Singularity

This post hardly dignifies a response, but here goes.  Microsoft is currently phasing out Win32/COM for application development in favor of WinFX.  Based on .NET, WinFX is the new application-level API that will address the concerns brought up in this thread.  Microsoft launched the initiative years ago with .NET.  Today, .NET is the predominant platform for new application development on Windows.  Therefore, Microsoft is well on its way in solving the alleged ‘crisis’.

To address the question of Windows in the embedded space, consider the success of Windows Mobile and XBOX.

Ok, before you guys try to tar and feature me let me make a few corrections. . .

booga,

You are correct when you said VMS and not Unix, I was thinking of something else. So, anywhere I said Unix needs to be replaced with VMS. Now, onto your other points.

The original VMS team that was hired tried to write something like NT from scratch. However, this would have prevented backwards compatibility so MS decided against that. Instead they took good size chunks of the VMS team’s code and merged it with some of their older DOS code. And yes, Microsoft does have a lot of money and market presence and they will be able to ride out the storm for a while. But it is going to be impossible to “Refactor” enough of the Windows code base to make it work. They sheer effort involved would take them the rest of this decade just to put a dent in it. Rather it would be easier if they just started over.

Ster,

I see your points but there is a huge difference between Vista and when OS X first came out. Vista is going to be late to the party even after cutting most of its features. Now, I realize that 10.0 wasn’t the best OS when it first came out but it did have the features it promised. And it has been steadily improved every year to two years. If I really thought there was a chance in hell that Vista would get a major update every 1.5 years like OS X gets then I might have changed the tone of my article.

Eron Wright,

Please don’t talk to me about the nightmare that .NET is. And as for WinFX, I will believe it when I see it. NOT when I hear promises of it for 3 years.

Here is a list of some embedded systems:

automatic teller machines (ATMs)

wearable computers

Out of the 20 on there you listed 2, game consoles and PDAs. Lets throw in cell phones too and that will bring the total up to 3. I think they have a ways to go.

REALITY!!!!!,

I didn’t mean the code looks the same as Unix I meant the functionality was the same, namely stability and security over speed and power.

And if you ever want to talk science with me let me know, I got a minor in Applied Physics.

And lastly:
I know I have done a good job if at least one reader tells me I suck. . .

James,
  You make good points, but our understandings still differ somewhat. 
1. Win16, which was the API for Windows 3.1 and earlier, only ran in emulation on Windows NT, just like MacOS X’s blue box. 
2. Although Win32 ran natively, it was a re-implementation and you’d often find API bugs on one platform that didn’t exist on the other (since it was largely a separate codebase.)  As well, NT didn’t support enough of the API to run most games when it first came out, although it improved dramatically with Win2K.
3. Win3.1/95/98/ME ran on top of a DOS-like kernel.  In NT/2K/XP, DOS runs on top of a VMS-like kernel.  Big difference for performance and stability.
4. WinFX isn’t going to replace Win32 overnight.  It may never.  (I have a lot of skepticism that Cocoa will ever “replace” Carbon, too.)  But it is the compatibility break that you’re looking for.
5. Carbon on MacOS X is like Win32 on Vista.
6. A rewrite generally just replaces bugs you know about with ones that you don’t, although only experience tends to teach you that.  It’s often one of the worst ideas in the world to replace old code with completely rewritten code.  One of the great strengths of open source is that replacement code tends to spawn, mature, integrate, THEN get bundled with a project, while with closed-source the “rewrite and ship” mentality is too easy.

As for your list of embedded systems, you’d probably be frightened to learn how many of them run a WindowsCE derivative these days.  When IBM finally deprecated OS/2 last year, a lot of banks and ATMs went Windows, for example.

A few notes:

1.) I don’t think the Windows NT history was cleared up quite completely yet. NT not only originated out of the MS-IBM split, it essentially evolved out of the OS/2 code and planning. You’ll note similarities, for instance, between the file systems in the two OSes.

