Talk:Microsoft Windows

From Wikipedia, the free encyclopedia

Jump to: navigation, search
Former good article Microsoft Windows was one of the good articles, but it has been removed from the list. There are suggestions below for improving the article to meet the good article criteria. Once these issues have been addressed, the article can be renominated. Editors may also seek a reassessment of the decision if they believe there was a mistake.

Archives
Archive 1 Archive 2 Archive 3
Archive 4 Archive 5 Archive 6
Archive 7 Archive 8 Archive 9
Archive 10 Archive 11 Archive 12
Archive 13 Archive 14 Archive 15
Archive 16 Archive 17 Archive 18
Archive 19 Archive 20 Archive 21

Contents

[edit] No criticism section

I can't believe there isn't a criticism's section

We've been over this before. There aren't any criticisms that apply to all versions, so we link to Criticism of Microsoft Windows, a disambiguation page in the See Also section. Josh (talk | contribs) 14:24, 22 November 2007 (UTC)
Oh that's utter pure bollocks. You work for Microsoft? You must. No products in any category of life on this planet have been so highly criticised as Microsoft's. None. You tell the world 'we've been over this before'. Guess what? It. Doesn't Wash. I hereby accuse you of being a dupe for the Microsoft propaganda machine. And likely paid for your efforts as well. —Preceding unsigned comment added by 90.5.128.188 (talk) 03:45, 12 July 2009 (UTC)
Of course, Microsoft Windows has been criticized numerous times. That's why we have and link to Criticism of Microsoft Windows. (Note that people keep trying to add information that has nothing to do with criticism to that page.) We just haven't documented any criticism that applies to every version. I tell "the world" that we've been over this before because we have. Could you please explain how having criticism in the articles for the versions they started in and not this generic article "doesn't wash"? - Josh (talk | contribs) 04:09, 12 July 2009 (UTC)

[edit] Windows 7 release date early?

I noticed that it says that there is going to be a new windows operating system, it doesn't have a name yet, but it has a code name, and it is Windows 7. Well the release date seems early, and I think it should be allot more than 3 years after vista.Foper (talk) 01:28, 16 January 2008 (UTC)

Microsoft says that they are "scoping Windows 7 development to a three-year timeframe", and that "the specific release date will ultimately be determined by meeting the quality bar." - Josh (talk | contribs) 01:37, 16 January 2008 (UTC)

I've heard 2010 from other sources, this seems valid as is.

However, shouldnt the following tag be added to the page?

123.243.27.198 (talk) 10:51, 29 January 2008 (UTC)

I've heard 2010 also. Does anyone know where Microsoft said it would be released in 2009? 24.151.249.73 (talk) 01:19, 9 March 2009 (UTC)

Its apparently going to OEMs and partners in very late august/sept then its going RTM, to business customers etc... in like october/november, with a consumer release in time for the christmas/new year (late dec - mid feb) period

Windows 7 will be released on October 22 ([1]) —Preceding unsigned comment added by 24.115.7.203 (talk) 18:46, 8 June 2009 (UTC)

[edit] Stable release

Shouldn't Windows Server 2008 be mentioned as stable release of Windows instead of Vista? —Preceding unsigned comment added by 83.7.14.207 (talk) 01:54, 15 June 2008 (UTC)

Yeah, good point -- done (diff). I've put both Vista and Server 2008 in the template, making a distinction between the latest client and the latest server release. -- simxp (talk) 03:48, 15 June 2008 (UTC)

Doesnt matter. They are the same codebase now. Server 2008 = Vista SP1. They are binary compatible. 134.36.92.18 (talk) 04:25, 29 April 2009 (UTC)

[edit] Just Wondering... .NET Framework

Shouldn't the .NET Framework be listed somewhere in here? I searched the page with Firefox and it came up with no results for .NET. Also, I'm wondering WHAT windows comes with the .NET Framework by default. 66.168.19.135 (talk) 14:22, 21 October 2008 (UTC)

