Toolkit

Story: Sun's plans...and how they could go wrong

• 
• 

George,

I really enjoyed reading your article on ZDnet about Sun's plans. I had a couple of comments to float by you about Sun's strategy and your take on it.

First, I agree that Red Hat is the weak rival in this match-up. My company is a technology consultancy that is solely focused on Linux and Open Source, and we are getting a lot of business now that the first round of Red Hat service contracts are expiring. Many large Fortune 100 and 500 companies we are working with are unhappy with Red Hat's ability to support their product.

Novell seems conspicuously missing in Sun's plans. Novell is a large enough company with a long enough history supporting large enterprises to give Sun some real competition in the services space. This is especially true, since partnering with Novell for directory services, Linux and business application support allows an enterprise to stay with their traditional hardware vendors like Dell, HP or IBM.

The other trend that Sun's plan is missing revolves around the fact that Linux is already past the point of no-return in the market. Sun has frankly already lost the Linux battle because they are not leading the engineering and technology development. The new threat is a multi-headed hydra in the form of wide-spread Open Source software development.

Sun is not the same driving force behind enterprise UNIX standards that they once were. Where are they driving the 2004 technology versions of the Pluggable Authentication Module (PAM), the Name Service Switch (NSS), or Network Information Service (NIS), or the ubiquitous automounter? These technologies were all invented by Sun in the early days of UNIX to enable the enterprise to more effectively use UNIX. The other UNIX vendors like IBM and HP were forced to adopt them to keep up!

Sun has a couple of technology areas that they could still be relevant, but Open Source software development is already starting to address these issues and is winning the market mindshare and momentum. Their new enterprise volume manager looks great, but they are coming late and the code needs to be released to help define their approach as the best standard for interoperability with Linux and other commercial UNIX vendors. Linux *already* supports using an LDAP directory as a replacement for the venerable NIS in the enterprise and the commercial UNIX vendors are struggling to keep up and to interoperate with each others implementations (they don't follow the rfc2307 fully in most cases). Open Source software is starting to pull the strings on emerging standards like the recent email SenderID fiasco in the IETF with Microsoft. They fact that Open Source was barred from using the patents influenced AOL and others to drop their support. This used to be Sun's domain; leading the development of open standards to advance their corporate position and UNIX in general.

I still remember justifying the selection of Sun as a vendor (when I was still doing UNIX administration) because they lead the technology development for the whole UNIX market. Now Sun is just selling overpriced commodity functionality. The innovation and mind share is coming from the Open Source space even in their own Java technologies. The enabling applications for development effeciency aren't coming from Sun. They are in the Apache group's Ant and Tomcat; in the JBoss application server; the JUnit and JMeter testing tools. Sun isn't even leading the development and use of their own technologies any more! They aren't fighting the good fight and people out here in the market know it.

I know they looked alive when they spoke to you, but I think you might be overly optimistic on their overall chance to succeed. They need to get back into the drivers seat and lead by example. They need to make UNIX work better in the enterprise and drive those standards into the Linux and commercial UNIX market. They need to identify those companies that are obsticles against open systems and open standards, like

• 
• 

• 
• 
• 

Full Talkback thread