I owned a NeXT. NeXT was never a direct competitor for MS. It was a workstation (remember those?) not a PC or PC equivalent. It was going after Sun's market. I did happen to own one as a mere mortal, but almost without exception, the only other standalone users were developers. The major installations were all running large networks by the standards of that day (NSA, derivatives players).

You are right re OS2. but that was late 1980s, MS hegemony less well established then. Never sure what Be's target market was, but I don't recall them as targeting broader PC market.

Not that Google is, per comments above, but the announcement did say they intended to then migrate later to bigger hardware.

The whole Linux open source is not targeting MS directly, since they don't operate from a business strategy headset, but a purity of code vantage. They could care less re user friendliness. Recall you 've had to have companies like Red Hat jump into fill that gap. If they had "get Microsoft" ambitions, they'd care about the needs of mere mortals.

Microsoft's core business is Windows and Office. Right now Google is targeting the Netbook market, which would only be a portion of the Windows market, so they are not exactly challenging Microsoft's core business yet.

As for other challengers, what about NeXT, Apple (why rule them out?), Be, IBM (OS/2), various Linux distros (SuSE from several years ago and Ubuntu from today come to mind), Sun's Java Desktop Environment, etc?

Chrome OS is more like WebOS. The actual OS runs off of the Linux kernel (so in that sense you could lump it into the Linux camp) and then launches Chrome, Google's web browser. Are consumers really ready for a Web-only experience? What happens when you don't have Internet access?

I have to wonder if Google is missing the boat here. Chrome OS reminds me of thin clients, which appeal more to businesses than individual consumers.

You Could not be more wrong if you tried. IBM created and opened the PC platform. MS had nothing to do with it. Linux takes nothing from microsoft. microsoft is the biggest parasite ever. I wish I had back all the time I've spent fixing stuff that happens because microsoft is so inept/corrupt.

It should be much easier to bullet proof the Linux kernel than to bullet proof Windows. For many years Bill Gates tightly coupled the parts of his OS, partially to improve performance and partially to prevent competitors from offering superior versions of those OS components. That architecture can't make bullet proofing the OS easier.

Gates did not invent the PC architecture, IBM did. Microsoft did not even write the first version of MS-DOS, they bought if from somebody. Gates did invent Basic, a horrible language for many years until it was forcibly cleaned up in the .Net incarnation.

.Net was not a major technical innovation. .Net copied the best ideas from the Java architecture. I will say that if you already know C++, it is much easier to learn C# than Java.

Eric Schmidt, Google CEO (IIRC), is on the Apple board of directors. You have to think that Apple and Google have complementary strategies that focus on drying up MS revenue.

I switched from Internet Explorer (MS browser) to Google Chrome. Google Chrome is way better in terms of speed and reliability.

There is no way that Google can provide human customer support for a product that they give away for free (presumably in a volume of 10's of millions if not 100's of millions). Google will have to construct/improve expert systems to deal with this issue.

Google's support of open source is awesome.

Even though I used to work for Intel, I don't think that we're stuck with x86 for the foreseeable future.

My understanding is that Google has said that Android and Chrome OS are completely separate products. It would really be interesting to know what, if anything, that they share (i.e. from a developer perspective).

Logically GOOG would attack MSFT at its weakest, most inflexible point: netbooks, where Vista doesn't run at all, and Windows 7 Beta is reportedly slow (and also unfinished).

Just as logically, GOOG would attack MSFT where switching costs are lowest. Netbooks are not used for heavyweight desktop software, and hence their users have lower OS switching costs than users of, say, Windows-based accounting software.

Why do it? Because user-friendly netbooks would hugely increase the total number of Google users. For now, netbook adoption is limited by the expense and slowness of Windows, and by the user-unfriendliness of Linux.

Linux distros, though super useful (I run five), have low consumer adoption because of less polished UI and peripheral support. These are solvable problems, but remain unsolved because no one entity has resources and interest to polish free software.

But given is an economic reason, it can be done. Apple built its slick UI atop BSD Unix to sell more hardware. If Google thinks it can sell more ads and services atop a consumer-friendly Linux, they certainly have the resources to make Linux friendly to Peoria.