From .NET_Framework#.NET_Framework_1.0:
"This is the first release of the .NET Framework, released on 13 February 2002 and available for Windows 98, NT 4.0, 2000, and XP."
IIRC, you would have received it as an "update" if you had XP back then, and probably received it OEM at some later point, including in XP SP2. Certainly shipped by default on recent and current XP and Vista machines. The "benefits" are listed here. You can sift and form your own conclusion. I came up with "useless bloat", but that's just a personal opinion. Cheers! Unimaginative Username (talk) 06:18, 23 October 2008 (UTC)
I see, because I was wondering where it came into play. On SP1, I didn't have it, so I was wondering when it came with windows. Even though it's useless bloat, some programs require it. For example, the CWCHEAT Database Editor. It's the same with the Java Runtime, some programs require that too. Though, both, I consider "bloat." This means that ReactOS doesn't need to have it packaged by default by using Mono or something. 66.168.19.135 (talk) 13:30, 28 October 2008 (UTC)

I think the only OS that definately shipped with .NET is vista/server 2008 (Net 3.5) and Windows Server 2003 R2 (As part of IIS). This version was 1.1 MS have a special page on thier lifecycle page saying how they will extend support for .NET 1.1 on Server 2003 until the end of life for that OS, even although .NET 1.1 on all other platforms is unsupported.

134.36.92.18 (talk) 04:27, 29 April 2009 (UTC)

[edit] 32 bit File Manager on Win3.11

Here's a citation for the line in the par. Hybrid 16/32-bit operating systems: "http://www.openwatcom.org/index.php/Exploring_Windows_3.x#Differences_Between_Windows_3.x_Versions". I couldn't add it because the page is semiprotected and I can't edit it (even thought I don't understand why I can't even if I log in with my account). Moreover, as long as we're talking about 32 bit disk access, this obviously can't be done through DOS that's a 16 bit OS. It's actually achieved with Win 311's VxD, that are 32 bit virtual device drivers.--Webwizard (talk) 09:21, 27 October 2008 (UTC)

It doesnt have a 32 bit file manager though. The only OS to have that was Windows NT 3.1, Nt 3.5 and Nt 3.51 —Preceding unsigned comment added by 134.36.92.18 (talk) 04:28, 29 April 2009 (UTC)

[edit] Citations needed for Security section

"Consumer versions of Windows were originally designed for ease-of-use on a single-user PC without a network connection .. Windows NT and its successors .. are not designed with Internet security in mind "

They why does this say "Windows 95/Windows NT The best client platforms on the Internet".[1]

What causal relationship is there between the design of Windows and the prevalence or lack of Internet usage. Unless you can provide verifiable historical citations as to their accuracy, they don't belong in a factual article.

emacsuser (talk) 20:32, 11 January 2009 (UTC)


I found this link that shows internet usage was less prevalent in the early 90s. http://www.allaboutmarketresearch.com/internet.htm 193.36.230.96 (talk) 11:20, 27 January 2009 (UTC)

Remember guys, correlation does not imply causation. --Scouto2 (talk) 16:51, 14 July 2009 (UTC)

In related logic that would mean that Windows got more secure as the Internet got more popular ? - yea ?? emacsuser (talk) 11:38, 11 September 2009 (UTC)

[edit] Programming language?

Why doesn't the infobox say in what programming language Windows is written in like the linux kernel article does? --BiT (talk) 03:54, 18 February 2009 (UTC)

[edit] Wine is Not an Emulator

The "Emulation Software" section refers to Wine and it's derivatives as emulators, when in fact Wine stands for Wine is Not an Emulator. http://winehq.org/myths#slow Cloud858rk (talk) 01:25, 7 March 2009 (UTC)

While the people that make Wine want to use the word "emulator" only for computer emulators (virtual machines), Wine does, in fact, emulate Windows. - Josh (talk | contribs) 03:07, 9 March 2009 (UTC)

Wine is an application layer, in the same way that the dos16, win32, posix and os/2 layers are in windows NT based systems (vista only features win32, all are present in nt4 for example) 134.36.92.18 (talk) 04:22, 29 April 2009 (UTC)

