Slashdot Mirror
ReactOS 0.3.15 Released
Beardydog writes "From the ReactOS.org bulletin, 'The ReactOS project is proud to announce the release of version 0.3.15. A culmination of over a year of development, 0.3.15 incorporates several architectural enhancements to create a more compatible and conformant implementation of the NT architecture. Perhaps the most user visible enhancement is initial support for USB devices, both storage and input.'"
ReactOS is a project to build a free, open-source clone of Windows, compatible with both drivers and userspace software. Why on earth hasn't this received more support from the OSS community? It's the only realistic chance of dethroning MS from the desktop in favor of an open alternative. Linux is fine for servers, portable devices, and embedded systems, but trying to stick it on the desktop is a foolish dream that has failed for over 10 years.
Please tell me you are trolling.
Windows is broken by design, I have no desire to run it.
Window's VMS derived architecture is pretty terrible.
Mod me down, my New Earth Global Warmingist friends!
Why do you think that would be the only realistic chance? Apple seams to do fairly well with OS X, and that's not at all like Windows.
In what way is it broken by design?
Linux is fine for servers, portable devices, and embedded systems, but trying to stick it on the desktop is a foolish dream that has failed for over 10 years.
Linux has worked wonderfully on my desktop for over 10 years.
Give me Classic Slashdot or give me death!
The graphic routines reside in kernel space? the drivers can kill the kernel?
All of my employees are living the 'foolish dream' of Linux desktops. It's free and works great for us.
You have no desire to run Windows until the moment you get a job in an industry whose standard applications are unavailable for GNU/Linux and rated "garbage" in Wine AppDB.
I can think of one reason: registry
Don't know about you, but I've been using Linux exclusively since 2007, and it is on a desktop computer. Personally I've been pretty happy with it! Can't afford to buy Photoshop, MS Office and all the rest, so I use what is available in the Repository and seem to get by pretty well!
That's not a reason to run windows. That's a reason to promote open standards everywhere. Why should open source enthusiasts lift a finger to make it easier for industry to use closed systems? It would be far better to write an open replacement.
Give me Classic Slashdot or give me death!
MacOS isn't too different from Linux. It's built from NeXTSTEP, FreeBSD and OpenBSD. NeXTSTEP was built on BSD to start with as well.
(yes I am aware that Linux and BSD are completely different, but they are both "unix related")
Somehow you just reminded me of GConf.
And that's different from a Linux framebuffer driver how?
It's obsolete, due to virtualization.
You're missing the point of ReactOS. It's like Wine, except actually an OS. If one day Windows bites the dust you can have this project for any legacy code, if you really needed to run something, and it could be patched and maintained forever. To be fair, yes this is a hobby OS, but to say that with disdain diminishes the value of a hobby.
Why should anyone care about making an open source Windows now, anyway?
Windows is dying. Its uses are rapidly deterorating and Microsoft's sped along that process with the Windows 8 disaster.
Good luck creating a replacement for AutoCAD or for the parts of Adobe Photoshop that GIMP doesn't replace using only hobbyist labor.
I applaud what the ReactOS guys are doing and wish them the best of luck, but on the other hand I believe they have chosen a very difficult path.
Linux is free software, which means that you can inspect the source code and prevent it from crashing the kernel.
This is why I keep one windows laptop around. I got a killer deal on it about a year ago.
17" quad core AMD, with 8GB RAM. Refurbished, $400 in Canada. It doesn't get used much, but nice to have the odd time I need it.
Buying a laptop in Canada it is almost always $100 more than buying the same laptop in the US.
If I were God, wouldn't I protect my churches from acts of me?
He means a relevant number of desktops, not three chucklefucks in their basement.
If there was a compatibility layer to run OSX applications on Linux, that might actually be a viable option. OSX has most of the big things people want: MS Office, Adobe Photoshop and friends, AutoCAD, etc. Conceivably, such a compatibility layer could be easier to write, debug, and maintain than WINE, since there is a lot less legacy baggage (and the underlying architecture is much closer to what Linux expects). But I am not aware of any such project so far, and I don't have anywhere near the level of systems programming experience needed to start it.
I can't stand the interface of OSX; it's even worse than Windows 8, and everything is in the wrong place and none of the shortcuts and gestures wired into my brain and fingertips work right. But with a compatibility layer on Linux, the apps could be run while allowing the desktop to be customized fully.
One thing that is absolutely non-negotiable, though, is that the font rendering needs to be fixed. In its current state, it's uttterly atrocious, the worst I've ever seen on any OS. The Microsoft core fonts expect aggressive hinting and snapping to the pixel grid, and Linux doesn't want to cooperate.
Gconf can't cause your system not to boot.
I've been running Kubuntu as my primary desktop since December, 2006, and I've loved it. Hell, my mom and uncle run Kubuntu, as well. Yeah, I'm the family tech support, but they rarely need help. Kubuntu has just been that smooth.
Whenever people claim that Linux will never be mainstream on the desktop, I laugh. What those people don't seem to realize is that, just because no one has made a "Linux for the layman" doesn't mean no one ever will. Android is Linux, and it's about as user-friendly as you can get. It can be done. Someone can take Linux and create a version that is accessible and easy enough for the non-tech people to use. Just because it hasn't happened yet doesn't mean it's a waste of time or a "foolish dream."
It can prevent your GNOME session from loading, and thus in practice cause your system not to boot.
The graphic routines reside in kernel space?
An absolute necessity for performance reasons. They tried doing it in userspace in NT4 and it just couldn't keep up.
the drivers can kill the kernel?
Windows 7 moved a lot of drivers to userspace. Yes, some code will still be run in the kernel. Some code is run in the kernel on Linux. The solution is for that code to be written well, not to give up and pretend that kernel mode doesn't exist.
The registry gets a lot of hate, but I don't see how it is worse than the alternative, which is tons of different .ini files (or equivalent) for each application and setting. At least on Windows, it's generally understood that settings should be exposed in some way in the GUI and that for all but the most advanced features, saying "go edit the registry" isn't really a good solution. On Linux, forcing users to manually edit config files is routine.
