Talk:Comparison of operating systems

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Computing / Amiga / Software / RISC OS (Rated Start-class, Low-importance)
WikiProject icon This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Start-Class article Start  This article has been rated as Start-Class on the project's quality scale.
 Low  This article has been rated as Low-importance on the project's importance scale.
Taskforce icon
This article is supported by WikiProject Amiga (marked as Mid-importance).
Taskforce icon
This article is supported by WikiProject Software (marked as Low-importance).
Taskforce icon
This article is supported by WikiProject RISC OS (marked as Mid-importance).

What decides whether an OS is worthy of being on this list?[edit]

There are plenty of major and very significant operating systems missing from this list, certainly from an historical perspective. I'm guessing that a lot of it was compiled by someone relatively young, which is fair enough. I think it's a worthy endeavor and would probably extend this list myself, but what goes on it?

For an OS to be distinct, it needs to be really distinct. For an example every should understand, take Microsoft Windows. There have been two versions; one was a GUI running on top of MS-DOS (up to and including Windows ME), and the other is Windows NT, which was a re-write (from 3.5 IIRC, up to the current 6.3 and 10) - Microsoft's commercial names aren't relevant.

To be noteworthy, you can't decide it based on peak installations or whether its current. For example OS/360 had a major impact on computing, but has been defunct for many years and as it only ran on mainframes (realistically) the installed based must have been in the thousands.

Of the top of my head, some major operating systems missing include CP/M, Multics, VMS, Wang VS, MVS, VM. Others would include AppleDOS, TRS-DOS, OS-65, OS/9, Mirage.

I'm only counting systems that are independent of the hardware and run on more than one manufacturer's kit. This is a pretty arbitrary distinction. I've also not added any major RTOS families (e.g. OSEK/VDX). And then there are problems like Digital Research's GEM. This GUI/Application environment ran on top of DOS in the same way as pre-WindowsNT Windows did, so how can you call one an OS and not the other? And where does X-Window fit in (the same sort of thing running on top of *NIX)? And Android is just a GUI running over Linux.

If I was to spend any time expanding this, a consensus on what counts as an OS is needed first! (talk) 16:33, 14 April 2016 (UTC)

Why is Android listed as open source?[edit]

I'm fairly certain Android is not open source. I know Google claims they are but that does not make it so. — Preceding unsigned comment added by (talk) 05:42, 7 February 2012 (UTC)

-- "Open Source" is a practical marketing term. I think that you and they are thinking of "Free Software" ("Not to be confused with freeware, open-source software, Free and open source software, or free software license.") Android is not Libre Software, Open Source is not "Libre Software". Open Source is little more than proprietary code generously shared at the moment for the purpose sole benefit of the party that created it. Any other benefit to anyone else is purely coincidental.

Almost all open source software is free (libre/as in freedom) software. Those terms are 2 somewhat different approaches covering almost the same set of software projects, though I would rather advocate to the free software ideology. Read this by Stallman, father of free software, regarding open source. Having said that, the very core of Android is released under the Apache license and that makes it both free/libre and open source; however phone manufacturers usually put proprietary drivers and non-free stuff on top of Android. There are some freer alternatives based on Android, like Cyanogenmod or Replicant. See [1] for more information. --Isacdaavid (talk) 20:37, 9 December 2012 (UTC)

Android IS open source. The source is available here: (at time of writing). (talk) 16:33, 14 April 2016 (UTC)

Windows IA-64 Support[edit]

I know that former versions of the consumer Windows operating system were ported to IA-64 (Itanium). However, that support was removed in Windows Vista. Since this article seems to deal with the most recent releases of operating systems, should that processor be removed from the list? (The most recent release of the Sevrer edition still has Itanium support, so the Windows Server entry wouldn't need to be changed.) 16:36, 13 April 2007 (UTC)

Yeah, makes sense. be bold and go ahead, and explain it in the edit summary. -- intgr 20:28, 13 April 2007 (UTC)

Inclusion of VMS[edit]

In Windows' Talk, VMS was mentioned as - principally - being a precursor (to Windows). There are already pages to OpenVMS and FreeVMS but I don't know enough of the subject to add it cojently... MonstaPro 23:24, 15 April 2007 (UTC)

Vista? XP? Pre-XP?[edit]