Emulate - To imitate the function of (another system). This is in contrast to simulate, which strives to imitate the internal workings of an other system. Wine is an emulator, because as well as providing an application layer (like Bochs does a BIOS) it also serves to support PE Files and mimic external interfaces of the Windows Operating System not adequately mirrored in Linux. Defintions are also taken from independant verifiable sources, not the producers. We don't want yet another Second Life ("videogame") cabel where the definitions is referenced and argued to the ends of the earth by sourcs all with a vested interest of bias. - Jimmi Hugh (talk) 19:30, 1 July 2009 (UTC)

Exactly. By definition, Wine emulates windows functions, the same way as DOSBox works. According to their website (which I read when my computer had Ubuntu on it), most programs are written to the lowest denominator (such as Windows 95, NT, 2000), and they use basically the same functions. So, Wine has those functions built in for the program to use. --Scouto2 (talk) 16:56, 14 July 2009 (UTC)

[edit] Vague?

Isn't saying that "Vista's successor, Windows 7, is slated to be released sometime between July 1,2009 and June 30,2010" a tad vague? I think that this should be deleted until we get a more definitive date range. BrownsRock10 (talk) 22:38, 26 March 2009 (UTC)

[edit] Citations

Under security, it links to footnotes 22, 23 and 24 as citations. None of them say anything about the subject. The only place the phrase "not until Vista was the default user not an admin" even appears, according to Google, is on this page and some pages that copied it. 192.139.30.1 (talk) 23:46, 7 April 2009 (UTC)

[edit] Windows designed without a network connection

[sarcasm] How dare you ask for citations. Windows was never designed for the Internet, it never happened, as Wikipedia says so and if Wikipedia says something, it must be true. Else there would be citations and evidence as to the opposite. It must of been on some parallel universe that I remember running Netscape and Eurora email on Windows 3.1, without a network connection. As to how we managed to boot into Novell Netware without a network connection, that never happened either. And if WinNT was not designed with the Internet in mind, what were the networking hooks doing in the OS? [/sarcasm] —Preceding unsigned comment added by Emacsuser (talk • contribs) 16:11, 25 May 2009 (UTC)

Umm there were a few years in there where there was plenty of networking and no one had any intentions of using Windows on Darpa Net or the Information Super Highway. Think pre-Bill Clinton/Al Gore. --Riluve (talk) 02:58, 27 July 2009 (UTC)

[edit] Windows 8

A job advertisement at Microsoft confirms that Microsoft will soon begin work on Windows 8 so this will need a page and section soon. http://members.microsoft.com/careers/search/details.aspx?JobID=524FE97F-DB02-4501-AD0E-8CCAF1719BB3 194.80.32.9 (talk) 00:45, 18 April 2009 (UTC)

There's no point in having a page about Win 8 before we know something more about it than the fact that it will probably exist at some point in the future. -- simxp (talk) 09:59, 19 April 2009 (UTC)
Early development work, feature exploration, mockup UIs and etc started on "Windows 8" (it has a name that is not "Windows 8") more than a year ago. When mainstream reliable sources start to discuss it there will be a place for it here. SchmuckyTheCat (talk)

[edit] Better Phrasing

  • "ME is also the last DOS-based Windows release which does not include Microsoft Product Activation."

I think this would be better phrased something like: "ME is both the last DOS-based version of Windows, and the last released Windows version not to include the Microsoft Product Activation, which started in Windows XP".

NONE of the DOS-based series has ever included it, and saying it is the "last to include it" is ambiguous as it suggests there will be a future DOS-based version. ME is the last. —Preceding unsigned comment added by 194.116.198.178 (talk) 11:46, 11 May 2009 (UTC)

What is this 'DOS based' nonsense? None of these systems are based on DOS and the NT family are largely compatible with DOS so what's the deal? You people need to leave this to the professionals and stop playing Computer Expert Sims™. —Preceding unsigned comment added by 90.5.128.188 (talk) 03:41, 12 July 2009 (UTC)
The Windows 95/98/Me boot loader, is, or is based on, DOS. - Josh (talk | contribs) 04:15, 12 July 2009 (UTC)

