If you are doing a near real time system... Windows 3.11 might actually work rather well, so long as the application wasn't trying to allocate too many resource handles.
Actually, I wonder why MS wouldn't release a non-preemptive Windows, just for this purpose. It would be a lot more reliable for some applications.
If all you need is a *near* realtime system, and you really want to use WIndows for some reason, then an NT derivative will do you fine: Priority levels 16 through 31 of the NT scheduler are soft realtime. And I'm sorry, but there's no *way* that 3.11 is more reliable than, say, a stripped down Win 2000.
(Of course, if you need *hard* realtime, neither NT nor 3.11 will do you).
Solaris and those others, OTOH will happily run for months and years without requiring a reboot. I recently ran across a system at work (RedHat 5) that nobody bothered with because it always did it's job. When I had to go look to see what the problem was, imagine my surprise to find it running RH5. Everyone that knew the root password had either quit or forgot they knew it, it had been sitting there running for several years. Windows will NOT do that.
Accroding to Netcraft, the server out there with the longest uptime is fp002.crayfish.net, currently at 1817 days (~5 years) of uptime and counting; running -- Windows 2000.
In the job I described above, all the code was written for Qt, on Windows and UNIX, because it was simpler, cleaner, and more reliable than the conflicting Windows APIs.
In recent versions at least, Qt in Windows works by using the Windows native APIs. So you're argument kind of nullifies itself -- you're arguing that using the native GUI APIs in Windows is difficult, (and, of course, impossible in Linux, since X doesn't have any), but then you admit that when you use a toolkit like Qt, developing a UI is equally easy across both platforms. So... do that then.
However I have two points right back at you.... Even is IIS7 ran as well as Apache or better (which I doubt) Apache is vastly more secure.
You can say that again. Why, in the year and a half since IIS 7 shipped, it's been deluged under a mammoth 1 Secunia advisories, whereas Apache 2.2 has had a mere 5 (2 still unpatched) in the same period. Clearly, Apache is vastly more secure. You sure showed them!
KDE? more modular? Kparts means that you can include practically entire programs (spreadsheets, browsers, editors) inside other programs - how much more modular can it get?
You say that as if this is some magical, never-seen-before new technology in KDE4 that indicates an unprecedented level of good design and modularisation. This is nonsense. Component frameworks have been a standard part of desktop environments for ages. Gnome has Bonobo and now D-bus; Microsoft has COM and now WCF; etc, etc.
Neither are sandboxed and both run with the same privs as the browser AFAIK.
you know wrong. ActiveX on Vista runs in a sandbox with lower rights than the current user.
Actually, the GP isn't wrong. But you're not wrong either. The GP is right that ActiveX runs with the same privs as the browser, and you're right that ActiveX runs with lower privs than the current user (on Vista); because the browser *also* run with lower privs than the current user.
Only an Apple hater would think Apple would purposefully expend developer time just to break an open source project that undoubtedly sold more iPods. I'm afraid you're in denial. It is well known that Apple have consistently and deliberately added layers of hashes, obstufication, and DRM to successive versions of the iPod in order to try and stop it being accessed from third party players. For example, to quote from a BBC news story about it:
There seems to be no reason for this change except to break the functionality of alternative jukebox software. It will not limit copying or restrict attempts to strip digital rights management code from tracks. It will not stop people adding non-DRM files they have downloaded from the internet to their library. All it will do is stop the third party players working and force anyone with an iPod to use iTunes. Independent enough for you? The BBC are hardly "Apple haters". And crying "Why would they do this if it sells more iPods?" is short-sighted: iPods are a low-margin product; most of the money Apple makes from the iP+iT Ecosystem comes from the iTunes Music Store, which is locked to iTunes.
None of which I use, or intend to - in every case, there exist Free apps that are better (except for the Mono development stuff, which by definition is not Free). "Free". You keep using that word. I do not think it means what you think it means. To me, the OSI, the FSF, and the entire free software community with the possible exception of you, it means "released under a Free Software license". Mono is dual licensed by Novell, similar to other products such as Qt and the Mozilla Application Suite; the compiler and tools are released under the GNU General Public License (GPL) (starting with version 2.0 of Mono, the Mono C# compiler source code will also be available under the MIT X11 license), the runtime libraries under the GNU Lesser General Public License (LGPL), and the class libraries under the MIT License. If you're disputing that any of those licenses are free, take it up with the FSF. (The majority of apps I listed are also licensed under the GPL or other Free software license).
The only thing I can think of that you could mean is that you've been lead in by some of the "patent encumbered" FUD. In which case let me disabuse you: the only parts of mono which might even be theoretically patent-encumbered are the parts that are not ECMA-standardized; most notably the Windows Forms implementation. Now, which of the apps I linked to used Windows Forms? Yup: None -- they all used Gtk#. So even by that definition, none of the apps are even in the slightest "non-free".