There seems to be lots of conflicting information for the Microsoft Windows row... How much distinction are we making? Firewall gets an "XP or later" tag and an "NT-Based systems" tag, but wouldn't those make sense to be cut being as you combined the rows? Also, security vulnerabilities points to an XP page, but pricing discusses Vista. I feel it should be either all Vista or add more lines for the different versions - presumably the reason for keeping XP around is how many people are running it, but isn't this meant to be a technical article? I changed the firewall and security link, but I don't know what else may need to be changed. EB0und 18:58, 16 May 2007 (UTC)

Costs of purchase of free software[edit]

OS/2 is proprietary so you are not allowed to sell copies of it. Most likely you don't even have the source an therefore the ability really copy it or to maintain it. The fact that you can download most free software at no additional costs doesn't mean, free software is gratis per definition. It can be sold at any price. Otherwise it would be proprietary. So {{free}} is just plainly false in this regard. It can only fit to proprietary products where the proprietator has set this price (zero). --mms 14:22, 21 July 2007 (UTC)

I don't understand your point. I don't see that the article equates free software with zero-cost anywhere, or asserts that one implies the other, it simply lists the official price(s) of the software from the publisher/main supplier. The {{free}} and non-free coloring was unnecessary and POV. There are many different prices and sometimes the same product has different prices for different versions (some even have free limited editions, or are free for students, or other variations): why should zero-cost be singled out among these? NicM 13:47, 22 July 2007 (UTC).
There is no official price for published free software. Even if there is one creator, he cannot establish the »official« selling price after he sold (or gave away) the first copy. After there is one licensee, there can only be selling prices but no official price. The coloring of the boxes can't be a POV ─ it helped to get an overview. I suggest to step back to the templates and use {{any}} for free software and {{free}} for freeware. Even if I must deflect opposition in this regard I insist that one can't impute free software to cost zero. We can write »any« or similar expressions or we can leave this field empty for free software as this category doesn't fit to it. And BWT, if you want one example of an »official« price for free software which isn't zero, take a look at the online store of the Free Software Foundation. --mms 14:21, 22 July 2007 (UTC)
You are missing the point. Free software can be sold for any price or for multiple prices from different suppliers, but so can anything - Windows has different prices in different countries, and copies on ebay often go at prices different again. However, each of the operating systems listed, free software or otherwise, have a single primary publisher, and it is their official price which is listed. The exception to this is Linux, and it already points to a separate article covering the many distro publishers. NicM 15:26, 22 July 2007 (UTC).
Non-free software can't be resold. Almost ever you don't even have the source code. So you haven't really ownership in the first place. I think what you mean is to just resell your single binary copy. This isn't distribution at all and especially not commercial distribution. --mms 16:03, 22 July 2007 (UTC)
That is not true, there are many people reselling commercial software, they are called resellers. Source code has absolutely nothing to do with it. NicM 18:24, 22 July 2007 (UTC).
In other words, just because I can take FreeBSD and resell it at $200, doesn't mean we should list its price as $200, or that the price column is meaningless. I could equally well come to an arrangement with Microsoft to sell Windows at $900 (and OEMs do come to agreements to sell Windows at lower prices with hardware). The official publishers are obvious - the FreeBSD project and Microsoft - and it is their direct prices that we list, not mine, not yours, not Dell's or anyone else's. You could make an argument for removing the price column entirely, but not for treating free software specially. NicM 15:31, 22 July 2007 (UTC).
FreeBSD is free software so the price could be any amount. The situation with FreeBSD may look more concise than with GNU/Linux but there are forks of FreeBSD, too. It is perfectly legal to sell thousands or millions of copies of FreeBSD at any price the market allows you. Your version wouldn't be less official than the one of the FreeBSD Project. It would be the same software with the same licence. This separates free software from proprietary software. --mms 16:03, 22 July 2007 (UTC)
You are still missing the point. It is legal to resell FreeBSD at any price, but it can also be legal to resell any software—free software is not special in this regard, if I reach agreement with Microsoft I can resell Windows at a higher or lower price. The point is that you are a reseller, you are not the primary publisher of FreeBSD. If two organisations did produce versions of FreeBSD, both with significant use and both called FreeBSD with different prices, then we would have to note that or list them seperately (like we do with Linux), but that is not the case. NicM 18:24, 22 July 2007 (UTC).
To state it more clearly, if someone did decide to publish their own version of FreeBSD for $100 a pop, it would be a different FreeBSD and would either require it's own line in the table or be listed on a separate page (like Linux). Because the license means that someone can do this, does not make the FreeBSD Project's FreeBSD has "any" price, it means that when someone does, we need to list them if they are significant enough to appear. Note that PCBSD has its own line, and OpenBSD is separate from NetBSD. NicM 18:39, 22 July 2007 (UTC).
The important thing is the distinction between reselling and republishing (ie, publishing a customized version). Reselling is irrelevant for this. With republishing, my point is that just because free software allows publishing customized versions at any price, does not automatically mean that the price is any. The table refers specifically to FreeBSD as published by the FreeBSD project, it doesn't mean FreeBSD published by anybody at all. If anyone else published a customized version, that would be a different table entry. Do you see my point? NicM 18:50, 22 July 2007 (UTC).
Also note that free software is not special in this regard, other than that it makes forks easy. Just because in theory I could sign an agreement with Microsoft to take their source base and build NicM's Windows and sell it for $9000, it doesn't mean Windows has "any" price, it means that Microsoft Windows has Microsoft's price, and NicM's Windows has my price. Once I stop being a reseller and become a republisher, then a new line in the table is needed. NicM 18:58, 22 July 2007 (UTC).
Oh, and the colouring of boxes can be and was POV. Why are we only highlighting an overview of free/non-free? Lumping all the varying non-free costs together is POV - it isn't as simple as a non-free/free split, and why should free get special treatment? NicM 15:37, 22 July 2007 (UTC).
The different values of the costs can be grouped to zero, non-zero and any. The question whether it is free or proprietary software is a totally different and is answered in Preferred license. A color scheme would be helpfull for this column. Especially the entry of Mac OS X (Server) is misleading as it is now. --mms 16:03, 22 July 2007 (UTC)
No, "any" is not appropriate here, all of them can have any price. As I have already explained, lumping the dozens of prices and variations under "non-free" and highlighting "free" is POV and unnecessary. NicM 18:24, 22 July 2007 (UTC).