And AutoCAD and Photoshop are the easy ones. They have a wide audience and it's generally understood, at least in a broad sense, what they're supposed to do and why it's important. Good luck rewriting a million different industry-specific niche applications for Linux. Better luck finding the coders willing to volunteer on obscure projects that neither they nor anyone outside the industry in question cares about.
Who said anything about using only hobbyist labor? Those companies that use those tools would probably save themselves money in the long run if they commissioned an open replacement.
Give me Classic Slashdot or give me death!
What's the point? Windows is a dying platform, why make a clone of it? I don't think the problem with running Windows apps is running Windows and need an alternative.
Also no point trying to dethrone anyone when the throne is on a sinking ship. Anyone vying to replace Windows on the PC desktop has not crawled out of their cave long enough to realize everyone has moved over to tablets and phones as their primary computing device.
I haven't thought of anything clever to put here, but then again most of you haven't either.
What do your employees do? I suppose if you have something like a call center where most operations can be done through the browser (including a web-based CRM system) and where you don't want your employees going off on tangents, it might work. Assuming they can resist the urge to claw their eyeballs out after staring at the horrendous font rendering all day. But for anything more than that, it's just a total nonstarter.
Why should anyone care about making an open source Windows now, anyway?
Because Windows owns the business world, most of the power-user world, and most of the PC gamer world. If you want OSS to make any inroads on the business desktop or with gamers, it has to run their software on their terms. And that means Windows binary compatibility.
It's the only realistic chance of dethroning MS from the desktop in favor of an open alternative.
It has no chance of dethroning Windows. Zero. Zip. Nada.
Look, no one will ever be as good at being Microsoft as Microsoft is. ReactOS may be eventually be 99 44/100 % Windows compatible. It may look like Windows, feel like Windows, and act like Windows almost all the time--but it won't be Windows. And sooner or later, anyone running it will run into some instance where Windows does this but ReactOS does that. Now, when this happens (when, not if) developers will say, "That's interesting, we should fix that." But regular users will think, "Serves me right for trying to use this cheap knockoff. Guess I'll just get the real thing." And if anyone asks them about their experience with ReactOS, that's pretty what they'll say.
ReactOS is an interesting project, and I wish them the best of luck. I'm sure it will find its uses. Taking significant market share away from Microsoft isn't one of them.
The same, BTW, applies to open-source clones of other Microsoft products, which is why it's kind of dismaying that the OpenOffice family (LibreOffice, etc.) tries so hard to imitate Microsoft Office interface standards. Those aren't the only way to design office software, and there's no reason to assume they're the best, either. The more you chase the market leader, the less chance of eventually becoming the market leader you have. Try to do something different and better instead. That's about the only way any piece of software has ever broken another's market dominance, and probably the only way it ever will.
The correlation between ignorance of statistics and using "correlation is not causation" as an argument is close to 1.
Windows 8
Ok boss, I need a few million bucks to have some guys write a replacement for this $5000 a year software. Or did you want me to just buy windows 8 and the software?
You use the word "rapid" to describe Windows' deteriorating use. Tell me, on a scale of "Watching grass grow," to "Why did the snail cross the road?" just how rapid are we talking here?
I'm not about to argue that Win8 is anything but a train wreck, but claiming that Windows' position in business is rapidly changing is a bit.. well.. dead wrong... Sure Win8 isn't being adopted in droves as MS might like, but of all of the WinXP-Win7 users who *aren't* upgrading [sic] to Win8, I don't imagine terribly many of them are jumping ship and moving to Linux or Mac. Some yes, but I doubt it's more than background noise on the stats.
Even if Microsoft didn't see a single new Windows license this year, their market share isn't in any immediate danger. (And by "new license," I mean an actual new person who wasn't using Windows before. Replacing an existing Winbox with a new one and paying the Microsoft tax or changing the installed Windows version on an existing machine doesn't effect the overall number of people using Windows and buying software to run on it.)
And perhaps worth nothing, this relatively MS-positive post was brought to you by a Mac user...
No one is saying you have to run it. But this would be great for anyone's technology-repellant relative who just wants Microsoft Word and a web browser. Not to mention it would put pressure on Microsoft by being a viable competitor in the "normal" consumer market.
Also, I (and I'm sure others) would prefer an OSS OS that could run today's games natively, since most games are still Windows-only.
Which businesses have moved their staff from 19" monitors to 4" mobile phones?
Well yes, but at the same time I only use Windows for one thing nowadays: running it in a virtual machine under Linux so I can run 3 programs that have no equivalent under Linux. If those programs work under ReactOS, I'll use that in a heartbeat. They must understand that beacuse the provide an already made VM among the downloads.
Non-Linux Penguins ?
Even newer versions of Windows have to run legacy Windows software. How do you think a competitor would manage to move into that same market without being able to do the same?
I heard hearings about ReactOS and Kickstarter They want to make some campaign
lol
still no modern file support, as fuse is still not supported
Back in the 1990's I was really excited about this project. I really hated how Microsoft had a strangle hold on the entire industry and there was no sign that it was going to change anytime soon. This project was promising in that it really offered a possible solution. But they're about 13 years too late. Far too little progress has been made. Microsoft has already been knocked off of its pedestal and now there are viable alternatives that consumers are embracing. Specifically, MacOS, IOS, and Android devices. Linux is still a niche for desktops. And the browser is really the thing that killed Microsoft more than anything else.
That's pretty much the kind of thing I was thinking of when I said, "I'm sure it will find its uses." It's very different from replacing Windows as the primary corporate and home user desktop OS, which is what I had the impression OP had in mind.
The correlation between ignorance of statistics and using "correlation is not causation" as an argument is close to 1.
Window's horrible cluster-fuck of screwing up the VMS derived architecture is pretty terrible.
There, FTFY.
"I don't know, therefore Aliens" Wafflebox1
First: Stop running GNOME.