I mean commercial, end user applications. For example, search "requires.NET" applications and look if they can ship to Linux thanks to Mono. As I said in my post, more than one of the apps I linked to are cross-platform. However, it's true they were mostly FOSS rather than commercial; and so I suppose if you have some wierd, skewed definition of "end-user application" that restricts itself to "commercial" programs only, as you apparently do, then my list wouldn't be very good at alleviating ignorance.
However, never fear, as five seconds of Googling that you are apparently unable or incompetent to do yourself yields lots of examples of commercial cross-platform mono apps; such as unity3d, plasticSCM, Versora, Voelcker
From Linux land, thanks to Trolltech Qt, a true multiplatform framework, Amarok 2 will release on X11/OS X and Windows using the same code. "From Linux land, thanks to mono, a true multiplatform framework, Banshee will release on X11/OS X and Windows using the same code." (Banshee on Windows, Linux). Many more examples are available at the end of a Google; just because you're ignorant doesn't mean they don't exist. Educate yourself!
My apologies, I did not mean to offend the people who control your mind. Sorry my mistake, I mistakenly assumed you could think for yourself and actually "had a pair". Enjoy your serfdom, it will last you till the day you die. I don't think there's anything I could say that would make that seem any more ridiculous than it already is;-)
Not to mention vast improvements to everything from Task Manager to Calculator (far more precise now, though the interface is unchanged) Maybe you've only just noticed it now in Vista, but the move from IEEE Floating point to an arbitrary-precision arithmetic library (infinite precision for standard operations, 32 bits for trancendental operations) happened back in Windows 2000.
or does anyone else take offence to politicians calling themselves "lords"?
Class distinction through self appointed title is bad enough, but to call yourself "Lord".... that's just rings of egoooooooo. I'm sorry, but that's just ignorant. Start by having a read through http://en.wikipedia.org/wiki/House_of_Lords.
I suspect they were relieved at the appeals - it gave them an excuse to keep on without publishing. No need to read semi-consipiracy theories into every nook and cranny. They can't publish a standard until it's ratified (otherwise there is nothing to publish). They can't ratify OOXML until the appeals process is over (and then only if the appeal is unsuccessful, obviously). This is all normal practice.
I will ask a single working, commercial application coded and shipped to OS X scene thanks to Mono or a single vendors end user application which is available on Linux thanks to Mono I'm not familiar with the OS X software scene, but w.r.t Linux there are many widely-used applications written in Mono. To claim otherwise is just ignorance. Thanks to Wikipedia for the following selection of popular mono applications for Linux (some Linux-only, some cross-platform):
Banshee music management and playback software for GNOME.
iFolder lets you share files across multiple computers and with other users through peer-to-peer or Novell's groupware server products.
libsecondlife is an open source implementation of the Second Life networking protocol.
Monodevelop is an [[Integrated development environment">IDE for creating Mono applications. It was originally a port of SharpDevelop to Gtk#, but is today developed on its own.
Muine is a music player with a user interface designed to be intuitive, developed by Jorn Baayen who also worked on Rhythmbox.
Second Life, the virtual world created by Linden Lab, will in the future be compiling all of the scripts in their own Linden Scripting Language to CIL. They will use an embedded Mono runtime to execute the CIL.
Tomboy is a desktop note-taking application which uses a Wiki-like linking system.
Unity is a game engine using Mono for game scripting.
Monotorrent (previously named bitsharp) a bittorrent library.
"After almost two years in development... Amongst the major improvements are a revamped rendering engine, massive increases in performance, EV and malware security features, synchronisation of bookmarks, a re-engineered mail back-end, improved address-bar searching..."
Doesn't sound like a "minor release" to me. Version numbers can be misleading: different organisations have very different ideas about when a new release should get a new version number. You can't compare versions numbers of different applications directly.
What do you mean?
HLT ...Are you seriously claiming that Windows doesn't use HLT? Don't be ridiculous. What on earth do you think NT's System Idle Process process is for, if not to issue HLT instructions when the ocmputer's idle?
On a laptop, Linux makes sense because if it has nothing to do, it sleeps. Windows, like rust, never sleeps. What do you mean? If by "sleep" you mean "Enter ACPI Standby mode", then both Windows and Linux have been capable of doing that after x minutes of inactivity for decades. If by "sleep" you mean "sit there and not do anything", then obviously both can do that -- it's not some special capability (though you may want to disable both operating system's respective idle HD indexing services (e.g. "kill trackerd" on Linux distros that use tracker)).