Josh is right - unless you are booting EFI on your Windows PC, it is "DOS Based". Unless you think it can magically get the memory map without calling a DOS INT15h - because it can't (and it does need memory). --Riluve (talk) 02:40, 27 July 2009 (UTC)

Sorry I may have misread something so I want to be clear -ALL PC Windows since '95 ('95/'98/ME/WFWG/NT/'00/'03/'08/Vista) all of them still use DOS functionality to boot. They use INT13h to read the drives, they use INT10h to display anything to the screen, they use INT15h to find out the memory map (including size), they use F000h to find ACPI and IRQ routing, they read ports 70h/71h and look for DOS values in CMOS - they boot in 16bit mode for god sake even when the hardware is 64 bit! --Riluve (talk) 02:54, 27 July 2009 (UTC)

There is a huge difference between doing the same thing DOS does and being based on DOS. - Josh (talk | contribs) 03:09, 27 July 2009 (UTC)

In some cases that may be true - for example, both Linux and DOS can read FAT. However, when you are calling DOS APIs - as basically everything I mentioned is - especially when you are calling the DOS equivalent of kernel functions - like getting the memory allocation table, it is impossible to say it is not DOS based. You might as well say that Linux drivers are not based on Linux. --Riluve (talk) 05:40, 27 July 2009 (UTC)

No...Linux drivers are not based on Linux. They are drivers for Linux. - Josh (talk | contribs) 06:06, 27 July 2009 (UTC)

You can draw all the semantic lines you wish - when a person designs a linux driver, they base their decisions on the established Linux API. They don't just randomly call routines with random parameters. --Riluve (talk) 06:56, 27 July 2009 (UTC)

I have removed this line completely because no one has any legitimate argument to support its claim (see DOS Patrimony below). That is, despite claims by the to the contrary, every x86 Windows that does not boot native EFI is based on DOS in a very critical way. This includes every official release MS have ever had for the x86 architecture, (Vista, 2003, 2008 - none - have a pure EFI boot support). Windows 7 may finally actually be the first x86 Windows to leave DOS behind - let us see if it happens. --Riluve (talk) 00:05, 7 August 2009 (UTC)

This is ridiculous. Being based on software means being based on it--actually taking its code and evolving it into the new software. NT was written from scratch. - Josh (talk | contribs) 20:41, 8 August 2009 (UTC)

No, it was not written from scratch. Beside marketing hype, do you have any evidence that it was written from scratch? I have shown the specific examples that show it has key elements based on DOS. If your argument were valid, you could also argue that free DOS and DRDOS are not based on PC/MSDOS - because they did not directly copy any MSDOS code. The argument is fundamentally flawed, however, because if a single line of code was used or not is irrelevant. The important/critical aspect are the specifications and functionality of MSDOS which are copied in FREEDOS and NT and thus makes them "DOS based" in an important and significant way. NT was written with the DOS specification sitting on the table - with DOS in mind.
--Riluve (talk) 21:48, 8 August 2009 (UTC)

[edit] Windows 7 Release Date

Windows 7 is being released on October 22.[2] Microsoft still hasn't made an announcement about prices. 24.115.7.203 (talk) 18:49, 8 June 2009 (UTC)jake

[edit] Windows 7 Release Date UPDATE

Microsoft has posted the October date on their Windows 7 site, it says :

When will Windows 7 be available?

Windows 7 will be generally available on October 22, 2009. http://www.microsoft.com/windows/buy/offers/pre-order-faq.aspx

[edit] Footnote 25

Footnote 25 is a 'plant' - it is phony, a fake. Footnote 25 is supposed to be a reference to how NT was 'designed for security' which of course is pure bollocks. NT is a makeover of Cutler's Emerald/Prism which in turn is based closely on his earlier VMS. The model Microsoft used has nothing to do with the model Cutler had in mind. NT and its successors have NEVER been certified as secure in an Internet environment. —Preceding unsigned comment added by 90.5.128.188 (talk) 03:39, 12 July 2009 (UTC)

[edit] Why is ReactOS listed as an emulator?

