I've thought a lot about this over the past few years. I believe there are a couple of cultural factors which contribute to it. On the large scale, Apple's "I'm a Mac. I'm a PC," made Microsoft bashing socially acceptable in the general culture. The second is more subtle, but more applicable to HN.
Microsoft is not a Silicon Valley company and it never was. One source of animosity that this has created is that it has been a powerful competitor to a great number of Silicon Valley startups. More deeply, as a bootstrapped company, it goes against the economic model upon which Silicon Valley's entire capital structure depends...YC cannot produce another Microsoft because YC companies take outside investment at day one. A startup which emulates Microsoft doesn't offer an opportunity to make early stage investors rich.
Maybe it's not just the "what" but the "how"? The perception is that Apple stomped RIM into the ground fair and square, while Microsoft mortally wounded (then beloved) Netscape through anti-competitive bundling and partner manipulation?
Google is every bit as much an existential threat to most small software companies now as Microsoft was at the height of its power, but we don't have any resonant stories of Google competitors being taken out unfairly.
Lest we forget, Netscape was acquired by AOL for more than $4 billion. One can blame Microsoft for it's demise, but both the acquisition and the anti-trust suit occurred in the same year. In other words, despite Microsoft's actions, there was a solid exit on paper.
Microsoft also doesn't do much that is interesting. They tend to imitate and their imitations are seldom better, and are often worse, than the originals. The only moderately recent exception I can think of offhand is Kinect. Are there other innovations they've brought to market over recent years? Microsoft's research division does amazing things, of course, but I'm talking about products.
In addition to the other examples mentioned, I would definitely call the F# and C# languages "interesting." I would also call their new approach to async "interesting." There are quite a few interesting things going on, but you don't find microsoft "interesting" you probably rarely hear about it.
On a pure technical level C# is better than Java, and F# is one of the most interesting languages in recent years. (SQL server really does have nothing going for it).
But the pure technical level isn't all that matters for a programming language. There's not the same library community around .net, because it came five years later and not enough better than java. And there's no migration path, because - which is worse - none of the MS stack interoperates with anything else.
I would, quite genuinely, like to do parts of my current job in F#. But there's no eclipse support for it. Even if I wanted to use visual studio, which I don't, it doesn't run on linux (and that's to say nothing of deployment - our ops team has been burned enough times trying to run windows servers). If I'm going to write things in F# there has to be a way for the rest of my code to use it, but F# doesn't run on the JVM so I can't write a library in it (or call JVM libraries), and when I looked for protocol buffers support all I can see is some third-party google code project.
I wouldn't want MS to give up on .net as it's actually a really good VM, so I guess that part makes sense, but windows is useless on the workstation and the server - and it seems like all the cool MS technology only runs on windows.
But lets be honest, Steve wasn't a paragon of virtue himself. Actually the more I read about him (just read the June Fast company article) the more I think he lets his emotions dictate his corporate actions. I mean, who ever publicly announces that they're starting a rival company in revenge for being kicked out?
I like (todays) Bill Gates as a human being. I like (the last current build of ) Steve Jobs purely because of his marketing skills.