Views on the Industry

Components

As is becoming increasingly evident to those in IT, you don't bet on OS', you bet on app. servers: http://www.businessweek.com/technology/content/jun2005/tc2005067_2133_tc024.htm

And this whole $50 million campaign is to keep the OS alive in time to prove the app. server:
http://www.sun.com/emrkt/share/

But the real news, which predictably flew below my radar is this, in particular project Nexus:
http://www.businessweek.com/technology/content/jun2005/tc2005068_5622_tc024.htm?campaign_id=topStories_ssi_5

Since 1999, we have been singing the tune of how to do enterprise software, and people like i2 and Ariba insisted on pushing into irrelevance by doing monolithic suites. Now the King of Monolithicity has gone the component route. That is more than what salesforce.com can say (kicking out Sueltz is not good PR for a CEO with too much to say, and do already - - enough with the yoga poses, no one cares). We did whitepapers, magazine inserts, conference speeches, and press releases:
http://www.sun.com/smi/Press/sunflash/2000-09/sunflash.20000926.i1.html

But no one wanted to listen. We signed up partners:
http://java.sun.com/features/2001/03/ejb.html

But no one wanted to follow. And now when all is lost, when BEA has lost its way (they should have been doing more not less component apps.), when IBM has abandoned its history (as mentioned before, where have you gone San Francisco?), when Oracle buys another monolithic suite and calls it Fusion (nice one, RS), and Microsoft has bailed out on Hailstorm, I will once again sing the familiar tune that apparently only Siebel (?!) has caught on to. Do modular, it is the only thing that works. JBoss is doing some infrastructure modularity, and I would love to see them to do some apps. along with it, and maybe that is what they mean by the "Framework" discussion. What the hell is a framework anyway, just make pre-built components.

I am going to have to take a little break soon from this blog, as I work on material for the next phase of it, and I thought it somewhat fitting to end the enterprise computing discussion on something near to me. I believe in Solaris' functionality, but not its pre-eminence. I believe in Sun's future, but not necessarily its leadership. I find Microsoft palatable for the first time since the Mac II, but still believe JAX is the one to beat. I hope you have found my insights useful or at least comprehendable. Everything I have said is verifiable through public materials, so let me know if you want to continue the discussion. Until then, continue to look to Ashlee for your information at El Reg. The only thing I won't be able to answer for you is whatever happened to Mr. Herring:
http://www.theregister.co.uk/2002/10/28/sun_aims_low_with_new/

Sometime down the road, all of these issues will become more clear...

posted by douglas dooley @ 8:32 AM