My understanding is that ReactOS executes Windows functions natively, without any translation or emulation.--Frozenport (talk) 00:32, 14 July 2009 (UTC)

It emulates Windows. - Josh (talk | contribs) 00:48, 14 July 2009 (UTC)

Then, any GNU/Linux is essentially an emulated version of Unix. —Preceding unsigned comment added by 190.234.26.32 (talk) 04:57, 2 August 2009 (UTC)

Plus One. ReactOS isn't an emulator, neither is GNU/Linux. 24.241.229.136 (talk) 22:52, 7 September 2009 (UTC)

[edit] DOS Patrimony

I removed the following claim which is shear non-sense: ", and was unencumbered by any Microsoft DOS patrimony"

Every x86 compatible versions of Windows -to this day- still has at least 2 direct and undeniable examples of DOS patrimony.

1- Windows can still only boot to a single fixed drive - device 80h (C:). It recognizes device 81h, 82h, 83h . . . etc, but even if they are bootable, it will ONLY boot to device 80h. Once it tries device 80h, it will never try 81h, or 82h. To grasp how lame this limitation is, compare this operation to removable media which is given device numbers starting at 00h (A:). If 00h is passed as the boot device, but turns out to not be bootable, it will try drive 01h(b:).

The only rational excuse for this behavior is that it is a limitation that DOS shared as well.

2- All PC windows still boot in real mode (16 bit mode) and read their media (device 80h, 00h, or 01h)in 16 bit mode and use the 16bit DOS interrupt INT13 to read the media.

3- Oh, I came up with a 3rd, Windows still gets its RTC from the location as specified by DOS and uses the DOS CMOS checksum to ensure the DOS compatible portion of CMOS is valid.

--Riluve (talk) 02:03, 27 July 2009 (UTC)

4- Sorry, came up with a 4th (technically you could consider this 2 separate things), PC Windows also requires the signatures _SMB_ and RSD_PTR_ to be located in E000:0 or F000:0 - when this area is already over-burdened. Considering NT, or 2000, or XP, or ME, or Vista can address at least 4G of memory, the ONLY reason it makes sense to use this specific 128k region, is because it makes the signatures DOS compatible.

--Riluve (talk) 02:14, 27 July 2009 (UTC)

6 or 7 or 8? Yes I came up with another and shame on me for not mentioning this 1st! Obviously I lost track of which dependency this is because many of my previous examples are more like 2 DOS dependencies. OK - even before I get started on this one I thought of another, but don't let me get side tracked - this is a doosie! The ONLY way for PC Windows to know the memory map (how much memory is available and which memory is valid) is through DOS interrupt - INT15h. Without that DOS compatible interrupt - NO MEMORY for Windows!

OK, I think I have made the case that all PC windows is still carrying around DOS baggage. Windows 7 may have an optional (and functional) EFI boot loader, but let them release it and test it first before they come back here with these claims of having left DOS behind. The testing is very important because Vista's EFI loader is not 100% DOS free, it still uses DOS INT10h for video! lord - see sorry it never ends!

BTW - I say PC Windows because Windows CE and Itanium Windows are very much DOS free.

--Riluve (talk) 02:28, 27 July 2009 (UTC)

Some of these are just aspects of the x86 processor; they have nothing to do with DOS. - Josh (talk | contribs) 03:07, 27 July 2009 (UTC)
Unencumbered by DOS patrimony means it wasn't based on--written from--DOS. - Josh (talk | contribs) 21:08, 8 August 2009 (UTC)
I have shown that it was based on DOS with clear, exact, verifiable examples. Your defense, without any evidence, is to say "no it isn't".--Riluve (talk) 23:18, 8 August 2009 (UTC)

Not a single thing I listed here is a limitation of an x86 processor that can run Windows. They are 100% DOS API limitations. In fact, the first operational code on the x86 -before MS DOS or PC DOS was written- only had one of these limitations. Please try to show that a single limitation I have mentioned is a limitation of a contemporary processor. I can show how they are all part of DOS. --Riluve (talk) 05:25, 27 July 2009 (UTC)

