Perhaps you can clue me in: What was Sun's plan with Java? For a company that sold hardware, making a runtime that works everywhere, and has cross-platformness as a core feature seems sort of counterproductive. Did Sparc offer some sort of competitive advantage, hardware wise (like hardware assisted GC) or something?

Edit: In Microsoft's case with the CLR, I can see the cross-platform goals being mainly technical, but MS sells software and developer tools. So it makes sense for them to get everyone using their platform. Did Sun have an aggressive plan to control the ecosystem and sell tooling and whatnot around it?

As an original member of the Java group I can tell you they didn't really have a plan. Prior to it going viral the plan was to flush it and move on, everyone had made plans for a new job come Jun 30th, 1995 (the end of the Sun Fiscal quarter and the point at which the funding was ending for the "LiveOak" group.

Once it was being seen as the "death" of Microsoft, taking away the desktop monopoly and moving it into the browser. Well then it became a hammer to beat upon the enemy, and beat they did.

That it was effective against Microsoft ensured it would get funding and support, it also prevented some really useful things from coming to market from Sun at least (like a native compiler, ala gcj, in which to build Java 'binaries')

"For a company that sold hardware, making a runtime that works everywhere, and has cross-platformness as a core feature seems sort of counterproductive."

This might possibly be connected to the fact Sun no longer exists as an independent corporation.

> ... no longer exists as an independent corporation

You can safely say just "no longer exists". Oracle made sure it erased any trace of Sun trademark from anywhere it could. Now it is Oracle Java, Oracle VirtualBox,... I always wondered why they chose to totally eradicate this trademark instead of keeping it. Does anyone know?

Relatedly, I was at sun in 2002 and went to one of the town halls that Johnathan Schwartz was doing in menlo park. Someone asked how we would make money from {thing} and he was told by Mr. Shwartz that engineers shouldn't worry about how to make money from things, they should focus on the code.

One of the many reasons our stock price went from $80 to $2 methinks...

I would disagree with that characterization. At the time it was released Sun was losing the battle for "Web" mindshare to SGI and was on the verge of becoming irrelevant as the 'wave' of the Web hit its crusty servers. There was certainly a lot of 'drag' as they say in terms of buying servers from the maker of Java. Especially as it became the defacto business logic implementation tool. So while I don't think licensing revenues paid the development bills (I may be wrong there) I know it kept Sun in the game during the dot.com explosion.

I want to know why Jini failed and if something like it (internet dialtone) either exists now or will exist.