% fortune -ae paul murphy

MySQL: what might have been

Today, I want to jump up and down and yell "I told you so" at Sun in the loudest possible way - okay, that's not nice (or sure to be right, either), but bear with me, there's a point beyond growing my own ego.

About six years ago I spoke to then Sun executive Rich Green and a few other people about two related projects - a Sun Ray demonstration project and their purchase and open sourcing of Unify Corporation's Vision and Data Server products at a combined cost I projected at about $25 million in the first year and about $12 million a year after that.

They did try to do the Sun Ray demo - but without any of the open source elements, with Wintel people, and essentially in secret - with the predictable result.

But they immediately rejected the notion that Sun should buy and open source Unify's intellectual property - in my opinion at the time in part because one of the junior players suffered severe not invented here syndrome and in part because Green and the advisors he listened to had no clues about the technologies involved and didn't understand how the Unify Vision technologies meshed with business uses for Sun Ray.

In 2004 I wrote a linuxinsider story about the Unify idea - and got no executive response. This year Sun got pushed into spending a billion dollars making sure that neither IBM nor Oracle could put MySQL out of business and thus cripple Sun's open source strategy.

So here's my, admittedly self-inflating, thought: compared to MySQL today Unify Data Server then was the better foundation product for complex, cross platform, applications - and Vision is still the best match out there for Sun Ray oriented business application development. Put those together with the quality and expectations of the technical people at Unify in 2002/3, and it's reasonable to think that if Sun had bought that company as a going concern in early 2003, they'd have spent, including acquisition costs, operating losses, and funding for a new development initiative a bit less than $70 million by now.

Meanwhile they might have spent another $15 million or so on the Sun Ray/Vision demonstration - and might now own a corresponding chunk of the market for business applications and their delivery infrastructure.

Jim Grisanzio to the contrary, the history to date suggests that this idea would have worked out well: Sun's commitment to the open source version of DataServer would have meant that the other guy's interest in shutting down MySQL would never have materialized - meaning that MySQL would still be an independent, open source, powerhouse; the tech world would have a major new open source database/applications environment to work with; and Sun would be ahead by about $915 million in cash plus whatever revenues the Vision/Sun Ray demonstration might have led to.

Part of the opportunity, that based on the 2002 demonstration project idea, still exists - and is still a no brainer. That's the point here, not the sheer chutzpah of arguing that NIH syndrome combined with narrow mindedness to cost Sun shareholders a billion bucks in cash and a major market opportunity, but underlining that Joe middle manager doesn't order a thousand Sun Rays on SPARC because he doesn't have a clue what they're good for - and what he hears from the PC salesman on the next barstool tends too have that distinct MSNBC quality to it.

And that's the real bottom line here: absent a well thought through, well managed, national scale demonstration project outside the secrecy businesses, customers won't know about many of Sun's better technologies - and people who don't know about them, don't buy them.


Paul Murphy wrote and published The Unix Guide to Defenestration. Murphy is a 25-year veteran of the I.T. consulting industry, specializing in Unix and Unix-related management issues.