An x86 processor does not switch from 16-bit mode until the OS tells it to; it is impossible for an x86 OS to already be in 64-bit mode when it starts booting.- Josh (talk | contribs) 06:06, 27 July 2009 (UTC)

Incorrect. Well you are mixing ideas. Here is the important part - after the reset vector, the processor can (and usually will) be out of 16bit mode in less than 15 lines of assembly code. Obviously that will not "count" as an OS - its probably another 100,000 lines of code after that until there is an attempt to pass control to any OS. Here is the part you haven't considered - how do you expect the BIOS to tell the OS how much memory there is if the BIOS itself is limited to 16bit addressing? Because the OS HAS to get memory information from the BIOS, the BIOS HAS to be able to address all of the memory first. There is no point in waiting around, so that is the first thing it does - get out of 16bit mode. So here is the frustrating part, after all that work - 100,000 lines of code later, the BIOS will go back to 16bit mode just before it passes to Windows because Windows is ONLY using the DOS API. Unless, as I mentioned, it boots a full UEFI solution, like OS-X does. Only then does the BIOS not have to go back into a DOS compatibility mode before passing control to the OS. --Riluve (talk) 07:47, 27 July 2009 (UTC)

Oh sorry, I must have been tired. If I wanted to be more clear, I would have just mentioned that OSX does "boot" in native mode (which means 32bit for 32bit hardware and 64bit for 64bit hardware). EFI will long have switched to native mode and does not need to switch into DOS compatibility mode to boot OSX. However, EFI does have to switch into DOS compatibility mode to boot any currently released PC Windows. Because both Windows and OSx both run on the same hardware (processor) it is obviously not a limitation of the hardware in any way shape or form. --Riluve (talk) 15:32, 27 July 2009 (UTC)

This conversation is hilarious, considering NTOS boots on SGI Visual series, DEC Alpha, PPC, IA64, etc, where these DOS "dependencies" don't even exist. In any case, this has nothing to do with improving the article. SchmuckyTheCat (talk)

So are you arguing that NTLoader is not part of windows or that Ntdetect.com is not part of NTLoader? I mean you have to draw a line someplace. Look at the freaking file name - its a .com file. The ONLY thing a .com file is compatible with is DOS or CP/M! It is 16bit real-mode relocatable code without any fix-ups! Wait, it might be that you are claiming Windows is compatible with CP/M.

At any rate, you blithely say NTOS boots on all of these systems, but you ignore that you are talking about completely different versions of the kernel booting from different loaders. Sure, MS can package 5 different kernels into a single zip file and call it one operating system, but on the technical level is it a single operating system? If you argue it is a single operating system, then the x86 "permutation" requires these DOS calls in order to boot - no ands ifs or buts. --Riluve (talk) 04:35, 4 August 2009 (UTC)

You are simply, and fundamentally, mistaken. SchmuckyTheCat (talk)

Can you provide anything besides your opinion to back up your claim? I have provided like 12 very specific examples. Let me be more specific. Please try to refute these clear and specific points which show that you are wrong:
1-Ntdetect.com is an integral part of Windows for the x86.
2-x86 Windows can not function without Ntdetect.com (or a functional clone).
3-Ntdetect.com is designed to operate in a DOS environment.

If you can refute these basic facts, then we are in a position to philosophize further. If you cannot refute them, you fundamentally have not even the simplest of cases. --Riluve (talk) 23:33, 6 August 2009 (UTC)

SGI Visual Workstation 540 SchmuckyTheCat (talk)
1, 2-x86 Windows Vista/7 does not include ntdetect.com at all. - Josh (talk | contribs) 21:08, 8 August 2009 (UTC)
Thank you Josh - I will take this as a clear admission that NTdetect.com is a fundamental, integrated part of previous version of Windows AND is functionally a DOS program. As for Vista still requiring DOS support, I can only give anecdotal evidence, as my direct knowledge is a corporate secret. Vista will not boot without some parts of the DOS environment - unless some update has rectified the situation. You can verify this deficiency yourself though by buying an Intel motherboard that is "EFI Optimized" (which means no support for DOS) and trying to boot Vista. Windows 7 should work - thus the crux of my entire argument.--Riluve (talk) 23:18, 8 August 2009 (UTC)

