I'm not familiar with Unix itself enough to comment, but with both Linux and *BSD you're able to start and restart services without a reboot and the architecture is such that you're much less likely to end up in a situation where you can't perform whatever action you need to in order to clear the error manually. I'm not sure how I would even go about looking up how to do a lot of that with a Windows box.
And windows does NOT have services.msc where you can start, stop and restart services? Like, apache or mysql, for example... Just sayin....
Really? Can you contribute code to H.264?
Can you contribute code to IEEE802.3? E.164?
Really H.264 may have been public but I would not call it open.
It was open in the sense that the major implementers and users collaborated under ISO stewardship to create it. Basically, everybody that mattered had a chance to collaborate. The fact that nobody came knocking on every basement-dwelling opensource evangelist's door asking "hey dude, wanna create a new video format? or do you just wanna rewrite a networking stack or reinvent the wheel or something?" does not make the process any more closed. Had it not been Google and pie-in-the-sky HTML5 guys creating a buzz over it, none of you dweebs would be the wiser. Nobody would even care, apart from the ones that have a video-based business to conduct (hey, what a coincidence - the ones that participated in the ISO process!). But now that sugar-daddy Google is troubled, every binary-faced "evangelist" feels the need to jump in and start fencing away "teh attakers of teh 0pen g00gle" not even understanding what it takes to create stuff like this. I mean, before sugardaddy released VP8+Vorbis+Matroska as gSpot..err...webm, everybody was howling about theora and its next incomplete version ptlaragtrewloa which was oh-that-great and should be standardized for ever and ever. Where is all this theora-praising now?
The aim of science is to seek the simplest explanations of complex facts. Seek simplicity and distrust it. -- Whitehead.