Save your hysteria for something genuinely catastrophic, like the loss of the print statement. It's not like you won't be able to print to a console; it's just being converted from a statement to a function for consistency. Is typing two extra parentheses really "genuinely catastrophic"?
Can you point me out websites where i can buy a keyboard without 'Windows logo' ? Might I suggest a permanent marker applied to the key in question, if it really offends your sensibilities that much, rather than buying a whole new keyboard?
As I understand it, the MinWin project was never about "rewriting the kernel from scratch to be more powerful"; that was exaggeration (and perhaps a little Microsoft fanboyism?) by the tech media. It was about modularising the NT kernel: consolidating the dependencies so that the core kernel can compile and run by itself independent of all the OS servers, subsystems etc. This is certianly no incredible new Windows 7 innovation; no matter what some Win7 fanboys may say -- look at how the Server 2008 kernel ("Server Core") compares to the Vista one: same basic kernel, only better compartmentalised. This is a good thing, and as a trend, it's something that's almost certianly going to continue in Windows 7. My guess is that's all MS were ever saying.
You do have to be careful with the tech media (inc Slashdot); you do get quite a lot of exaggeration and overdramatisation. This whole thing can prob be summarised as (forgive the l33tsp34k): Microsoft: Dudes, MinWin; we're compartmentalising the k3rn3l, man! Tech media: OMG Windows kernel rewritten from scratch kewl! Microsoft: Woah -- hold your horses, dude; it's still an evolution of the V1st4 & Server08 kernel! Tech media: OMG MinWin's been canc3lled!!1111one1!
Is that what Windows 7 is supposed to be? Vista + New Input Human Device Driver + Funky Taskbar? As a side comment, I do love how so many people seem to think that if their knowledge of Y is limited to A, B and C, then Y must consist exclusively of A, B, and C. I suppose it's kind of existentialist, in a cognitive dissonance sort of way: "Anything I don't know about must not exist"...
I've used the same technique before to restore an XP system that no one knew the admin password to. Changed the default screensaver.scr to cmd.exe. Just booted to login, made a coffee, when I got back, cmd was open in System user. Funnily, I think the user rights for System in XP were limited below Administrator Why bother? If you can reboot the computer, you can just boot into single user mode and change the password directly, on any operating system I've ever used (Windows: press F8 on bootup; Linux: append S to the GRUB kernel line, etc.).
... But you have to respect that Linux distros can do what they do and still remain with the very flexible and well-known X, all the while remaining completely open.
There is nothing inherently wrong with the tools and UI available in Linux distros when compared to MacOS.... Yes there is. You've just cited one example: X. It may be "very well known" and theoretically flexible, but good compared to a modern windowing system like Quartz it ain't. Ever tried to set up dual monitors on Linux? (Using the nVidia binary driver settings utility is cheating.). Compare that with the experience on a Mac, or Windows.
(If your answer to that is "Yes, and it was relatively easy, because it was within the last year and so since XRandr 1.2 was released, and I have xrandr-supporting drivers", then I'll raise the problem to getting three monitors to work, at all, somehow, ever. Considering that xrandr only supports two monitors and any drivers which support xrandr don't work with xinerama in any non-pathalogical way, good luck! (Maybe, in a few more years, xrandr will be able to handle more than two screens, and X will be where Windows Mac OS were... 10 years ago...). ).
Slashdot Mirror
If you are doing a near real time system... Windows 3.11 might actually work rather well, so long as the application wasn't trying to allocate too many resource handles.
Actually, I wonder why MS wouldn't release a non-preemptive Windows, just for this purpose. It would be a lot more reliable for some applications.
If all you need is a *near* realtime system, and you really want to use WIndows for some reason, then an NT derivative will do you fine: Priority levels 16 through 31 of the NT scheduler are soft realtime. And I'm sorry, but there's no *way* that 3.11 is more reliable than, say, a stripped down Win 2000.
(Of course, if you need *hard* realtime, neither NT nor 3.11 will do you).
Solaris and those others, OTOH will happily run for months and years without requiring a reboot. I recently ran across a system at work (RedHat 5) that nobody bothered with because it always did it's job. When I had to go look to see what the problem was, imagine my surprise to find it running RH5. Everyone that knew the root password had either quit or forgot they knew it, it had been sitting there running for several years. Windows will NOT do that.
Accroding to Netcraft, the server out there with the longest uptime is fp002.crayfish.net, currently at 1817 days (~5 years) of uptime and counting; running -- Windows 2000.
In the job I described above, all the code was written for Qt, on Windows and UNIX, because it was simpler, cleaner, and more reliable than the conflicting Windows APIs.
In recent versions at least, Qt in Windows works by using the Windows native APIs. So you're argument kind of nullifies itself -- you're arguing that using the native GUI APIs in Windows is difficult, (and, of course, impossible in Linux, since X doesn't have any), but then you admit that when you use a toolkit like Qt, developing a UI is equally easy across both platforms. So... do that then.