Well, Linux can be sold for a price and some vendors do (Red Hat?)... Who is the primary vendor of Linux? Is Linux free? Is Torvalds the primary vendor? Is it Stallman? Or is it Red Hat, or even Shuttleworth? Not to mention licensing. The kernel may be under GPL, but what happens when we load a proprietary ATI driver into it? Furthermore, back to cost, what currency are we talking about here? Many of these things are listed on local sites as costing twice what we've specified in the article! I'm just putting some extra stuff on the table. I guess: What is the "Cost/Availability" column? A lot of these have specific dollar-amounts listed but there's no link to any pricing page telling us any of the prices - this makes it very hard to confirm and very hard to contest the accuracy of pricing in Wikipedia's list.

I propose: throw away the "Cost" part and keep the "Availability" part. If it's available from the vendor for a fee then mention that it is available "For a fee", do not mention how much that fee actually is. GPL software is always free, regardless of the primary vendor's charge (because the purchaser is permitted by the license to redistribute it) - is that not correct? (talk) 03:39, 3 January 2009 (UTC)

What about user rights management[edit]

It seems, after installing and using basically every MS operating system since Windows 3.1, that Windows 2000 is the only MS operating system in which user rights can be flexibly defined, just as on any Linux system. I.e.: administrators can define for ALL directories on a machine which users have reading, writing and executing rights. This has been made much more "fool proof" (or "illiterate user resistant") since XP, but this has been achieved by eliminating all flexibility to give users for instance "read only" user rights for other users document root subdirectories. To me it is unclear why MS didn`t include comparable functionality at least in the "Pro" versions of XP and Vista. Thus keeping W2k still the only viable option for shared workstations... —Preceding unsigned comment added by (talk) 06:08, 28 April 2008 (UTC)

Linux vulnerabilities[edit]

The article represents linux as a whole os but the listed vulnerabilities are only about the kernel. That's unfair for the other OSes. I would suggest that there are no vulnerabilities listed for linux and a link to an explanation why to be added, or get rid of the whole vulnerabilities column anyway cause it's useless. It is not feasible to count vulnerabilities for what is called GNU/linux cause there is no standard distribution. Well done. (talk) 22:16, 11 September 2008 (UTC)

Comparison of Free OSes?[edit]