I removed this nonsense "Me is also the last DOS-based Windows release which does not include Microsoft Product Activation.", because the best case that people have for this claim is because "they say so". --Riluve (talk) 23:55, 6 August 2009 (UTC)

Oh, sorry Schmucky, I completely missed your attempt to make a point there, i.e. the SGI Visual Workstation 540. In fact, this "example" you are trying to use actually proves my case - so thank you for bringing it up (but you should have done some more research before you did). Turns out, when investigating the NT/2000 bootloader, SGI actually ran into the case where Microsoft had mistakenly IFDEF'd DOS compatibility into the x86 architecture, making the same amateur mistake that Josh did earlier in this argument, i.e. assuming the DOS limitations were x86 limitations. They are not.

So as I specifically mentioned the requirements would be, SGI created a layer on top of the standard ARCS IPL interface, to provide NT with its required DOS interfaces - so that NT could boot. I will give you credit for this very good try though - it shows that you almost know what you were talking about. However, your inability to break any one of the 3 fundamental points I mentioned earlier shows that you are simply wrong.

Another interesting point is that SGI had to include a DOS virtual machine into the boot sequence so they could support PCI cards with x86 BIOSes on them, because - what do you know, they all run under the DOS environment. A standard x86 box does not include this DOS virtual machine, ONLY because it is exactly running a DOS environment, at least until right after NTdetect.com has done its job. And as it should finally be clear to you, NTdetect.com is a DOS program. Point-set-match.
--Riluve (talk) 16:27, 8 August 2009 (UTC)

Microsoft assumed the limitations it had written into DOS were limitations of x86? That's just silly. - Josh (talk | contribs) 21:08, 8 August 2009 (UTC)
Agreed, it was silly. But to be clear what they really did was not account for the idea that someone would boot an x86 without supporting DOS. Thus, they treated two logically separate ideas -DOS limitations/dependencies and -x86 hardware specifics with a single code base/build switch. If you think it is so implausible, perhaps you can explain why the 320 & 540 both had a HAL - no other x86 platform needs a proprietary HAL to run Windows.
--Riluve (talk) 23:18, 8 August 2009 (UTC)

Oh dang Schmucky, I have to take back the credit I gave you for having a clue. If you had read the article you linked to, you would have seen that SGI had to provide its own HAL, to make up for their lack of full DOS support in the boot process. Yes, ok, I forgot that detail to, but it strengthens my case beyond measure! Not that it needs any help. I thought this might require me to add a new stipulation to my 3 previous points, but looking back, I have it covered nicely in #3:
3-Ntdetect.com is designed to operate in a DOS environment.

So the HAL helps to provide the environment - in places that the ARCS-DOS compatibility layer fall short (e.g. ASL). OK - what else did I miss anything?
--Riluve (talk) 16:56, 8 August 2009 (UTC)

So, besides being factually wrong, you are stubborn, rude, and insulting. This conversation is not about improving the article. You removed one sentence, which didn't actually have anything to do with this conversation thread. I'm done. If you have something else to say about improving the article, you should start a new subject heading. SchmuckyTheCat (talk)
I have begged you to provide a sliver of evidence to support your claims, but you stubbornly refuse. You seem to think you can wave facts away with your mere opinion. As for rude, don't be upset that I responded to your behavior in kind.--Riluve (talk) 06:36, 9 August 2009 (UTC)

[edit] MS-DOS is merely a bootloader in Windows 9x --> not true

There's a sentence in the Windows 95 section of this article stating that MS-DOS would have been degraded to a mere boot loader for Windows 4 in the Windows 95 bundle. Around this time, there has been alot of research of this Microsoft-originating claim, and it was shown, that this statement does not hold any truth in it. Therefore, this MS-propaganda-influenced sentence bares substance and should be changed or removed.

