[ThrowawayB7]

> "...embrace-extend-extinguish against Java..."

Early Java was horrid for everybody except the architecture astronauts who could cram ten GoF design patterns into a hello world program. It only got traction because a different wannabe monopolist, Sun Microsystems, spent heavily to get it pushed into CS curriculums. Fortunately, the one-two punch of Linux and Intel killed Sun or we might all be cursing them today instead of Microsoft.

[AnthonyMouse]

Sun was actually a decent company once upon a time. Their problem was that Microsoft excluded them from the market and starved them for money for so long that their hardware stopped being competitive, so that by the time Java made it so you could run some software on it, nobody wanted their hardware regardless.

It was only after they went bankrupt and got bought by Oracle that things like OpenSolaris getting killed off and Java lawsuits started happening.

[ThrowawayB7]

> "Their problem was that Microsoft excluded them from the market and starved them for money..."

That is not what happened. Sun Microsystems had immense revenue and clout in the server and enterprise space because of the dotcom boom, so much so that their advertising declared "We're the dot in dotcom." Microsoft was trying to duke it out with them in the server space but Windows Server was just barely starting to become decent at that point so MS didn't get all that much traction.

When the dotcom bust hit, Sun went into a tailspin because of the glut of Sun server hardware from dead dotcoms at bargain basement prices. That eventually passed but by that time Linux + Intel was good enough to undercut both Sun and Microsoft in the server space. With no way to compete with free as in beer software, Sun was doomed.

[AnthonyMouse]

> Sun Microsystems had immense revenue and clout in the server and enterprise space because of the dotcom boom, so much so that their advertising declared "We're the dot in dotcom."

Which is why Microsoft had to use such dirty tricks to prevent them from making inroads into workstations and desktops at the point that they still had competitive hardware.

> With no way to compete with free as in beer software, Sun was doomed.

Sun was a hardware company that did everything it could to commoditize software. That strategy works extremely well for hardware companies -- Intel successfully did the same thing for many years -- as long as their hardware is competitive.

They were perfectly content to sell SPARC hardware with Linux on it. But to do that they need to sell enough of it to keep up the R&D, i.e. they needed to ship desktop chips in similar quantities to Intel instead of only servers.

[defrost]

Original Sun hardware was rugged, operating mounted in truck beds running across corrugated roads rugged, optional full Faraday shielding for box, cables, and monitors tough. Really positive "never going to let you go" slotting of expansion cards into motherboard that resisted vibration issues. Fans that could go in a building HVAC.

That came at a cost and the market size of people that really really wanted / needed that field toughness was considerably smaller than the general office usage market.

[AnthonyMouse]

Because that's the market you can win when you're locked out of the mass market. But it's also not sustainable because then the mass market product will move up (Xeons based on x86) and you still can't move down with Microsoft blocking you.

Whereas if not for that, you could do both. Design a solid chip and then put dozens of them in a big iron cabinet for big money but also offer desktops with just one of them for prices that compete with Dell. Except that Dell's customers expect to open their existing Office documents and run their Windows API proprietary software and then won't buy from you.

[ThrowawayB7]

> "Which is why Microsoft had to use such dirty tricks to prevent them from making inroads into workstations and desktops at the point that they still had competitive hardware."

Huh? Sun Microsystems pretty much owned the workstation market by the mid to late 1990s, having beat out HP, DEC, IBM, etc. It was their game to lose, which they did. And I can't think of anything Sun ever offered in the desktop market in their heyday that was credible; their lunchbox SparcStations cost too much and delivered too little.

The main buoyancy behind Microsoft's push into the workstation market around Y2K is that Windows 2000 was much cheaper and ran on also much cheaper Intel processors. Compared to the ridiculous enterprise pricing from Sun and the other UNIX OEMs, the TCO made them very compelling.

[AnthonyMouse]

> And I can't think of anything Sun ever offered in the desktop market in their heyday that was credible; their lunchbox SparcStations cost too much and delivered too little.

When they were selling Unix and the alternative was DOS, Unix was better. Except that Microsoft did everything they could to keep everything tied to their own OS.

> Compared to the ridiculous enterprise pricing from Sun and the other UNIX OEMs, the TCO made them very compelling.

Sun was already on the decline by the time Windows 2000 was released. How many people were playing PC games on Sun hardware in the 1990s?

[jsiepkes]

None of your opinion on Java changes the fact that Microsoft used it's monopoly to execute an embrace, extend, extinguish strategy on Java. It is well documented since they lost an anti-trust court case on it [1].

[1] https://en.wikipedia.org/wiki/United_States_v._Microsoft_Cor....

[ThrowawayB7]

Back when the word hacker still meant something, that was the opinion of most hackers. Microsoft being bad guys did not make early Java versions good.

[EDIT] I'd actually say MS losing the J# lawsuit was a net positive since it gave Hejlsberg the opportunity to create C#.

[croes]

What does it matter if Java was good or bad? The point is Windows API didn’t win on merit but because MS attacked every competition

[bartread]

That’s not really true: it just didn’t win on the merits you care about. But it did have other merits: backwards and forwards compatibility, excellent developer documentation, and no need for expensive workstation grade hardware.

I don’t disagree on the skullduggery, but it’s worth bearing in mind that Sun, Oracle, IBM, and all the other big vendors of that period engaged in similar skullduggery. Microsoft was simply better at it, had better timing/luck, and therefore more resources and reach to play those games.

But one of those companies was always destined to become a scumbag monopoly: if it hadn’t been Microsoft, it would have been one of the others.

And the war to kick them back into line through OSS would have played out similarly, I suspect.

[dlcarrier]

That lawsuit was overturned on appeal. The DOJ settled with Microsoft's offer to publish some APIs, instead of launching another trial.

The Java stuff wasn't even the craziest part. The whole thing from investigation to the appeal took the FTC and DOJ 11 years, where they were unable to kill of Microsoft's dominance of Internet Explorer through lawfare, but it only took Mozilla and later Google six years, nearly half that time, for an open-source web browsers to have more market share than Internet Explorer.

It turns out that a better product was all that was needed. It's too bad that the Mozilla Foundation has changed course and is now adamant that Firefox be as unusable as possible.

[dlcarrier]

It went the other way around. Object-oriented programming came out of academia, before Sun adopted it into Java. They adopted it full-force, making programmers jump through all of its hoops whether they wanted to or not, unlike C++ and Python where for most programs, the only hint of the language being object oriented is the syntax.