Just wondering if there should be an article that compares free OSes, rather than putting them in with the big boys. Because right now, the table cannot sort by free (since F is not a number.) Or am I missing an article that shows the comparison of free OS Software that I overlooked? (talk) 12:18, 16 October 2008 (UTC)

You may be overlooking the "sort" button on the "cost" column of the table -- clicking it sorts "free" ones together.
Also, you may have overlooked the "See also" section of this article, which includes a link to comparison of open source operating systems -- is that what you wanted? -- (talk) 20:29, 6 November 2008 (UTC)

Known unpatched vulnerabilities out of date[edit]

Most of the links have a different number than listed in the article. Aside from the fact that comparison of the numbers is meaningless for systems that have different security research focus, outdated information are even worse. I know that this was discussed before, but this columns should be removed altogether, already for the reason that they do not get updated regularly. (Even of the popular ones, none was correct. Windows, Mac OS X, and Linux are still wrong. I fixed FreeBSD by changing it to "unknown".) (talk) 14:49, 15 December 2008 (UTC)

Known unpatched vulnerabilities should be removed[edit]

  1. As noted, most of them are out of date.
  2. Most of them are closer to the number of Secunia "Advisories" than they are to the number of Secunia "Vulnerabilities" - for instance Linux says "13" whereas in Secunia we see "10" "advisories" and a far greater number of "vulnerabilities"... The information in Wikipedia absolutely does not correspond to the information given by the source.
  3. As noted, the Linux data is only accurate for the Kernel - and only for one specific kernel version - not everyone is using it; remember that usually you need to turn the system off (only) to update the kernel (or its drivers) and many Linux machines will be up and running for far more than the lifespan of a kernel version. They will suffer from different vulnerabilities. I imagine that it's the same for, say, BSD boxes. Are we talking about Operating systems people can start using now, or what they're already using?
  4. Linux (and BSD etc) distribution vendors may patch their kernel versions, blocking vulnerabilities at different rates than what Linus Torvalds does. Hence their binaries may have more or less vulnerabilities than what the vanilla kernel source does.
  5. As noted, the Linux count is only against the kernel. This seems unfair as the counts for Windows (for example) may be against removable components (such as services that must be switched off to enhance security or registry settings that must be tweaked)

So basically:

  1. The counts are not current and do not reflect the state of real systems
  2. The counts do not reflect information from the reference!
  3. The counts are not measured or gathered consistently; they measure different kinds of information

I will (try to) remove them for the time being. If I'm making a mistake then I apologize, in that case please revert and update. (talk) 03:22, 3 January 2009 (UTC)


The fact that footnote numbers are repeated within this page (e.g. there are three footnote #8s) causes many of the links to go wonky. I may fix this when I have more time, otherwise someone else can. EvanED (talk) 21:28, 14 April 2009 (UTC)


In the security table, a division should be added called "user permission restrictions". This vital part is entirely not described; perhaps that part of the section: Operating_system#Security can be moved to this article. According to this site, windows versions even completely lack it, meaning that any user can wite or overwrite almost anything ont the computer and install software.

Focus on personal computers[edit]

The article is focused too much on personal computers. For example, the diagram only shows a comparison for these machines. The story is different for servers, ... Also, any computing device can have a OS installed, including eg a playstation 3, robot, ...

  • Yes, agreed. Now, what to do about it. Propose to make this a more general page and split out pages for comparisons for desktops, servers, user appliances (consoles, phones), embedded, realtime, and industrial uses. What else? (talk) 10:35, 19 October 2009 (UTC)

Split RISC OS branches?[edit]

RISC OS has branched into two differently licensed branches with somewhat different features, might be a good idea to split them into the ROL and CTL/ROOL branches, which diverged after the Pace/ROL 4.02 release. Bhtooefr (talk) 17:25, 6 August 2009 (UTC)

Support: The current table layout doesn't permit use of both {{Nonfree}} and {{Free}} unless separate cells are used. --trevj (talk) 09:44, 3 March 2011 (UTC)
Yes check.svg Done with this edit.
  • Prices still need checking/changing;
  • RO5 coloured cyan in 'Cost' column, which seems to correspond to "Free" (non-commercial use noted);
  • 'License' links to RISC OS Open and states "Shared Source" (but is not Microsoft's Shared source).
Improvements welcomed. --Trevj (talk) 14:54, 1 July 2011 (UTC)
PS I've just added Pace for RO4. --Trevj (talk) 15:09, 1 July 2011 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── I guess Comparison of operating systems#Technical information is next, i.e. splitting 26- and 32-bit. --Trevj (talk) 15:02, 1 July 2011 (UTC)