Second: There are not many things that most places couldn't do on Linux. Please, tell me some. We're not using AutoCAD here, and graphics team, while still wanting PS and AI, doesn't really care if it is on Mac, Windows, or IRIX (our lead graphics designer used to use PS and Maya on SGI hardware), and they have different hardware needs then most users anyway.
I'm starting to think GNU is the problem with "GNU/Linux" these days.
This was solved on unix in the mid 70s. System configuration files exist within directory structures in /etc
User configuration files exist within dotfiles (hidden files on unix) in the user's home directory, ie /home/JDG1980
You can make all the fancy UIs you want, and that's great, but I only care that they read from and write to the textfiles residing in /etc somewhere. Yeah, that's tons of files. I'm used to working with tons of files. Find, grep, sed, awk, bash, perl, et cetera ad infinitum work just fine for me even with filesets ranging in the thousands.
The graphic routines reside in kernel space?
An absolute necessity for performance reasons. They tried doing it in userspace in NT4 and it just couldn't keep up.
Hummm, what?!?! You know it is the same computer, right? If it is killing performance, then the scheduler is fucked. Another proof windows is broken.
Linux had some performance problems with database (Oracle specifically) because of it running in userspace. They changed the scheduler and some other stuff (rawio, etc), and it works flawlessly now. Based on your rationale, the solution would be to move Oracle to kernel space.
pretend that kernel mode doesn't exist
You know what should run in kernel space? The kernel.
morcego
To be fair, yes this is a hobby OS, but to say that with disdain diminishes the value of a hobby.
Nice point. Amateur means "one who loves" (literally), it should never be disparaging to be called an amateur. Hobbyists are "amateurs" by definition. All the great Renaissance thinkers were amateurs across a wide range of fields, but often to great depth. Hence we owe much of modern thinking to amateurs.
The registry gets a lot of hate
Yeah. And then those same people who keep hating windows registry go and implement the same thing for Gnome, in a even more crappy way than windows did.
Gnome is, AFAIC, the current bane of Linux.
morcego
But there are limits on what ReactOS can do - and pretty serious ones. For one, it can't use NTFS w/o violating Microsoft patents, so fat chance seeing a modern filesystem on it.
I have previously suggested that the project be split into 2 or 3 parts - one win64, another win32 and a third win16. Have the win64 project aim at Windows 7, and target 4GB of RAM. Have the win32 aim at XP, and target 256MB of RAM and above, and for the win16, try something like a 16-bit thunked version of it. The goal of the win32 & win16 would be older boxes, and it should aim at being compatible w/ any win32 drivers, be it XP, NT, 2000 and so on for the win32 spin, and w/ Windows 95 for the win16 version of it.
But to those who've been claiming that ReactOS has a moving target, no it doesn't. Already, most people are finding Windows 7 good enough, but Microsoft wants to take us to 8. So here, a 64-bit ReactOS would be perfect.
Oh, and some long term suggestions (for an OS that's taken eons to develop that make it the envy of HURD). Make the user interfaces of all the versions that Windows ever had, and make them selectable regardless of which OS is underneath. In other words, a win64 version of the OS could have the classic NT 3.5 interface, or a win16 version of the OS could have the XP interface - just the same way Linux guys can pick DEs. In other words, decouple the user interface from the underlying OS, so that people can pick & choose the best of all worlds.
The other thing I'd suggest - some of the modern OS features, such as IPv6 support - make them available to all the OSs. Also, they could take NTFS, make the 32-bit file system 64-bit, and then make that the file system for the OSs. That could enable them to sidestep the patent issues, and things like it. Also, I'd like it to be ported to some other CPUs other than just Intel - not just ARM, but MIPS and PPC as well.
2 years behind? Only "initial support" for USB devices? That's more tahn 15 years behind the times.
Or you can just run Windows in VM?
How is that the machine not booting?
It still boots just fine.
Actually, many Windows drivers now run in user-space where they don't even need to context switch, while Linux runs many drivers in the kernel space.
Config file files are easy to edit and simple to automate. They are also generally well documented.
The registry none of those. In practice stuff gets thrown all over the place based on the programs creators with often no good way to automate program setup.
Because the computer is just as unusable at that point.
On the other hand, at least ini files are portable. The registry's sole purpose is to replace ini files with an identical structure in a less portable binary database. I'm not sure how not being able to copy all of an app's setting without playing "find the fucking registry keys" is better than having a few more files on disk.
Because using Windows requires you to buy Photoshop and Microsoft Office? Since when?
Not at all, I use a great many computers that have no X at all.
You can even login and fix the issue unlike a machine with a borked registry.
A few million bucks? You're doing it wrong.
The catch-up argument no longer holds. The current project is aimed at XP, and once it's done, they'd have a mature win32 OS that can address all x86 boxes. If they do a follow-on project that is aimed at Windows 7 and uses win64, that's all they'll have to do. There won't need to be a Windows 8 based OS - Windows 7 will be good enough. So future libraries that Microsoft introduces will be irrelevant.
The legal hurdles can be real, though, but since this project points out that it does not use any MS source code but just their published specs, they have nothing to fear there. The parts that are patented and blocked, such as NTFS, they're not doing.. But the project does have good reasons to work, but first, they need to properly staff that operation, give themselves deadlines and stop treating it like they have all the time in the world.
GDI was moved to the kernel back in the 90s wasn't it? Won't things change now almost everything renders via WDM and not by GDI?
ReactOS is.
To clone a system we can't get rid of on so much new computers. What is benefit there?
To be free of token license fee Microsoft demands for basic OS, when Office & MSQL & ... is what they really are leveraging for. And ReactOS is as good for leverage into M$ dominance as M$ OS is.
And so on.
Stupid. Waste. Of. OSS. Talent. ReactOS.
http://opencm3.net, http://www.nongnu.org/gm2/
My employees do typical office stuff - web browsing, spreadsheets, email, faxing, printing, word processing etc etc. No need for Windows so why pay for it? I've never had anyone complain about fonts either - everything works just fine.
And its entire life, started as a single-user system, means the whole damn thing is broken as far as multi-user goes.
