News on the next Windows OS?

Somebody mentioned today that 2009 was announced as a release-year for the next version of Windows...I haven't heard this anywhere else though. Paolo, is this true?

If it is indeed true, how soon until we see CTP's and other releases? When does the community to getting full of talks regarding the next great OS?

jsampsonPC wrote:

Somebody mentioned today that 2009 was announced as a release-year for the next version of Windows...I haven't heard this anywhere else though. Paolo, is this true?

If it is indeed true, how soon until we see CTP's and other releases? When does the community to getting full of talks regarding the next great OS?

I want to know too. I recall Steve Ballmer saying that MS will start to ship a new OS every 2 years or so.

But what is going to be new in the new OS? Will MS take our advice and start changing the kernel so that its more secure?

Will the UI change such that its more cool? More customizable? For instance I want to be able to rearrange the open windows tabs in the task bar , just like one would arrange tabs in a browser.

I mean the little things and the big things. Vista was better than older versions, but still its not where people want windows to be. I hope the next OS will improve on Vista such that its more secure, more customizable and more fun. Perhaps by the time the next OS ships MS would have incorporated a new way to take advantage of Mutli-Cores using Transactional Memory, and other good stuff.


What ever micro-surgery MS will do to the next OS will define life for us devs in terms of programmability in Multi-Core world. This is the time that makes or breaks a product. I hope Windows does not get broken because of the shift from vertical performance gains (Ghz) to horizontal ones (Cores).

My 2 Cents.[A]

SecretSoftware wrote:

Will the UI change such that its more cool? More customizable? For instance I want to be able to rearrange the open windows tabs in the task bar , just like one would arrange tabs in a browser.

There is actually a free mod that you can download that does that.  Works very well, although I cannot remember the name of it at the moment.

I did post the question here some months back, and that was where I got the answer from.

Who wants to bet the next IE will be released around the same time? It's looking that way.

[C]

jsampsonPC wrote:

Somebody mentioned today that 2009 was announced as a release-year for the next version of Windows...I haven't heard this anywhere else though.

Nobody said something like that.
At the moment Microsoft is working on:
- Windows XP Service Pack 3
- Windows Vista Service Pack 1
- Windows Server 2008

Most of them will be finished around the end of the year and then they can start thinking about a next version.

SecretSoftware wrote:



jsampsonPC wrote: Somebody mentioned today that 2009 was announced as a release-year for the next version of Windows...I haven't heard this anywhere else though. Paolo, is this true? If it is indeed true, how soon until we see CTP's and other releases? When does the community to getting full of talks regarding the next great OS?

I want to know too. I recall Steve Ballmer saying that MS will start to ship a new OS every 2 years or so.

Awesome! (legal) free windows forever! In the brief amount of time between release candidates and betas, you have the 30days of free trial use!

Unlikely. The last official roadmap I saw had about 3 years between Vista and "Vista R2" followed by another 3 years before the next major Windows upgrade.

More like Vista LOL2. Before ever prodding a stick at Windows again, I'll wait until the new shell is introduced. And even then.

SecretSoftware wrote:

Will MS take our advice and start changing the kernel so that its more secure?

Right, because they've never thought of that before. Kernel improvements are a continuous process. A lot of work was done for Vista to make the kernel more modular and securable, but it's something that time, resources, and backcompat make impossible to do all at once.

There's a kernel video with smart people who acknowledge what they did wrong and are going to fix it.

Charles: I wouldn't mind seeing an update video, the old one is very good but it would be interesting to see how their plans are solidifying.

We cannot say if we are/we are not working on the next Windows, but one thing is always puzzling me: you say that the kernel should be more secure. What exactly is not secure in the Windows kernel?

PaoloM wrote:

We cannot say if we are/we are not working on the next Windows, but one thing is always puzzling me: you say that the kernel should be more secure. What exactly is not secure in the Windows kernel?

I was refering to the fact that the WinKernel , or much of its code, runs still in Ring0, and MS did not move it  from Ring0 to Ring1 in the Privilege Layers, or the kernel onion. Moving much of the kernel code from Ring0 to Ring 1 would allow PatchGuard to then secure Ring1 Code. This way you get rid of the impracticality of securing code with same privilege level that exist in Vista.

Or, MS could introduce Zones within the Kernel layer, where one zone would have more Privilege than the other (layers within layers). Kind of like the Throne and the King servents. Or the nucleous in a cell. Zone 0 Zone 1 Zone 2. Zone 0 Runs hyperviser and hurestics, Zone 1,2 run Kernel and other stuff.

Right now there are rootkits that can install in Vista and you cannot detect it easily nor kill it easily because it runs in Ring0 or 1. PatchGuard cannot do anything because this code run in its same privilege level or above. Hence you have a false sense of security.


I know that MS can say, well we have given the user tools to stop installation, but thats like saying the user should know how windows works or they should know Assembly to read binaries of  what they want to install and see if they are safe or not. The user does not know and the system should know to defend itself and alert the  user. System should have heuristics to tell the user meaningful things to be able to act upon it by denying install or not.

Sometimes code can install without a prompt, using specially crafted exploits.


More in this article : Remediation of Application-Specific Security Vulnerabilities at Runtime

ZippyV wrote:

Nobody said something like that.
At the moment Microsoft is working on:
- Windows XP Service Pack 3
- Windows Vista Service Pack 1
- Windows Server 2008

Most of them will be finished around the end of the year and then they can start thinking about a next version.

You don't honestly believe Microsoft is clueless about the next version of Windows until they've released service packs and server versions of the OS, do you?

I believe their strategic thinking is slightly more far sighted than that. I hope so. There must be some grand scheme, that may change and evolve underway, but which is a path that they want to follow.

I thought it was clear that MS is now working on a vista successor behind closed doors with a targeted release of 2009 to 2010, as I read this weeks ago...

And I also read, that they want to do more than just cosmetical changes to the UI since the maintainer of the new office 2007 UI changed over to the Windows Team to do more than little changes...

Speaking of behind closed doors, I think that Microsoft's extreme "transparency" with Longhorn betas showing up in 2003, almost four years before RTM, hurt them more than it helped them because it let people see each feature getting axed (WinFS for one). Release a beta when the feature list has been set in stone and they're mostly working.

PaoloM wrote:

We cannot say if we are/we are not working on the next Windows, but one thing is always puzzling me: you say that the kernel should be more secure. What exactly is not secure in the Windows kernel?

Badly written kernel-mode drivers. (I'm looking at you Nvidia)

SecretSoftware wrote:


I know that MS can say, well we have given the user tools to stop installation, but thats like saying the user should know how windows works or they should know Assembly to read binaries of  what they want to install and see if they are safe or not. The user does not know and the system should know to defend itself and alert the  user. System should have heuristics to tell the user meaningful things to be able to act upon it by denying install or not.

Funnily enough, I'm working on that at the moment. Sadly, however, you can't say very much about a compiled binary. You cannot (in general) say that a program will be able to write files, access the network etc before it runs, unless you change your compiled binary language, and this is beyond the scope of modern operating systems (and there's a hint as to what I'm working on).

SecretSoftware wrote:


Sometimes code can install without a prompt, using specially crafted exploits.

In Vista only by running an unelevated program.