Rowspan doesn't work properly in sortable tables when the rows are split by the sorting. Click the column headings in [2] to see the problem. I have replaced rowspan by repeated cells in [3]. You are free to add something to the cells to distinguish them. PrimeHunter (talk) 02:05, 12 July 2011 (UTC)
Face-smile.svg Thank you for correcting the issue I'd introduced. Sorry - I'd missed the note at WP:TABLE. Discussion of Bug 8028 indicates there could be a workaroud available at some point. --Trevj (talk) 04:55, 12 July 2011 (UTC)


Derp. An operating system costs $100. That's nice. What currency? I don't mind it being in US dollars, I mean it's hard to justify why it should be in any particular currency, but US dollars seems to be the most universal.... But nevertheless, it needs to be stated. Either every entry should say $100 USD or the column should be called "Cost (in USD)/Availability" (talk) 11:23, 28 February 2010 (UTC)

Windows NT Native API[edit]

This page lists both the NT native API and Win32 as 'native' APIs of Windows NT, where in reality Win32 is implemented as a subsystem. References to this are numerous throughout the internet but here is an example.

--Mappy wp (talk) 02:20, 12 June 2010 (UTC)

Modern OS Only Option[edit]

Many people coming to the page are only interested in comparing current OS to decide which to use. However, because all of the comparison tables are multiple screens long and compare every OS that has existed, they find the page to be mostly useless. We need a way to show only current information for those who want it, in order to increase utility. A method for allowing the viewer to sort information and suppress what they don't need would also be good.

This is a general comment applicable to all comparison articles. —Preceding unsigned comment added by (talk) 22:39, 15 October 2010 (UTC)

No info about the HURD[edit]

I couldn't find the string "hurd" in the article, although it is certainly an OS and has an article in the wikipedia.

Maybe to get on this list there has to be a certain number of users?

Son of eugene (talk) 04:41, 4 November 2010 (UTC)

z/OS pricing, joke?[edit]

Was somebody joking about 130 a month? 30 thousand a month, maybe, and that doesn't include compilers or anything else. Maybe somebody knows actual numbers and will tell us, but nothing IBM sells on the mainframe is 130 a month, certainly not z/OS since it runs on boxes costing a million dollars or more. Even the z/PDT (software alone!) is 25,000 minimum one-time charge. This number should be verified or removed. Cheers Gavin (talk) 18:55, 10 April 2011 (UTC)

BSD is an OS X predecessor.[edit]

Don't you think that BSD should be listed as an OS and iOS predecessor, since Darwin and some parts of NeXTSTEP derived from it? --Isacdaavid (talk) 21:36, 9 December 2012 (UTC)

I would rather say that it is a mix of a modified mach and BSD. the system use mach as it's base kernel (the servers run in kernel mode) with BSD kernel as a server. Even if this is BSD API which is available, and things FS and things like file systems are in BSD part. You may pay attention to other things like the executable format which is mach-o.
So iOS is rather mach based which is indirectly BSD in itself.2A02:8422:1191:6E00:56E6:FCFF:FEDB:2BBA (talk) 22:05, 6 April 2013 (UTC)


while I was wondering if systems calls are faster in microkernel, I didn't find anything.

What about writing the average system call time for each OS (where system calls exist). It may also answer others related questions2A02:8422:1191:6E00:56E6:FCFF:FEDB:2BBA (talk) 20:41, 6 April 2013 (UTC)

iOS's package manager(s)[edit]

iOS has 2 different package managers that are responsible for installing apps and files. Whenever an app is installed (through iTunes, the App Store or other means), the IPA file is sent to installd, the Installer Daemon. On jailbroken iDevices, the package manager is dpkg and, by extension, apt-get. DEBs are installed onto the iDevice through those programs. Either can be interfaced by a terminal application, SSH, or through the GUI frontend Cydia (I don't know if installd can be interfaced directly by the user in such a way). It has been this way since iOS 2 for installd (I think), and since iOS 1 for apt-get/dpkg. Windoze96 (talk) 01:25, 27 November 2013 (UTC)

