"Microsoft has said it expected to release a new version of Windows approximately three years after the introduction of Vista in January 2007. A company spokeswoman said Gates' comments are in line with a development cycle that usually releases a test version of the software before its official introduction."

So, that's 2010 at the earliest....not even including beta stages. It'll be a while.

windows 7, codename 'ultimate insult'

wouldn't really say "given up on vista" we don't say that novell has given up on openSuse 10 because openSuse 11 is about to come out out.

:P

I doubt it. For major software like this, you can bet that the next version has already been under planning/development long before the current version is released.

I can agree with Ak Random It can take years just to put some of the
major bugs in the code. Just to get it to the standards of some of the
older Windows.

... My laugh for the evening!

I think you guys are not being fair. It is a fine line between a bug and a feature. It pretty much depends by whom it is used. A little security issue for most users it is a just a bug, for a hacker, it is an amazing feature.

So, with that in mind, it is safe to say that Windows has more features than Linux does...

I think there's a misconception on actual number of bugs versus number of bugs perceived. For instance, there are by far more Windows users than there are Linux users. As such, more bugs will be found because Windows is used by more people. If just as many people were using Linux, there would be a lot more bugs found too. Anyway, this goes back to my original point, which is that that any large piece of software (e.g., Windows, Linux distribution, Firefox, OpenOffice, etc.) is complex, and the next release of that software will already be in planning/development when the current version is released.

BZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZT. Sorry, this is an old canard that needs to die an agonizing death. On the desktop the numbers are true, but Linux is the majority of server installs, and still Windows and IIS come up as the major areas of bugs and security issues. Since the underlying OS is the same for the desktop, it is probably a reasonably safe assumption that windows itself is buggier/less secure than Linux. Of course then there is your definition of a bug. To me, ActiveX is one gigantic bug, but Microsoft seems to see it as a feature.

The fact of the matter is that if you have software, you have bugs. Period. The question is how severe are they, do they allow your system to be compromised, and how quickly do they get fixed.

This reasoning is quite flawed because of two reasons: 1) There are by far more computers running as desktops than servers, so it's easy to keep the narrowing the scope of the problem until it's in favor of whatever you are arguing for; and 2) Extending what is observed in the server market to the desktop market is comparing apples-to-oranges, even if the underlying kernel is the same.

Agreed. But that's besides the point I was originally making.

Trying to use "number of bugs" as a metric is meaningless. To illustrate: suppose you were interested in buying a car, and could choose from two otherwise identical models. Also assume that car "A" had 6 identified bugs, and that car "B" only had one. Now, which is better car to buy? Obviously the real answer depends on the severity of the bugs, not the number -- if the 6 bugs with the first car were that a headlight was burned out, all four tires had low air pressure, and the ashtray was missing, but the one problem with car B was that the brakes didn't work, it would be ridiculous to claim that car "B" was superior due to it having 'fewer bugs'.

That is increasingly untrue. Server farms are getting huge, and have the added attraction of 24/7/365 uptime and are usually connected to the Internet continuously as well. Those are the kind of machines bot farmers really want. Also, increasingly, apps are being pushed onto servers (at least in the field I work in) and end users are getting lighter machines that really only run a web browser and an office productivity app.
No, actually the comparison is very valid. The premise of the old numbers argument is that all OS are pretty much the same when it comes to security and bugs and the prevalence of Windows OS on the desktop is the only reason why Windows keeps being found with more vulnerabilities and bugs. The argument further claims that as soon as OSX/Linux/SomeOtherOS reaches Windows-like numbers, they too will reach Windows-like numbers of bugs and vulnerabilities. The argument makes no reference to how the OS is being used, which is actually a good thing because it doesn't matter. What the argument conveniently ignores is that in areas where Windows is NOT the major OS, it is STILL the leader in bugs and vulnerabilities. In fact, I would argue that the server market is in fact a much better test of bugs and vulnerabilities because those machines tend to be used MUCH more heavily than desktop machines and are certainly exposed to a wider variety of malicious behavior. In other words, servers are being used much harder, and therefore have a higher chance of running into a bug in the underlying OS.

Ummm... when you say "more installations," you must stop and consider the environment you're referring-to. For instance, the majority of servers on the Internet run Linux, not Windows. Also, "most bugs" are not actually found by people, but by automated testing.

Yes, the production cycles overlap. Several product-lines are in development at one time and the cycles can be very long.

This reasoning is increasingly flawed. Although the size of server farms are increasing, the reason is because they are serving more and more clients. How much of those clients are other server farms versus how much of those clients are end users of some kind?

This is exactly the wrong reasoning. The two are different beasts, and can't be compared to one another.