems for MySQL

    Recently, the first database servers optimised for MySQL were made available. For MySQL users who have moved beyond initial deployment and are now looking for high performance servers with rock solid support at great price points, these are excellent. They are optional, but I'd wager most people will save money and create more value by graduating to them for some applications.

  • Unified Storage

    Today's huge news is the release of the new Sun Storage 7000 Series. These new storage appliances create value by combining open source software with commodity hardware and very clever programming and hardware design to deliver low cost storage appliances with great performance. And the use of open source means the extra access protocols other storage vendors try to charge for are included free.

There's plenty more to say on this subject.  For Sun, open source is not a matter of warm statements of alignment while we carry on with the same old business or keep our core products proprietary. I hope it's becoming clear that the Sun Model is a directional matter.


You forgot Step 4:

Install MSN toolbar.

Java + MSN toolbar = match made in heaven.

Posted by Arrack Osama on November 10, 2008 at 12:51 PM PST #

Yeah, I too was put off by Sun lending a hand to a convicted monopolist that constantly attacks Free software. Where's the logic? It seems more like Microsoft hubris, characterise by its "Slog tactics" < >.

Posted by Roy Schestowitz on November 10, 2008 at 07:26 PM PST #

I quite agree - see my last link today. Having said that, the desperation involved in MSN Search using the much-hated Java as a vector for self promotion is also a telling irony.

Posted by Simon Phipps on November 10, 2008 at 10:22 PM PST #

That's another way to look at it. :-)

"Don’t encourage new, cross-platform Java classes, especially don’t help get great Win 32 implementations written/deployed. [...] Do encourage fragmentation of the Java classlib space."

--Ben Slivka, Microsoft

Posted by Roy Schestowitz on November 11, 2008 at 02:51 AM PST #

In case you didn't realize it yet, the MSN toolbar is a vector for installing Silverlight (JavaFX killer) - yeah, that's what the new MSN toolbar is built in :-)

Posted by Steve Dallmer on November 11, 2008 at 04:58 PM PST #

Phase 4 wasn't supposed to be Ants' world domination? <> :)

On a more serious line of thought, I am keeping writing about the Sun Model, and I think that phase 1 and 2 deserve more attention.

Posted by Roberto Galoppini on November 12, 2008 at 01:06 AM PST #

I miss the cohesion between the steps. In fact I don't think they should be steps, they are like pieces of a puzzle; they interoperate and are in fact completely interdependent.

A thriving community of developers is not just something nice to have, it has a serious influence on code quality. Of course, this only happens if the code is actually available during development and not just "published as GPL". The open source development model.

Obviously, non-OSS additions and tools don't get extra eyeballs on their code at all. And that's where I think things go weird. Why does a company use an OSS dev model.... because it's a neat marketing trick (adoption), or because it produces better code? Even if you like the former, the latter is still very important!

Then the "oeh but we must have leverage to make money" debate begins, and suddenly the "better code" part disappears from the table for the proprietary extensions. All this may make short term business sense for the vendor, but from the perspective of a savvy potential client, it looks like the vendor either might not understand what the OSS dev model actually does, or chooses to ignore those aspects to score revenue. The client however does not give a stuff for the vendor's revenue, the client cares about code quality.

So, I think the "but we must make money" argument for proprietary extensions is a poor one, a bit more imagination and daring is required. An OSS company has to rethink how to drive decisions, for sales in particular. You can't simply aim for maximum revenue (commission based sales force also automatically creates this even if you don't explicitly decide for it) as it will warp the development side of things. An OSS company has to also innovate on the sales side.

If you truly believe that an OSS development model is better (code quality), then producing proprietary extensions is actually telling your clients "hi here we want you to pay us money for a component of lesser quality".
And that's weirdness....

Posted by Arjen Lentz on November 12, 2008 at 06:50 AM PST #

Post a Comment:
Comments are closed for this entry.

Thoughts and pointers on digital freedoms and technology markets. With a few photos too.


« July 2016