Looking Beyond Vista To Fiji and Vienna

Vinit wrote in with an article that describes Microsoft's strategy for future versions of Windows. It begins: "As we all know that Microsoft Vista was originally scheduled to be released in 2003, after two years of Windows XP, but it got delayed by over five years due to various reasons. Definitely, Vista is very very improved OS over the previous versions, but the delayed in the launch has cost Microsoft, billions of dollars. Now the question at the moment is, what exactly after Vista? Microsoft can't afford to wait another five years for an operating system. People are becoming more aware of the choices they have, and Linux is no longer a hobbyist OS, and that day isn't far away when it becomes simple enough to be a viable alternative to Windows. The competition is fierce. That is why, to stay at the top, Microsoft has planned a 'Vista R2', codenamed 'Fiji' which will be released some time in 2008. And after Fiji, there will be Windows 'Vienna'. Windows Fiji, will not be a totally different OS from Vista; but it will be an add-on. Whereas Vienna will be totally different from Vista."

Comments

Actual article

[VAXGeek]

REAL article with actual meat: http://jameskyton.wordpress.com/2006/12/29/beyond- windows-vista-fiji-and-vienna/

Don't you hate reading the whole thing and getting to the end and seeing SOURCE? I wish I could digg this article DOWN!

Re:Five years?

[johnw]

But the article says it was delayed by 5+ years, not that it came out 5+ years after XP.

Perhaps what the author meant to say was that the intended 2 year interval between releases became 5+ years.

other reasons why vista is inferior

[mr_death]

Take the points in the parent posting, and add:

50+ millions lines of code bloat

lots of stupid, unnecessary eye candy

alleged security features, some that have already been broken ("most secure o/s ever", my ass)

a virgin ip stack

DRM silliness

kernel restrictions that keep third party security systems out -- said systems having done a much better job than Microsquish at keeping the bad guys out. You can, of course, pay extra for windows "defender" -- somewhat like buying an antidote from the people that poisoned you in the first place

As Ren and Stimpy might say to Ballmer, "you eeeediot!"

Re:64bit linux world-domination-201 by 2008

[TheRaven64]

Eveyone will switch to 64bit hardware by the end of 2008, it's impossible otherwise Upgrade cycles have been slowing down a lot for a while now. Many people I know are still using machines from 2001/2002, and are only upgrading because people on faster upgrade cycles are giving them newer kit. This is true of corporate machines too; faster doesn't give much more productivity anymore, so machines are only replaced when they break.

The moore's law tells us what will be the memory capacity by 2008. Moore's Law tells us one thing; the number of transistors that it is possible to put on a chip for a fixed price will double every 18 months. This doesn't necessarily mean RAM sizes will double every 18 months, it could mean that RAM prices will half every 18 months. In practice, it's somewhere in between. A little over ten years ago, people put about £125 of RAM into a £1000 computer. Now, they put about £30 of RAM into a £300 computer for the same market segment. A lot of users don't really need more then 256MB. It's not sensible to buy less than about 1GB of RAM these days, so putting that in a machine gives them some headroom. People will only buy more than 4GB of RAM if it is cheaper to buy more because no one is making the smaller modules, or if they actually need it. Otherwise, they'll just go with a cheaper machine.

And with 32bit it's impossible to address more than 4GB Good thing every chip since the Pentium Pro has supported PAE then. This allows you to address up to 36GB of memory. Individual applications can only address 4GB unless they are specially written to use multiple segments, but then, most applications don't need to address more than 4GB of memory.

Re:Shared libraries? So now DLLs are good?

[be-fan]

Linux shared libraries are quite different from DLLs. The shared library mechanism on *NIX systems has features that mitigate a lot of the problems of "DLL hell".

Re:Shared libraries? So now DLLs are good?

[iabervon]

Our shared libraries support useful versioning. A program gets linked against a library by name, but it records the major version of the library that it used. When you run it, it looks for the newest library with that name and major version. Libraries get new major versions when they change in non-backwards-compatible ways and only new minor versions for bug fixes and backwards-compatible improvements. Also, when a version is supposed to be backwards-compatible, it's generally actually backwards-compatible.

DLLs are only bad because you can't set up a system with a sufficiently complete collection of them at the same time that every program will get the DLL it needs. Just because Microsoft's implementation of something is terminally broken doesn't mean it's not otherwise a great idea.

Re:Fiji

[alc6379]

Up until Windows XP, there had always been two separate codebases-- the 9x and the NT codebases. NT was for "business use" and 9x was for "home/small business" use, basically. XP changed that-- both the Home and Professional versions use the same NT codebase that was started in NT, and moved into 2k.

2000 is technically NT 5.0.

XP is technically NT 5.1

Server 2003 wasn't filler, it was designed to fulfill and entirely different role-- serving. It's the same NT codebase as always, it just has enhancements/modifications to better support serving and scalability. It's basically XP without all of the userland GUI stuff in it. Technically, it's NT 5.2

For that matter, 98 wasn't really filler, either. It was how they should have done 95 in the first place! ME, yes, that was filler. I will give you that much.

For more information on how things actually are/were, check this page out:
http://en.wikipedia.org/wiki/Windows_NT