Summary: they have no intention of moving up the ladder to heavyweight PCs. Instead, this is an "Innovator's Dilemma" move, creating a low-end mass market product that will always remain economically unviable for Microsoft.

Agree that we're stuck with variations on the x86 for the foreseeable future.

kievete: I run a mixed server environment and was woken up (again) last night to fix the Windows box (again) because Windows does not have some features already available on Linux. Do I have other issues with the Linux boxes? Sure. But I have *far* less babysitting with the Linux servers.

In regards to "innovation", this is a vague metric. Most ideas are refined from some other project so no OS project can be described as "more" or "less" innovative.

Stability isn't an issue for any of the OS's on topic. The rest of your comments are not accurate.

Microsoft is in a very weak position in this netbook space. XP is an orphaned project yet it's the only one that can possibly run on netbooks. Yes, they have WinCE. But it's not an XP desktop at all. Getting a project ready for netbooks would be a huge commitment and years of work at Microsoft's scale.

Is it "the end of Microsoft?" No way. Are they being slowly encircled? Yes.

Does Google have an automatic win with this one? No. Especially if they are using their phone OS. It's a Linux kernel, but nothing at all like the typical Linux distro.

1 – Point taken, but Linux covers a wider range than just chasing Windows. It can run embedded wonderfully (and does, in many routers, small devices, etc.) Windows CE doesn't even count here for competition. Linux also runs on supercomputers/clusters well (see all the slashdot jokes about Beowulf clusters). Linux covers a greater range of hardware in general, so it isn't just chasing Windows in that regard (hence Google chose it for what will initially run on small devices, but ultimately targets real PCs).

2 – NT architecture, at the kernel level (a microkernel as opposed to a more monolithic kernel like Linux) is indeed solid. However, it is in no way more advanced than the Linux kernel. Tickless running, O(1) time scheduler, FS support for vastly more filesystem types, faster FSs than Windows, etc, etc. If anything Linux as a kernel has gained an edge technology-wise, because the development of it is more fluid.

As for .NET, that has nothing to do with the kernel (unless you are playing with their experimental OS that essentially has a .NET VM embedded in it, but they aren't shipping that). .NET is a VM, a programming framework, and a set of tools. It's decent, and I really like C#. But I can develop for it to great effect on Linux under Novell's Mono. In fact I do, and Mono generally supports more languages, more platforms, and has additional APIs that are more comfortable than WinForms to create applications in. Again, Windows has very little if any advantage, and this is in the application layer.

3 – Microsoft Office applications are indeed featureful and would be hard to create elsewhere. They're also bloated, have generally awful UI design, are buggy, and full of security holes. You can get used to any number of other office/productivity apps that are free and divorce yourself from MS Office without any material hit to being able to get *real work done*, even in a commercial setting (like, say, where I work we've already proved this is true).

4 – If you know anything about what Linus and the kernel developers choose to work on and their criteria, you would know they don't spend much time thinking about chasing anybody, much less Microsoft's OS development team. They have their own goals, feedback from users/companies using Linux, and plenty of academic theories they try out in branches to prove new concepts. In the application space on Linux, they tend to do a lot more tailgate chasing, but definitely not in the kernel.

As for no innovation? I mentioned a couple of OS features that Linux has that I don't think the NT kernels have, and they're all recent developments (in the last 5 years). OS kernel development of necessity is slow, based on small advancements, and when there are big changes they are vetted over a period of potentially years to determine rock-solid stability. The same goes for the NT kernels, if you haven't noticed (the kernel in Vista vs. XP, for example, is probably not very different, and for good reason).

-

My point is, all of the "Linux development model is awful" or "Linux just chases windows" arguments are factually on very shaky ground. I have developed both applications from high-level (.NET-style and web apps) to systems-level (driver-ish to bare-metal computer graphics applications) software for Windows XP/Vista, Linux of various flavors, and Mac OS and Mac OS X, and I also have my own operation systems toy project that I work on from time to time to keep my mind tuned to how to deal with OSs properly as a software developer. I can conclusively say all this subtle and not-so-subtle Linux bashing in these comments are unfounded.

Google isn't stupid. They chose Linux for good reason. If I were working at Google, I absolutely would have advocated Linux. Kudos to them for choosing an open-source kernel, and contributing to it, because that makes all of us better off in the long run.