That was only true of Win9x, and the last version of that was discontinued about 10 years ago. Windows NT (which includes 2K, XP, Vista, 7, and 8) was built from the ground up as a modern, multi-user OS with full support for security built in. In fact, the NT security model is slightly more sophisticated than the Unix model (though not as good as SE Linux). Both do share the same flaw: from a security POV, the program is the user and can do whatever the user wants. This is something Android got right, granting permissions on a per-app rather than per-user basis.
A lot of people ignored the NT security provisions up through XP by running as admin all the time, but UAC mostly killed that. People hated it, but it gave the developers a much needed kick in the butt to stop breaking stuff by requiring root.
Which doesn't really work out. If every 5 years it was only 2 more years behind it would have already had USB since that would only put it 6 years behind (since it's only around 16 years old). Which would mean it would have more than "initial support" since even USB 2.0 came out 12 years ago.
I concur. Insofar as the registry removes the need for a metric crapton of config files, I am in favor of it, but it stops there. The manner in which they implemented it and documented it has turned it from something potentially useful into something even worse than 200 config files.
If they had done it right, they could have standardized good practices, and even allowed for the creation of useful configuration management utilities. Since they didn't do it right, it's just a pile of confusing and useless crap that you can't avoid.
This depends on Microsoft cleaning up its act on Windows 8. But really speaking, it'll be more & more difficult to move users to future versions of Windows, except maybe to end support for the existing ones. And that's where the opportunities for ReactOS will open up - once users are forced out of their comfort zone, they'll consider ReactOS, and if it works more or less the way Windows used to, it'll displace Windows in those places. Corporate sites may stay w/ Microsoft due to support contracts, but outside that, I don't see Microsoft being successful in forcing users to move along w/ them.
If there was a compatibility layer to run OSX applications on Linux, that might actually be a viable option. OSX has most of the big things people want: MS Office, Adobe Photoshop and friends, AutoCAD, etc. Conceivably, such a compatibility layer could be easier to write, debug, and maintain than WINE, since there is a lot less legacy baggage (and the underlying architecture is much closer to what Linux expects). But I am not aware of any such project so far
Well, there's the Darling project. I get the impression it's very much a work in progress, however.
You never read the Song of Fire And Ice then by G.R.R Martin. That's the whole point of the series IMO.
21st Century Renaissance Man
We moved most of our staff from 19" monitors to 10" iPads. Not saying it was a great move, but we did it anyways. (Of course most of our staff is outside sales reps, and carrying an iPad is a lot easier than carrying a laptop or desktop.)
21st Century Renaissance Man
It doesn't matter if the driver is in kernel or user space when your graphic card can DMA to/from your main memory space anyways.
You are not forced to edit config files any more often than windows forces you to make manual registry changes...
The primary reason that technical people will choose to edit config files instead of using the gui is because it's much easier to explain in either a textual (website, forums) or vocal method. Telling someone to transcribe what you're talking about is infinitely easier than trying to explain over the phone how to navigate a gui, and in a textual medium you can even include examples which the user can cut/paste.
Also text based config files usually have comments where you can explain why you made a change, or where the authors of the program can explain what settings do and give examples. The registry has nothing like this.
Text based configs are also very easy to back up and store in a revision control system, that way you can roll back changes, see what/when changes were made etc.
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
The graphic routines reside in kernel space?
An absolute necessity for performance reasons.
Oh, that's why Value got more FPS on games natively on linux vs windows, right?
Linux has a macro kernel... all the drivers are part of the kernel and run in kernel in kernel space...
Of course Unix wasn't a single target, it was a whole family of operating systems. Linux was no more different from those Unix systems than the Unix systems from each other.
The Tao of math: The numbers you can count are not the real numbers.
Don't forget, Microsoft's own Windows NT4 with no USB "plug-n-pray" or mass storage device support without a special driver from the vendor was still being used and supported up until 2004; they were even offering extended support contracts to customers with large NT4 install bases throughout 2006.
Windows 95 OSR2 had USB device support, it also was pretty terrible by most standards for doing anything more than running a couple of applications at a time or playing a game. I guess you could say Win2K had USB support, but virtually nobody used it in the consumer space due to it's lack of game support. It wasn't until XP came out that MS had a decently stable OS with real protected memory, preemptive multitasking and also had decent 3rd party graphics support.
grep -iw skynet
It was just as unusable to begin with because it was running gnome...
Gnome is, AFAIC, the current bane of Linux.
A good argument for switching everyone to Unity.
Who would win this election: Andrew Weiner vs Andrew Weiner's weiner.
The kernel was designed with security and multiuser in mind, but the ui and apis on top of that were not and thus neither were a large number of applications. You have a lot of areas where things were obviously kludged in at the wrong level and are thus easily circumvented, eg see group policies such as command prompt restrictions.
The security model may well be more sophisticated, but more complexity is NOT a good thing. There is a reason why the vast majority of linux users do not use selinux, and that is because the overhead of learning and maintaining it outweighs the benefits in most cases.
A simple security model suffices for most use cases, is easy for people to understand and likely to be used effectively...
A complex security model either gets in the way (and thus people find ways to circumvent it like running as admin) or the complexity causes people to make mistakes when configuring it (often because its too complex to understand) and thus introducing new holes.
Look at the way services are run as unprivileged users... On unix you just setuid(), which is simple and seemingly insecure... On windows you must "authenticate" as that user, which means storing the user's password on the system. End result? Either programs don't bother, and run as system, or they leave a password easily obtainable from the system (which may be valid on other hosts, especially if your using a domain user).
Similarly, many of the windows networking protocols let you authenticate with the password hash, as such the hash is the equivalent of plaintext so the passwords are effectively stored as plaintext on the host. Compromise one host, grab hashes, attack other hosts.
You can also extract plain text passwords of currently logged in users from memory...
So overall the linux security model is MUCH better... You have a simple model that the vast majority of users can easily understand and use effectively, and a complex model that can be used for those specialised niche cases.