An interesting list[edit]

An uninteresting collection of disassociated, unrelated bit and pieces of information comparing apples, to oranges, avocados, aardvarks, orangutans, and Volkswagens. Useful in what manner? Was somebody bored? These are THE Operating Systems I get to choose from? Should have been CSD'd a long time ago. :- ) Don 07:26, 25 October 2014 (UTC)

Firefox OS missing?[edit]

Firefox OS seems to be missing in the table or does it for any reason not belong onto this site? Retrovertigo en (talk) 03:30, 28 November 2014 (UTC)

I don't want to start a huge debate but...Linux is a kernel...[edit]

...Not an operating system. And it doesn't make any sense to say the predecessor of "Linux" is the "Linux Kernel". I'm going to change it to "GNU/Linux" because, while yes I'm aware that much fewer people say that and it's a mouthful, it's the more technical/accurate term. And this is an encyclopedia after all. I'll also change predecessor to "none". For the same reason, I'll also change Android's predecessor to "None" and GNU/Linux's price to "Varies".

Doing this will also help with consistency with Android and ChromeOS. They are Linux-based too, so why should they be in the table while other Linux-based systems aren't? My edit solves this since if we specify that "Linux" is "GNU/Linux" then it makes sense to differentiate between Android/ChromeOS and all the other Linux-based OSes, since the Google ones aren't based on anything from Gnu.

I agree with the guy above me that we should include FirefoxOS as well.

Also, is it totally correct to label ChromeOS as proprietary? Isn't all of the core stuff carried over from ChromiumOS still open source? — Preceding unsigned comment added by (talk) 00:44, 18 July 2015 (UTC)

Also I alphabetized the list. Many things were out of place. I put all the BSDs together at the "B spot". — Preceding unsigned comment added by (talk) 01:28, 18 July 2015 (UTC)

Final Note: In the last two tables, about security and commands, I used the term "Linux-based" instead of "GNU/Linux" in order to include the Google ones. — Preceding unsigned comment added by (talk) 01:44, 18 July 2015 (UTC)

So how is the z/OS nucleus different from a "kernel"?[edit]

How does the z/OS nucleus differ from what's described in kernel (operating system)? As far as I know, it's code running in privileged mode, with user-mode code entering nucleus code through a privilege-changing instruction (SVC or PROGRAM CALL). Guy Harris (talk) 05:40, 17 January 2016 (UTC)

What is a "Predecessor"?[edit]

The "Predecessor" column in the "General information" table seems to be used for several different purposes:

  • For AIX, HP-UX, and IRIX, the "Predecessor" is the AT&T UNIX release from which the OS is derived;
  • For Android, it's the kernel it uses;
  • For various BSDs, it's the BSD from which it forked off;
  • For IBM i, it's the name of the OS before it was changed to "IBM i" (i.e., OS/400 isn't an earlier OS, it's an earlier name for the OS)
  • For MVS, it's the OS from which it took a lot of its code and APIs, although its immediate predecessor was OS/VS2 (SVS), which was a descendant of OS/360 MVT;
  • For OS/390, z/OS, z/VSE, and z/VM, it's again the name of the OS before it was changed;
  • For OS/2, it's the OSes from which it took some ideas;
  • For Windows NT, it's 1) the OS from which it spun off (NT was originally going to be a New Technology version of OS/2), 2) an OS that came out after the first version of Windows NT but that contributed to NT 4.0 (Windows 95 came out 2 years after Windows NT 3.1), and 3) an OS that I guess represents the roots of classic Windows;
  • For ReactOS, it's the OS that it's trying to clone;
  • For HP NonStop it's, I guess, what the OS was called before the "OSS personality" was added;

so what exactly is it supposed to be? Guy Harris (talk) 05:17, 20 January 2016 (UTC)

Indeed! And how exactly is Windows 3.x supposed to be a predecessor of OS/2? Did it fall through a time-warp? It was actually based on MS-DOS if any one thing, and VMS if you want to see where the ideas came from. Windows 1.0 could arguably have been based OS/2 Presentation manager. I was there. (talk) 15:57, 14 April 2016 (UTC)

Windows MTP[edit]

Windows 7 (at least) supports MTP (TC) 09:21, 3 August 2016 (UTC) — Preceding unsigned comment added by 2001:44B8:802:1100:1C2D:9220:DB96:30FD (talk)