I'm changing the established order of this week's column to move up the punch line to the lead paragraph: The next time somebody offers $450 million to buy your infant company, just take the money and run.
More than a few present and former insiders who own stock in PointCast wish that Chris Hassett had taken that advice to heart when he was negotiating a possible deal to sell his operation to Rupert Murdoch's News Corp.
It was a sweet deal that never got signed and so instead of finishing up like Kim Polese, a one-time fellow apostle of "push," Hassett's winding up his 15 minutes of fame as a latter-day Gary Kildall.
Perhaps more than any other executive in the history of the software industry, Kildall, the founder of Digital Research, committed the most famous flub of flubs when he blew a golden opportunity to supply IBM with the operating system for its PC.
Then there's Polese, who recently raked in the big payday when Marimba went public. Polese, the public face for this Kleiner Perkins-backed startup, smartly toned down the push angle in favor of developing a technology that allows corporations to automatically update and distribute their applications.
Watching the relative success enjoyed by Marimba, the second-guessers in the PointCast amen corner are understandably having a field day with this cautionary tale.
"It's criminal," said one stakeholder. "It should never have happened this way.
Let it be said up front that this one exec -- who is clearly in the know about how the way PointCast ran -- was no great admirer of the way Hassett ran the company.
"They thought they knew better and could parlay this into the biggest IPO in the history of high tech," he said. "It was pure vanity and greed."
Still, it must be said that there's no shortage of either in Silicon Valley, where more than a few outrageously Gordon Gekko-like executives are running first-class outfits.
And with the benefit of 20-20 hindsight, there's a lot of blame to pass around. Indeed, some of the company's more bizarre moments came during the tenure of Dave (the Rave?) Dorman, who earlier this year jumped ship to take a nice paying job as CEO of a newly created, $10 billion joint venture between AT&T and British Telecommunications.
First, PointCast decided to file for a public offering without coming up with a fresh, better story to sell to Wall Street. Then it abruptly canceled the offering, ostensibly to pursue a merger. That was a nice fig leaf to hide the embarrassment of an IPO surely doomed to become a bust.
The shame is that PointCast probably could have salvaged the situation by moving with more alacrity to resolve the lingering performance issues
One suggestion made at the time was to completely redefine the product, stripping it down to what it was supposed to originally be: a utility. (Remember that the original idea behind PointCast was how to build an effective advertising model online that met traditional advertising requirements.)
Some insiders contended that PointCast needed to put the product into a window rather than leave it floating around as an irritating screen saver. Even as late as a half year ago, they believe, the company could have successfully provided the market with an office-oriented instrumentation system that helped people better do their jobs.
Former PointCasters could spend the next eternity in pub debates about what went wrong. But I think the answer is a lot simpler: Where Hassett went wrong was in believing his own press clippings.
Push was an interesting technology, but Hassett got too caught up in the moment as the media babbled on about how it would change computing. Just like Marimba, the folks at PointCast needed to morph their company into something different.
They were just too slow and that's a recipe for disaster when the rest of the world is on Internet time.