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
That was less than 5% just a few years ago...
And doesn't run on the cheapest tiers of hardware...
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
Funny thing there... I have 4 windows open right now: one is OS X, one is Windows 7, one is Ubuntu 10 and one is Ubuntu 12...
Ubuntu 10 is slightly blockier than the others, but with the right fonts, it's not that noticeable.
I'm definitely doing a lot more than staring at browser windows :)
Business users are gradually moving to web based applications, which at least when properly designed are platform agnostic...
On the other hand, ReactOS could be very useful for supporting legacy applications, which many companies find themselves locked into and end up having to keep ancient hardware for.
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
$5000 could easily get most industry niche apps running on wine.. where they will stay running on new and fast hardware/linux unlike the current niche apps which in some cases STILL need DOS.
Disney ported Photoshop 5 (I think it was 5, too lazy to look) to be used by their animation department for $20,000 if I recall correctly. Most industry niche software is nowhere near as complex as Photoshop 5.
Most of that niche software is stuff like estimating, inventory, retail serial programmers, specialized accounting, that sort of thing. The software itself is fairly simple. You could probably have someone describe the software functions, menuing, features, etc from the niche software you use and have what amounts to a clone built from scratch for $5000 - $10000 in most cases which isn't bad. Go to a trade show, find a few like minds, split the cost and it becomes pretty much no cost.
Linux has worked wonderfully on my desktop for over 10 years without rebooting.
Because,
1. By the time it's got a decent compatibility with Windows XP, we'll be on Windows 12.
2. It's only legally safe because it's only a novelty. Should it become mature enough to use in a production environment, Microsoft would surely wish to assert about a bajillion patents. Most of them rubbish ones, but enough to cripple the project with legal costs and scare away developers.
It has no chance of dethroning Windows. Zero. Zip. Nada.
Look, no one will ever be as good at being Microsoft as Microsoft is. ReactOS may be eventually be 99 44/100 % Windows compatible. It may look like Windows, feel like Windows, and act like Windows almost all the time--but it won't be Windows. And sooner or later, anyone running it will run into some instance where Windows does this but ReactOS does that. Now, when this happens (when, not if) developers will say, "That's interesting, we should fix that." But regular users will think, "Serves me right for trying to use this cheap knockoff. Guess I'll just get the real thing." And if anyone asks them about their experience with ReactOS, that's pretty what they'll say.
That's exactly why Linux failed to replace UNIX. A knockoff can never succeed.
A knockoff that's competing with a family of OSes that aren't 100% compatible with each other at the source level, that run on machines that are typically more expensive than the primary class of machines on which the knockoffs run and that don't even have the same instruction sets as each other (so that binary compatibility is out of the question), and on which a lot of the software is either open-source or written in-house so that it can be compiled and run on the knockoff, could succeed.
A knockoff that's competing with a single OS that has a ~90% market share and that has a huge collection of binary-only packaged applications that might depend (explicitly or implicitly) not only on documented behavior but also on undocumented behavior is a different story.
You haven't gotten out much. GU/Linux is being used by grandpa, grandmas, school kids, and techys. All others not so much. But what you shouldn't underestimate is the size of the population which is retired. It's a growing base.
And your office staff? How are they finding working on a much smaller screen?
Nothing is 99 44/100% windows compatible...
Apps that ran on xp sometimes fails on 7 etc... There are many apps out there that require specific versions.
So given that no version is fully compatible with all windows software, you just need a version that is compatible enough with the specific applications you want to run, which might actually require you to have multiple versions (eg in a vm). If reactos is good enough for your particular use case, then the fact it's free and open gives it an advantage.
Windows must be about the most complex os in existence, and i'm sure much of that was intentional to discourage cloning. Dos was simple, and had numerous clones... Ofcourse simple is good, as it makes the system easier to maintain and use.
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
Aren't most open source projects amateur projects, or at least start out as one?
The efforts of amateurs and their willingness to share what they have loved has added so much to the world.
Linus Torvalds is usually considered the prime example of this, especially if you read his first Usenet post about the primitive Linux kernel he was developing on his 386.
The only use I can see even if it were mature would be handling ancient legacy apps that won't run on a new version of windows. If the program demands XP, then the options will be either XP (with a ton of security holes MS no longer patches) or ReactOS. And if MS stops selling XP licenses, there may be a very business-specific applications end up on it. But it's a very small niche.
How will you activate newer versions of windows if the activation servers have been turned off?
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
Linux has a macro kernel... all the drivers are part of the kernel and run in kernel in kernel space...
Linux is currently a mix of macro (monolithic) and micro kernel concepts. Not all drivers run in kernel space. I'm sure you were remembering the old Linus vs. Tannenbaum disagreement when you wrote that one.
Regardless, the focus of this discussing is graphic routines which, except for a few proprietary cases (most notably, nVidia), run in userspace. Which is one of the problems people have with the proprietary nvidia driver (another is it not being free, but w/e).
So, anyway, not ALL drivers are part of the kernel, more and more are moving out of it as time goes by. But yes, many drivers still are. Our Minix legacy.
morcego
How do I setup an NFS mount point under Linux without editing /etc/fstab by hand?
"Ignorance more frequently begets confidence than does knowledge"
- Charles Darwin
Careful with those arguments! They're antiques!
"Ignorance more frequently begets confidence than does knowledge"
- Charles Darwin
I think "sudo mount -t nfs server:/share ./location"
Well. This is funny. If you put the effort which goes into ReactOS into developing a driver layer for the linux kernel which allows to expose windows driver functions after loading windows drivers to sw which may be interested and polishing wine a little bit, you have much better chances of developing a working Desktop replacment.
Moreover: Linux may be failing on the desktop (somthing to be disputed), but linux is awfully sucessful everywhere else. The turning point probably was that sony startet to put linux in all their products (from e-book readers to televisions and cameras) and linux (android) is replacing windows systems at a tremendous rate. I must be really bored to fire up the email reader on my laptop. Usually all my communications go via my phone, and thats what i observe anywhere. I see people buyn tablets with android who would have been hard-core "i need MS Office" followers.
