That was an interesting article, though I'm not sure I agree completely with his Microsoft history. This isn't the first time backwards compatibility was scrapped. It happened when developers moved from 16-bit Windows 3.1 APIs to 32-bit Windows APIs. You'll notice a lot of DLLs with a 32 at the end of their name (before the extension) for that very reason. The difference is they had to 'reinvent the wheel' then, not so this time around. MS dug their own hole when they decided all apps should be deployed as web services. It didn't dawn on them that if they were, Windows wouldn't be needed. After all, it's not that much more difficult to run a browser on Linux or OS X. So now they have to reinvent the web with proprietary APIs that require Windows. I think this link has been posted before, but it seems particularly applicable to the discussion, so I'm posting it again: http://news.netcraft.com/archives/20..._and_mono.html

Holy cow. There is so much that I do not know about .Net, XAML and WinFX. I really need to read up on that.

From what I can gather, is it that Longhorn is supposed to be an OS that is tied to the Internet?

If that is the case, then yes Linux sounds like it may have a hard time to fight that...of course they have to get Longhorn running first...

Mike

I'm not as worried as the mono guy seems to be. Firstly, not only has MS alienated many "old school" developers, they're going to do it again to early adopters of dot-net when Longhorn is released (as ShakyJake's link pointed out.) What's that saying, "once bitten twice shy?" Twice bitten should really make them think then. Secondly, my personal opinion is the web services "revolution" has proven to be a lot of hype. The internet has certainly been revolutionary, but fat apps (what the other guy calls "rich apps") are still very popular. I really don't think most users want to give up control of their computers and data to application service providers, regardless of how easy it is to do so. Of course, as always, I could be wrong. ;)

Webmin is a master piece of Web-based application example. I would take Webmin any time ratter then any fancy slow richer applications...

Crito: as people that use IRC a lot, and people that use p2p a lot start, finding there downloaded media (or maybe "illegally downloaded media") disappear and cops at there front door, i think people will ditch windows/M$ OS's rather fast (after all, computers are all about chatting and sending files to one another for most people, )..,, i know i would, well, i already have .. but still.......

Webmin is a truly outstanding web app, but an internet connection isn't required to use it and it has to be installed locally, so I'd argue that it's not a web service. I'd also argue that the webmin GUI isn't "richer" or faster than something like Mandrake's Control Center, but it's certainly much easier to deploy and maintain, which are the primary advantages all web apps share. There's no need to port code to other platforms and there's no need to update clients to support new features.

I think SciYro grasps the central issue is trust however. It's obvious big corporations don't trust users. The government certainly doesn't trust users. Yet big corps and governments think users should trust them unconditionally, despite repeated violations of people's privacy. Sure, the justifications, stopping crime and terrorism, are good ones, but once individuals lose control it's virtually impossible to get it back. I'm going to need a lot more convincing before voluntarily giving up any (more) of my rights.