Friday, April 4, 1997 by Dave Winer.
I think it was Yogi Berra, the great Yankees catcher of the 1950s, a lovable teddy-bear of a guy, who said It's deja vu all over again.
It's funny how the package doesn't always predict the intellect. To look at Yogi, you'd think he's kind of a slow guy. But he's deep. He came came up with It ain't over till it's over and You can observe a lot just by watching.
Right on Yogi!
Yesterday I ranted about Sun, but the real problem is the technology press. If only they were as challenging and astute and deep as the baseball press. Understand that it's a game, there are rules, strategies that work and strategies that don't.
A good baseball manager saves some pitchers for the next day. If I were a sports reporter, covering a baseball team, and if yesterday they used all their starting pitchers as relievers, I'd predict a losing outcome for today's game.
The technology press must be very confused about Java. All the pubs were represented at the Sun press conference on Wednesday, but where were the reports? Why was there no coverage of Sun's Java announcements in the New York Times or the Wall Street Journal or the San Jose Mercury on Thursday morning?
The Java licensing scheme, and its connection to their proposal to ISO, is an important story. Dig a little deeper guys and gals.
They aren't telling all about Java. There are secrets. What are the terms of the license agreements that Sun has struck with Netscape, Apple, Oracle, IBM, etc? It's confidential.
Even so, we can figure out what's happening. When the facts aren't in evidence, there's always the tea leaves and human nature, and the attitude of Silicon Valley, and unhappy employees, to fill in the blanks.
Let's say I want to do a license for Java. Get out the checkbook and start writing. Add another zero. Add another. Next year, more millions. Add it up. JavaSoft has 400 employees. That's a big burn rate.
OK -- what do I get for those millions? The right to use the Java brand name, for sure. Java is not the name of a language, but a trademark owned by Sun Microsystems. I pay Sun money for the right to say that this is Java.
There's a feedback system, according to Sun. First they listen to licensees, then they listen to the general world. Pay us money, and we'll implement your ideas. I suspect that the more money you pay, the more they listen.
James Gosling is the original Java guy. I heard him speak several times on Wednesday. As a language developer myself, I see lots to admire in him, but I also see his struggle.
If anyone is, Gosling is the soul of Java. He's still at Sun. If Java is a cash cow, he's the one who sold it. I don't begrudge him the money, he deserves to be a rich man. But how empowered is he at JavaSoft? How well do Alan Baratz and Scott McNealy listen to him? If he said "Java must be set free now" would they do it?
I watch body language. I watch Bill Joy look at Gosling for approval, while he explains how he was involved in the development of Java. I listen to Arthur Van Hoff of Marimba, another original member of the Java team. I've talked with Arthur one-on-one several times. I see a different Arthur on stage, a strong, enthusiastic, opinionated developer. I believe I see the connection between Gosling and Van Hoff, it's student and teacher -- there's respect there, both ways.
Gosling and his project were minor corporate footballs inside Sun. Not nurtured, forgotten in a quiet cul de sac while Sun fought other battles, while Bill Joy was the technical spokesperson for Sun. Even though legally, Sun owned Java, they had, on a personal and professionl level, disowned Gosling and his team.
Off on the side, working month after month, Gosling was in solitary achievement mode. That's why Java is a good language! That's how all excellent software happens.
Then, thru an accident of history, and a marketing manager who was paying attention to the outside world, an opening was created, and Java and was thrust into the opening. There was follow-thru. A small part of Sun came awake, decided to win, all the planets were lined up, and winning happened.
Java was a reaction to the wizards of Wizzy. A vote for the geeks. A necessary balance to the world of easy-to-use software.
Java was also love at first sight. Like the happy computer that greets Mac users, Java loves the world, and the world loves Java!
But that was just the beginning...
Like Java, Macintosh was an accident of history. Apple was dead in 1983 when the Mac came together. The IBM PC was a juggernaut, the Apple II was was a true cash cow, dying as a platform, and Lisa was a failed corporate head trip.
Macintosh filled a void. A friendly computer. And it had a long life ahead of it because it wasn't limited to 640K. And it was designed, not thrown together. Many of those advantages are still evident in the Mac platform. Multiple monitors, easy plug and play, interapplication communication.
By 1987, as the corporate credit-takers took over, the seeds of the downfall of the Mac fell into place. By 1990, it was evident how the Mac would end. When Windows 3.0 came out, and Apple failed to respond, it was clear that Mac was being viewed as a cash cow, not being managed or nurtured, but being allowed to serve egos, not a community, tethered to a corporate earnings statement and big engineering egos, when so much more was possible.
Java is in the corporate ownership phase of its life. Being set up to be assumed by Microsoft. This loop is closing right now. And, as I said yesterday, it's a shame. It's still possible for it to be much more than that, but there isn't much time left.
It's interesting to note that Larry Ellison has a different idea for Apple than Steve Jobs. A new set of public quotes from Ellison say that Rhapsody is a loser, that Apple's only valuable asset is the current Macintosh OS, and that all of Apple's remaining energy should be focused on that. The quotes have appeared in print, and are being confirmed by reporters who have interviewed him.
If this is his true intention, I support it. Rhapsody is a total Hail Mary play. Apple has never supported System 7. It's their only product. Focusing on that is the only way for Apple to remain a viable platform vendor.
Back to Java.
Simply put, Java needs a great 1997 graphic interface. Do we trust the engineers from Netscape, IBM and Sun to do it?
Microsoft has a deep base of know-how in easy to use graphic interfaces. Forget corporate reputations. This is 1997, the people who did the Mac have left Apple. Some of them work for Microsoft now. (For example, Steve Capps, one of the original Mac team members, now works for Microsoft.)
The next few months are critical. Sun has put its exclusive bet on the "Swing Set" team from Netscape and IBM. But what if they don't produce something as pretty as Microsoft? And is that good enough? No, we need better choices. This is 1997. Director happened. The web happened. I saw the demos on Wednesday. Microsoft's interface is much more attractive than the Sun-developed one.
Instead of having two choices, we now need fifty choices. Then let the market shake out. With any luck, given the huge opportunity, a more than adequate interface will emerge, a true alternative to Microsoft's.
The Sun-led strategy is a total Hail Mary play. Bet the whole outcome on one group of engineers. It isn't the quiet period anymore for Java. This ain't a rehearsal, it's the real thing.
I want a shot at this juicy opportunity. Hey, so does Arthur! Look at Bongo. Why doesn't Sun look to Marimba to solve this problem? See the bigger problem? Why should all the rest of us bet so heavily on the wisdom of Sun's management team? They're so overworked. And they're in the way.
When will Silicon Valley get the message that great software is created by people not companies? When will the financiers of Silicon Valley set up corporate systems that serve the products, instead of vice versa? And when will creative engineers stop selling their babies to the miserable petty agendas of small thinking corporate credit-takers?