ReactOS is a nice idea and an interesting project. But it you ask me for the shortest route to replacing windoes desktops, it wont be involved.
When you've used linux enough, and "get it", then you'll have no desire to do otherwise.
I am a life long mac user (since 1985, MacPlus), and love writing usable guis; however, I still prefer the power of text files for configuring my computer. A bad gui is terrible, and a good gui has a ridiculous amount of state to handle. Config files give you all the power you want, with the addition of allowing you to use unix tools to glue together various processes. I believe the main reason people don't write guis to edge config files is that they wouldn't use the resulting gui anyway.
Like all pain, suffering is a signal that something isn't right
By using the crack that all the pirates do.
I said gconf CAN'T cause the machine not to boot. Whereas the registry can and usually does at some point.
Windows 98 had USB support. Also, what relevance does NT4 have to do anything? Most of its development was done before the USB 1.0 spec was even finished. So it's pretty silly to bring it up at all.
Why should anyone care about making an open source Windows now, anyway?
Because Windows owns the business world, most of the power-user world, and most of the PC gamer world. If you want OSS to make any inroads on the business desktop or with gamers, it has to run their software on their terms. And that means Windows binary compatibility.
Which is why we have WINE -- we want binary compatibility, not the ability to load up an entire OS that looks like Windows 95 and behaves like Windows XP.
IMO, WINE got it right; it provides a compile target for developers who don't want to re-develop for Linux, and a runtime wrapper for developers who don't even want a Linux build. Plus, WINE itself is platform agnostic, so it'll run on any POSIX compliant OS (read: almost every OS that isn't Windows, and even Windows for that matter), providing binary compatibility to everyone, not just the adopters of a particular distribution.
Of course, WINE and ReactOS have an interesting co-dependant relationship; ReactOS is a great platform for testing stuff out before it gets rolled into WINE, and that's the reason I'm very happy it's still around (even after the whole source integrity issue a few years back).
Hummm, what?!?! You know it is the same computer, right? If it is killing performance, then the scheduler is fucked. Another proof windows is broken.
Nope. The problem is with the cost of a context swap. When usual kernel code executes, the code still executes in the context of the current process. There is no MMU context swap, just some registers get saved to stack, and the protection level is changed. When you run it in a separate process, you not only to invoke the scheduler (which isn't invoked when you merely do a syscall), you have to swap out the entire MMU context. This is very expensive, and only gets relatively more expensive compared to running code with more advanced processors.
A successful API design takes a mixture of software design and pedagogy.
A "permanent" mount point?
"Ignorance more frequently begets confidence than does knowledge"
- Charles Darwin
You misunderstand *completely*.
The person I replied to said "You are not forced to edit config files any more often than windows forces you to make manual registry changes" which is demonstrably wrong as my example shows.
What I prefer has NOTHING to do with it.
"Ignorance more frequently begets confidence than does knowledge"
- Charles Darwin
That's when you get a VMware license :)
A successful API design takes a mixture of software design and pedagogy.
And run what in VMware? Running Windows in VMware is still running Windows, which is what h4rr4r is claiming to be able to avoid.
The code that managed context swapping is part of the scheduler, at least on Linux. Yes, it can be costly, which is why it needs to be implemented correctly, and why you keep getting alternative schedulers (not as often as you once did, it was crazy back in early 2.0). There is classic problem with Intel-HT and Postgresql that caused context swapping for database I/O to be extracostly, as you probably recall. And it can be done correctly, as was proven in this case, and then again for Oracle.
It is absolutely possible to have high performance userspace graphics, as was proven with some of the more up-to-date drivers. I think it was ATI that first did it, by the way. The trick is to keep as much as possible in userspace, but that requires a change in mentality for developers.
morcego
At least you can do the rest of your job hopefully better and easier by being otherwise freed from Windows.
A successful API design takes a mixture of software design and pedagogy.
You can even login and fix the issue unlike a machine with a borked registry.
How many individual users who are not IT professionals are willing to learn to log in and fix a broken GConf or do anything else on a desktop PC running GNU/Linux* without X?
End result? Either programs don't bother, and run as system
As I understand it, programs needing privileges are supposed to run as system (or a similarly privileged user) as a service, and then applications are supposed to make requests of that service.
We can all point to examples in various OSes where you need to use the CLI to do something. WTF is your point ?
If there was a compatibility layer to run OSX applications on Linux, that might actually be a viable option. [...] But I am not aware of any such project so far
GNUstep aims for source code compatibility with Cocoa. In theory, the publisher (other than Apple) of a Mac OS X application could use Cocoa when porting it to GNU/Linux.
Totally true, I've always wondered about this myself. It should also be noted they use WINE quite a bit, so in supporting ReactOS, you would also be supporting WINE because they give back to the WINE community. Which reminds me, while I have been submitting App reports to the WINE Project, I still have a closet full of Games I need to get busy testing and making reports on. =)
If about thirty of you Linux devs would join in part-time, I bet they (ReactOS) could get just a little closer to their Goal; if you have some spare time of course. ;)
could use Cocoa
should have been "could use the subset of Cocoa supported by GNUstep"
Wine is binary-compatible with many Windows applications. ReactOS aims to be binary-compatible with both applications and drivers. This way, printers with no CUPS driver and scanners with no SANE driver will still work.
Bad exampel NFS was never intended to be a userfriendly network filesystem. A ordinary desktop user would most likley use cifs samba with a nas and use the gui to mount. However some linux distributions will let you set up nfs mounts during the install without edititing fstab manualy.
What those people don't seem to realize is that, just because no one has made a "Linux for the layman" doesn't mean no one ever will. Android is Linux, and it's about as user-friendly as you can get.
Android has what I believe to be one fatal flaw: all maximized all the time. Because of a poor decision that Google made in the Honeycomb era, namely allowing applications to assume that the screen area never changes after installation, the user can't split the tablet's display down the middle and run a phone app on each side.