However I have two points right back at you. ... Even is IIS7 ran as well as Apache or better (which I doubt) Apache is vastly more secure.
You can say that again. Why, in the year and a half since IIS 7 shipped, it's been deluged under a mammoth 1 Secunia advisories, whereas Apache 2.2 has had a mere 5 (2 still unpatched) in the same period. Clearly, Apache is vastly more secure. You sure showed them!
KDE? more modular? Kparts means that you can include practically entire programs (spreadsheets, browsers, editors) inside other programs - how much more modular can it get?
You say that as if this is some magical, never-seen-before new technology in KDE4 that indicates an unprecedented level of good design and modularisation. This is nonsense. Component frameworks have been a standard part of desktop environments for ages. Gnome has Bonobo and now D-bus; Microsoft has COM and now WCF; etc, etc.
Neither are sandboxed and both run with the same privs as the browser AFAIK.
you know wrong. ActiveX on Vista runs in a sandbox with lower rights than the current user.
Actually, the GP isn't wrong. But you're not wrong either. The GP is right that ActiveX runs with the same privs as the browser, and you're right that ActiveX runs with lower privs than the current user (on Vista); because the browser *also* run with lower privs than the current user.
The only thing I can think of that you could mean is that you've been lead in by some of the "patent encumbered" FUD. In which case let me disabuse you: the only parts of mono which might even be theoretically patent-encumbered are the parts that are not ECMA-standardized; most notably the Windows Forms implementation. Now, which of the apps I linked to used Windows Forms? Yup: None -- they all used Gtk#. So even by that definition, none of the apps are even in the slightest "non-free".
However, never fear, as five seconds of Googling that you are apparently unable or incompetent to do yourself yields lots of examples of commercial cross-platform mono apps; such as unity3d, plasticSCM, Versora, Voelcker
From Linux land, thanks to Trolltech Qt, a true multiplatform framework, Amarok 2 will release on X11/OS X and Windows using the same code. "From Linux land, thanks to mono, a true multiplatform framework, Banshee will release on X11/OS X and Windows using the same code." (Banshee on Windows, Linux). Many more examples are available at the end of a Google; just because you're ignorant doesn't mean they don't exist. Educate yourself!
"After almost two years in development ... Amongst the major improvements are a revamped rendering engine, massive increases in performance, EV and malware security features, synchronisation of bookmarks, a re-engineered mail back-end, improved address-bar searching..."
Doesn't sound like a "minor release" to me. Version numbers can be misleading: different organisations have very different ideas about when a new release should get a new version number. You can't compare versions numbers of different applications directly.
HLT ...Are you seriously claiming that Windows doesn't use HLT? Don't be ridiculous. What on earth do you think NT's System Idle Process process is for, if not to issue HLT instructions when the ocmputer's idle?
As I understand it, the MinWin project was never about "rewriting the kernel from scratch to be more powerful"; that was exaggeration (and perhaps a little Microsoft fanboyism?) by the tech media. It was about modularising the NT kernel: consolidating the dependencies so that the core kernel can compile and run by itself independent of all the OS servers, subsystems etc. This is certianly no incredible new Windows 7 innovation; no matter what some Win7 fanboys may say -- look at how the Server 2008 kernel ("Server Core") compares to the Vista one: same basic kernel, only better compartmentalised. This is a good thing, and as a trend, it's something that's almost certianly going to continue in Windows 7. My guess is that's all MS were ever saying.
You do have to be careful with the tech media (inc Slashdot); you do get quite a lot of exaggeration and overdramatisation. This whole thing can prob be summarised as (forgive the l33tsp34k):
Microsoft: Dudes, MinWin; we're compartmentalising the k3rn3l, man!
Tech media: OMG Windows kernel rewritten from scratch kewl!
Microsoft: Woah -- hold your horses, dude; it's still an evolution of the V1st4 & Server08 kernel!
Tech media: OMG MinWin's been canc3lled!!1111one1!
... But you have to respect that Linux distros can do what they do and still remain with the very flexible and well-known X, all the while remaining completely open.There is nothing inherently wrong with the tools and UI available in Linux distros when compared to MacOS.
(If your answer to that is "Yes, and it was relatively easy, because it was within the last year and so since XRandr 1.2 was released, and I have xrandr-supporting drivers", then I'll raise the problem to getting three monitors to work, at all, somehow, ever. Considering that xrandr only supports two monitors and any drivers which support xrandr don't work with xinerama in any non-pathalogical way, good luck! (Maybe, in a few more years, xrandr will be able to handle more than two screens, and X will be where Windows Mac OS were... 10 years ago...). ).