A column needs gimmicks to succeed, but this column hasn't had many gimmicks so far. I'm not so smart, not such a good writer, and am thin on sex appeal if nothing else. About all I've had going for me, in fact, is an uncanny ability to piss people off, which generates reader mail, which generates more readers, etc. Here's a secret: Want to kill-off a columnist? Don't complain, just stop reading him/her. They'll be dead by morning.

The one really smart gimmick I did think up was my "Lunch with a Bigshot" feature, which you'll find at the bottom of this page. The idea was for me to go to lunch every week with an important industry person, to tape record the whole damned lunch, burps and all, then post that tape as a streaming audiofile on this page. It sounded like a great idea and still is, but it is a lot harder to do than you'd ever guess.

First, you have to get a bigshot to lunch. These people eat lunch, I'm sure of it. I have seen them eat before. But after I invented "Lunch with a Bigshot," they stopped eating. Why? Because the lunch file is completely unedited. Whatever they say, they can't take it back, can't deny it, and can't hide behind a publicist. If a publicist takes part in the lunch, you'd hear them madly back-pedalling and spin-doctoring for their doltish bosses. Publicists like to sound like themselves even less than do high-tech CEOs, which makes sense, since CEO's at least perform a business function.

So the fact that nobody wants to go to lunch with me ("Next month, I promise," they all say) is problem number one. Problem number two is audio quality. These lunches take place in restaurants where other people are eating lunch. And at least as far as tape recorders are concerned, the most important sound in any restaurant is the clinking of dishes. Conversation? Who needs that?

But it's getting better. I started with a pathetic microcassette recorder, advanced to a stereo cassette recorder and have just acquired (but not yet used, as you'll notice) a Sony Minidisk recorder with wireless microphones. It's the individual mikes that I believe will be the key to unlocking the true power of lunch -- that is, if I can get anyone to eat with me.

Well, last week someone finally did eat with me. It was only my second lunch with a bigshot, but it was a good one. The first lunch, which you can still suffer through, was with Kim Polese of Marimba, a person who for many many reasons will always be on my personal list of heroes. And now, lunch number two was with Marc Andreessen of Netscape.

The next big chunk of this column will be devoted to my impressions of lunch with Andreessen for those who won't bother to listen to the whole clinking hour. If you are one of those hardy souls who actually will listen to the hour, then fast-forward to the Power Computing stuff below.

Image -- Marc Andreessen is this cornfed kid from the Midwest who was plucked by Jim Clark from obscurity at the University of Illinois to co-found Netscape at the age of 23. Reality -- Marc Andreessen is a 40 year-old industrialist stuck in the body of a very tall 25 year-old. He is simply the most articulate person I have interviewed since -- well, since I interviewed Kim Polese. But Polese dances better than does Andreessen.

Marc Andreessen is being continually compared to Bill Gates, but I think it's a poor comparison. Gates has made more mistakes than Andreessen and has learned more lessons. Gates is scared while Andreessen has the false confidence of youth. They are equally smart and equally close to the technology (that is, not all that close at all but they can talk a good game if their conversation partner is an idiot like me). Where Gates has the advantage is in his killer instinct. Andreessen has others at Netscape to supply that for him, notably CEO Jim Barksdale. If Microsoft wants to kill Netscape, the key is hiring away Barksdale.

There were no PR people at our lunch to tell either of us not to curse, and no back-pedalling. Andreessen kept comfortably ahead of me the whole time and never broke a sweat, which explains yet again why I am nearing retirement age. I, however, ate the healthier lunch.

Listen to the audio file, there's lots of good stuff in there, but here are a couple of items I found to be especially interesting. Of course, they concern Microsoft.

Andressen declared that ActiveX was effectively dead, claiming that Microsoft had fully embraced Java as the superior technology. ActiveX events for developers seem to have halted as Microsoft pushes its J++ version of Java. And he's right. It has become more important for Microsoft to steal Java than to push a fully proprietary alternative with compromised security. This is good. Andreessen thinks it waill fail, though, as most companies reach for the Holy Grail of One True Java. We'll see.

But Andreessen went on to claim that this is just another example in a pattern of Microsoft backing-down. He painted a picture of Microsoft as the old IBM, tied too closely to legacy cash cow products that will eventually shrivel as Java takes over. I'm not sure he is right about this for two reasons. First, the One True Java is still far from a sure thing. Second, Microsoft has an uncanny ability to change positions overnight. Microsoft is not the old IBM or even the new IBM. Microsoft is a reflection of Bill Gates, who is willing to do whatever it takes, to change whatever must be changed, to win.

Listen to the audiofile.

Now on to Apple buying Power Computing. This story gets curiouser and curiouser. Steve Jobs, who for the moment has total control of Apple while owning only one share and not being on the payroll, has declared war on Mac cloners. This is war on Motorola, Power Computing, and Umax. Steve wanted all three companies to just go away, but it's not that easy, so he spent some of Apple's cash to "buy the core assets" of Power Computing.

Buy the core assets? What the heck does that mean? It means Power Computing, which had already announced it would begin building Windows PCs, will do just that, abandoning the Mac market. Power Computing owner Steven Kahng gets to keep his company's name and its factory. In effect, Apple paid him off, financing Kahng's move back to Intel PCs -- a move he had to make since it was clear Apple would never give Power a license to use MacOS 8.0 or Rhapsody and also clear that the Apple market was in big trouble. They bought from Kahng a business that effectively had no value specifically because Apple took away that value. If Apple's market share numbers were still visible, actions like this might look like restraint of trade and get some attention from the Feds.

What Apple gets from this deal is almost as much as Steve Jobs thinks. Apple gets back some engineering talent that a couple of years ago left Cupertino in disgust for Power Computing. Apple gets a team of workers from Texas who are used to a lean, mean, hard-working operation. Jobs thinks these workers will somehow innoculate their Apple counterparts, but it won't happen: they'll all be gone in a year. Apple also gets some people who know how to sell PCs over the phone, which means Apple will set up a direct marketing operation (Apple pretending to be Dell) and screw its dealers. Expect the dealers to howl, then die.

And it might even work, though not without making lots and lots more enemies for Apple. Thanks, Steve.

But there are a couple of wildcards in this hand that Jobs has dealt -- Steve Kahng and Motorola. Kahng will have more than 4 million Apple shares, making him for a time the company's largest shareholder. Jobs thinks Kahng will do what Jobs, himself, did earlier this year -- sell nearly everything before it's worthless. Jobs thinks he has Kahng under control, but with Kahng's future tied to the Windows market, not to Apple, I'm not so sure. Given the right external incentive (see below) Kahng might surprise everyone by how he votes those shares.

And that brings us to Motorola, which has been pissed at Apple for a long time. Because he successfully bullied Motorola over the years, first at Apple and then at NeXT, Jobs thinks he has Moto's number, but he doesn't. Jobs thinks that if Moto gets mad enough they'll just buy Apple outright, which they could do but probably won't.

Umax is not a major factor in this equation, but they will follow Motorola's lead. Motorola could buy Apple, perhaps with Steve Kahng's help. Or Motorola could use its muscle to steal from Apple the Macintosh standard, which I think is more likely. They'd do this either through using the BeOS with Mac emulation or Linux using ARDI's Executor Mac emulator. These discussions are being held right now, and the bottom line is that Motorola is determined to make and sell PowerPC machines no matter what Apple says.

This story is far from over. Look for fireworks soon.