Windows RT doesn't run legacy Windows software. It doesn't have an x86 emulator. Nor does it even allow developers of desktop applications to recompile them for ARM; the only desktop applications for RT are IE and Office.
But the user would still see ExFAT and NTFS formatted removable media failing to mount.
Because the code base is horrible.
Because the community isn't welcoming.
Because those things which need most change and are most doable for newcomers are permanently checked out but stagnant.
Because the core coders on the project seem to hate OO as a matter of principle.
Because there are better things to do than fight their process.
The turning point probably was that sony startet to put linux in all their products
Huh? Sony took Linux out of the PS3 as of system software 3.21 and sued George Hotz for putting it back in.
I must be really bored to fire up the email reader on my laptop. Usually all my communications go via my phone
Good luck typing a three-paragraph reply on a 4" sheet of glass. And good luck seeing your e-mail and something else on the screen at once. Or do you carry a Bluetooth keyboard for making replies?
... in the command terminal for about 90 percent of your time! ;-)
I guess I make the second. Who's the third chucklehead who uses Linux on the desktop?
They'll never be able to get the same number of blue screens as Windows gets, so there is no chance of ever catching up.
Because using Windows requires you to buy Photoshop and Microsoft Office? Since when?
Since every last "Why Linux Will Never Succeed on the Desktop" post that's ever been posted to Slashdot, that's since when!
Who is Value? If you mean Valve, you should remember that they were using their ancient Source engine, which is not GPU bound.
I call BS.
Win2k at least as of SP3 had full usb support, full direct3d support up to 9.0c, and support for SMP (but not hyperthreading, which was an XP+ only feature, and turned out to be unnecessary once multicore procs came out in the 775 P4/Core 2 era.)
Additionally, until XP specific feature began becoming mandatory, it was much faster than XP in almost all ways, just like win9x had been prior.
Honestly the only reason I stopped using it was because ATI dropped support for it on hardware >HD3xxx series and I'd just gotten a 4000 series card.
Given the 2k->xp hacks they've created since I could probably upgrade and continue running it today if I didn't have an XP license and only need one computer for gaming.
I'm pretty sure the graphics drivers in Linux run in kernel space...
Michael J. Ryan - tracker1.info
Many apps will store their settings in the user's AppData directory structure... For the most part I haven't seen much more in the registry beyond app install/uninstall information. There are some apps that are more badly behaved then others... also, you can export a key from the registry for import on another system.
Michael J. Ryan - tracker1.info
What you are talking about is pretty much what WinRT is... though MS only released it for closed deployments to locked down ARM devices... :-(
Michael J. Ryan - tracker1.info
Right. Linux users ragging on Windows 8 for a supposedly disrupting context switch going from desktop to the start menu, but switching between multiple OSes is perfectly fine and smooth...
It's not even that high. As of April 2013, Mac OS X has a combined market presence of about 7%. To put that into perspective, Windows has a combined market presence of about 92%.
Linux is fine for servers, portable devices, and embedded systems, but trying to stick it on the desktop is a foolish dream that has failed for over 10 years
While I can't speak for anybody else, more MORE than 10 years I've been running Linux, not only for servers, but also on desktops
Many desktops
Yes, my company does have Windows machines, but they are there because we can't find proper alternatives to some software that some of our people need, and that's the one thing the Linux community needs to improve on --- to persuade software developers to port their software onto Linux
Especially now that the market share of desktop computer is shrinking, software companies should understand that if they port their software into Linux, they gonna get a lot of new purchases from Linux users
Most Linux users of today are professionals - and have the money to pay for licensed software
In the case of our company, if only we could find software which runs on Linux, we wouldn't need to get Windows machine anymore --- plus we could use the money we save (from not having to pay the M$ OS tax) to upgrade our Linux machines
Muchas Gracias, Señor Edward Snowden !
Graphics drivers on Linux run in the kernel.
I bet you wish one could delete posts on /. ;)
"Politicians and diapers must be changed often, and for the same reason."
If he keeps coming up with terms like "chucklefucks" I heartily encourage him to stay under whatever rock he's stuck under. That was worth reading this entire flamewar thread right there.
WinRT is a locked-down anti-desktop walled garden to please Microsoft and friends. That's the exact opposite of what I would want to see. I'll take an NT kernel with a *real* desktop. One that doesn't purposely lock out potential non-Microsoft applications. RT is just a vehicle to get Metro going and to secure ARM so Microsoft can have it all to theirselves.
Most hardware engineers use linux exclusively... there frequently isn't even a Windows option...
I think his point is that he doesn't really need those applications to use his computer, and so Linux works fine for him. If you don't need windows specific applications, then you really don't need windows, now do you?
I cannot understand why anyone would call parent a troll. He is making a good point (in a genuine tone for those of you who care). A lot of us would love this OS.
I'll say it again for the remedial English crowd. That Linux *does* force you edit config files more often than Windows forces you to make manual registry changes.
"Ignorance more frequently begets confidence than does knowledge"
- Charles Darwin
It's plan 13
Plan 1-3 were various timesharing OS
Plan 4 was DOS
Plan 5 was Amiga OS
Plan 6 was Windows 9X
Plan 7 was Windows NT
Plan 8 was OS X
Plan 9 was Android
Plan 10 is being revised it was going to be Vista & Windows 7 but Windows 8 hasn't taken off
Plan 11 is Under NDA
Plan 12 is even more NDA buried
Plan 13 are 'far out plans if other plans fail'
Plan 13b is to create a NT clone and sell^h^h^h^hNOW FREE! on CD's^h^h^h^hDVDs^h^h^h^hDownload
No no no.
I've been using Linux for a long time. Editing config files is a must for everyone.
Unless:
1) You own a computer with an exactly perfectly supported hardware configuration (unlikely)
2) You never wish to install your own/non-package-managed programs
3) The default network configuration is perfect (and the NetworkManager/control panel hasn't screwed it up already)
4) There are no bugs in your distro to work around
5) Gnome never poops on your configuration files
This just never happens. Every time I do a "desktop linux" install, I try, really hard, to not edit configuration files. It ALWAYS comes up.