The most obvious to see fact for normal end-users is that it is very easy to stop MS-DOS 7.xx showing its "Windows splash screen" at start up, so you could see the DOS drivers loading, boot to the COMMAND.COM shell and start the GUI by typing in "WIN" at the prompt, just like you did in previous versions. However, it could be argued (and was!), that this was a special mode for compatibility. It isn't, but it requires much deeper knowledge of the internals of the system to prove, that, even when the Windows 4 GUI is up and running, it fundamentely depends on DOS actively running "underneath". It has been shown by experts, that Windows 4 will fatally crash immediately (within milliseconds), if the DOS underneath would suddenly stop servicing the constant stream of Windows API requests, clearly contradicting the idea of DOS being nothing but mere bootloader for Windows. In reality, DOS is a vital part of this hybrid architecture and Windows 4.00 (Windows 95), 4.10 (Windows 98), 4.99 (Windows ME) could not live a single second without it. Actually, this was one of the many issues discussed and demonstrated in court in the Caldera vs. Microsoft case, and anyone interested in the technical details can find very in-deep discussions of this on the net. Why is it important to keep this fact straight? It is important from a historical point of view, because by - first - denying its existance - and later - by undervalueing the importance of MS-DOS in Windows 9x, Microsoft basically destroyed any market for alternative DOS operating systems, which offered additional functionality and would have been beneficial for plain DOS users as well as for Windows 4.xx users (more free resources, significantly better memory management, smoother multitasking, smarter command line tools, just to name a few). 84.63.86.38 (talk) 20:01, 27 July 2009 (UTC)

[edit] Windows 7 Picture Caption

Why did someone edit the caption of the Windows 7 screenshot in the infobox? Windows 7 Ultimate edition has a different desktop wallpaper than Starter edition. The caption should clarify the fact the screenshot is of Windows 7 Ultimate editon. See here. It has a picture of the Starter wallpaper, clearly different from the other editions, and the caption states "The Windows 7 Starter desktop, with final wallpaper that is unique to this version." This is from Paul Thurrot's "WinSuperSite", which is a source used multiple times in the article. It is a reliable source. Please clarify in the caption. Thanks in advance. This is TechOutsider. 70.153.241.200 (talk) 02:00, 30 August 2009 (UTC) —Preceding unsigned comment added by 68.154.249.132 (talk)

[edit] ReactOS

ReactOS isn't just aiming for XP; it's also aiming for Vista now too. 24.241.229.253 (talk) 01:49, 26 October 2009 (UTC)


[edit] Server 2008

The article states that the latest version of Windows Server is "Windows Server 2008". I always thought that it was "Windows Server 2008 R2". Should this be changed? Correct me if I am wrong. Twistor96 (talk) 20:50, 28 October 2009 (UTC)

[edit] First version of Windows to be a true OS?

I couldn't help notice this sentence: "Windows 95 also removed the reliance on the MS-DOS command prompt, making it the first true operating system in the Windows family." Um, what? Reliance on a command prompt means it's not a "true" operating system?? Althepal (talk) 05:51, 30 October 2009 (UTC)

I think the sentence is using "true" to mean "standalone": it's just pointing out that, unlike Win 3.1 & earlier, Win95 isn't just a shell on top of DOS underneath (though it still used DOS as a bootloader). In any case, it's been removed now. -- simxp (talk) 12:00, 1 November 2009 (UTC)

[edit] Article update + An objection

Aren't we going to update the "Windows History" section to include Windows 7? Besides, I'm not sure about Windows 8. The link mentioned above leads me to MS Careers, and it has no info about Windows 8. TKhaldi (talk) 13:14, 1 November 2009 (UTC)

[edit] reference in Security section isn't formatted correctly

It's showing up in the text as: <refSchneier, Bruce (2005-06-15). "Crypto-Gram Newsletter". Counterpane Internet Security, Inc.. http://www.schneier.com/crypto-gram-0506.html. Retrieved 2007-04-22. </ref>


Retrieved from "http://en.wikipedia.org/wiki/Talk:Microsoft_Windows"