2.) Lisa was not, as commented above, the hugely popular predecessor to Macintosh. It was a very unpopular product that was developed at the same time the Mac was being developed. Eventually Apple added a Mac emulation and placed it under the Mac name plate. James is thinking of the Apple ][.

3.) Someone mentioned MS buying a GNU/Linux distro. That might make sense, although honestly it’d make more sense for them to build a BSD-based OS, as Apple has done. MS has spent too much time attacking the GPL license for them to build on top of Linux without a ton of flak.

“Whereas when PC’s get too infested with viruses these days their owners just throw them out.”

Sure. Yeah. PC owners are too stupid to know how to prevent viruses and/or clean them up.

It’s really quite amusing to hear people criticize Windows who haven’t used it since ‘95, and who refuse to believe that there are millions of people using Windows right now! this very second! productively, safely, and securely.

Is is possible to envision a future wherein people continue to have a choice of operating systems? I would no more want a Windows-only world than a Mac-only or Linux-only world. Competition is the only force that reliably drives innovation.

Mr. Stoup,

Whew! Sorry to get you so riled up!

I gather I came across to you in my post as some sort of expert. I am not, nor do I deliberately present myself as such. I speak only from my own experience.

I work at a public library. I am not an IT professional. I did not recently take an introductory course in economics. I have been a Windows user for approximately 8 years. I am in charge of maintaining my library’s 18 workstations, for both public and staff. In 5 years we have experienced one (1) virus attack (due to a staff email) and one (1) fatal blue screen (due to third-party software). We run a mixed XP-2K system, with a Linux Red Hat server. I spend much more time dunning people about overdue books than I do in maintaining and updating our PCs.

At one point I became enthusatic about exploring the use of Macs in our library.  I have yet to learn anything from Apple to lead me to believe that they have any interest whatsoever in supporting their product in public libraries. I am now concentrating my search on several Linux distros, as I have found the Linux comminity to be very supportive, polite, and accessible—quite in contrast to many people in the Mac world.

I attempted to voice an opinion that a diversity of choice in OSs is a good thing. I take it that you do not share my opinion.

I hope I don’t come across as anti-Mac. OS X is a really good OS, one that more Windows users would be delighted to try, if they had an easy way to do so (IMHO).

Trying not to hijack the comment thread, but a really big barrier to a wider adoption of the Mac OS is the problem of older hardware. Again, I’ll speak only of my current situation—a small public library. We have several old PII’s in the basement that are ideal candidates for a light Linux distro. Half of our public workstations (now running 2K) will not be up to snuff to run Vista, so no point in upgrading the OS there—they also make good candidates for Linux installations.

None of these workstations will ever run any Mac OS, yet they are still too functional to throw away.

Windows is simply too hard to secure for (or should I say from) public users. Using Group Policy is a nightmare, and the PAC tools available are either too expensive or too unweildy. Thus, a Linix environment seems ideal for setting public user permissions.

Out staff workstations are less than 2 years old, so we will not be replacing them for at least 2 years. At that time, we might look seriously at Apple—except that, as I mentioned earlier, there doesn’t seem to be any interest on Apple’s part to put their machines in “government” for a discount, as they do now with education. For any public institution, price is pretty much the only factor that can be considered.

So who cares what kind of computers we have in public libraries? The people who use our public workstations fall into two groups: 1.) People who don’t have their own computer and 2.) People who can’t use their home computer, either because it’s broken or they’re travelling. In both cases, it would be a great oportunity for them to try out a Mac—and they might be disposed to buy or upgrade to one when the time comes.

Buying a new computer is a major purchase for the “average” consumer. I’m skeptical about recent reports of the number of “switchers” to Macs, mainly because it’s not really that easy to abandon Windows in favor of another system. It’s quite an investment in time and money for an established Windows user—remember how hard it is to change old habits, no matter how bad that habit may be. People still smoke, for god’s sake!

There is an incredible opportunity to bring Macs to a wider audience (including disgruntled Windows users) via public libraries, but without strong support from Apple, that will never happen—and Windows will continue to be the dominant OS for some time to come.