I don't hate it, but let's not be delusional. The linux gui tools are not perfect. End of story.
If you were to ask my opinion on a solution:
1) Big distros need to get together and spend a lot of time on configuration management and configuration tools.
2) Store all config in a locally run light SQL server. The access is a local XML socket. This makes it really easy to design all sorts of neat tools to manipulate the machine's configuration.
Yes that's idealistic and would take years to work out.
I think the point here is that the design isn't broken because you *can* do so. The system will still be functional enough to allow someone with the right knowledge to fix it (remotely if need be) and in situations where this matters, people with the right knowledge are a phone call away and would not need to move to fix it.
If a linux box has a borked gconf, some dude across the world can make it magically boot a few minutes later. Not to mention that the user can only bork *his own gconf*.
If a windows box has a borked registry, you need a human with some level of technical skill at the computer or to pay extra to have some sort of lights-out management built into the computer which is no excuse for poor design.
Mind the frickin' laser...
sudo echo "server:/share /location nfs 0 0" >> /etc/fstab
Mind the frickin' laser...
If a linux box has a borked gconf, some dude across the world can make it magically boot a few minutes later.
If a Linux box behind NAT has a borked gconf, then how is "some dude across the world" going to connect to said Linux box to make it boot?
No games.
Your argument against ReactOS falls apart with your second stated reason, nice job.
"ReactOS (as well as mono and wine) are destined to be playing catch-up with whatever Microsoft does. This puts them always one step behind their competitor."
Clearly you don't see the advantages to this, but the ReactOS folks do, seeing the MS has continually brought out "broken by design" software and then scrambles to fix things because the user base is slow to 'adopt' their crapware, the ReactOS group will have plenty of time to "fix" it well before MS can react.
You see liabilities, they see low hanging fruit to be picked.
This written using Win7 Starter, a crippled crap OS that came on my Toshiba netbook, so I'm no LINUX 'fanboi', just a realist.
Wow. You used NFS as your use case? Really? Have you had the misfortune of actually using the half-hidden Mac OS X NFS mount configuration tools? Or perhaps the UNIX Tools for Windows package? fstab is heaven compared to those. I'm not even entirely sure whether you're trying to argue for or against configuration files with that kind of question. Heck, NFS has had absolutely zero focus on ease of use since it's inception and yet it looks like you're trying to use it as some kind of proof that Linux forces you to use configuration files. That seems a bit disingenuous, if you don't mind me saying so. That brush is a teensy bit broad.
Alright, so you found a situation where editing a text file is the primary method of accomplishing a goal.
Now, how do you set up an NFS share on a Windows machine, and does it take longer to do this than to edit fstab?
P.S. - Don't try and change the problem to a simple Windows (SMB) file share - most Linux distros nowadays can configure Samba from the file manager. You specifically asked for NFS :).
The registry and config files are not really an apples to apples comparison.
In the first place, the windows registry isn't meant to be modified by the user directly. The naming conventions are short, difficult to understand, and sometimes entirely opaque to the user. The meanings of constants can be impossible to discern (bit flags, masks, GUIDs/UUIDs, etc) and there's little variety in the types of values leading to sometimes drastic measures to fit as much parsable data into a field as possible while not worrying about user friendliness. Not to mention there's very little documentation as to how each program uses the registry, and if you actually found some, then you're pretty lucky.
Configuration files in comparison are intended for user consumption and are far more verbose. The purpose of configuration variables and constants are generally easier to understand and very verbosely documented. Their layout is also generally logical and in the case of very large config files, sectioned and with clearly indicated defaults. Really nice projects will also usually include an entire manual for each major configuration file.
The only thing that the registry and config files really have in common is that they both happen to store user settings. While I agree that the registry gets a lot of hate, comparing the registry to config files is no less misguided. If you want a real comparison, I would suggest at least taking a look at gconf or gsettings first.
"The program 'regedit' is currently not installed. You can install it by typing: sudo apt-get install wine1.4"
Tomorrow is another day...
YaST works for most of the stuff you would want to do in a desktop operating system. The SuSE Linux Enterprise Desktop (SLED for short) distribution is the best example I have of a decent no-nonsense Linux desktop.
It works well even for basic server configurations of Postfix, Apache and Bind and it detects (and respects) when you have manually modified config files.
The only instance I had to edit config files on SLED was when I had to configure SSSD to authenticate on kerberos and lookup for user and group information on LDAP.
Not having to use MS Windows to run these programs would be great.
"The likes of Facebook and WhatsApp are free to those whose privacy is of zero value."
Is joke, right? The whole idea of dethorning MS is passe now. MS is dethroning itself.
The code that managed context swapping is part of the scheduler, at least on Linux.
Well, yeah, it's in the scheduler, but since it's a single instruction there's not much scope for fucking it up. The problem is that switching process context means changing CR3, which means blowing out the TLB, which means that you have to re-load the damn thing when you're finished, and that's not a cheap operation.
Having said that, it is possible to get good performance with userspace display drivers, but it has nothing to do with scheduling. The trick is to pull the driver into a library in the client process, with a minimal set of stubs in kernel mode. Even then you have to trade off latency and throughput far more carefully than you do in a pure kernel-space scheme.
Linux is fine for servers, portable devices, and embedded systems, but trying to stick it on the desktop is a foolish dream that has failed for over 10 years
Hmm. not sure if you are trolling but I have been using Linux as my principle desktop for over 5 years and as a professional engineer I can easily work within a Microsoft centric environment. Some of the main reasons why MS Windows dominates over Linux on the Desktop is policy making Management in the majority of cases don't know there are alternatives or the company is effectively locked into using MS Windows. For the home user most people use MS Windows because they already have it installed on their PC (The Microsoft Tax) and don't really know or care about installing a Linux distribution since it may damage their brain cells :). Of course if the home user is a "Gamer" then in the majority of cases they want to play "Games for Widows" which even though some games can be run under packages such as WINE most gamers will not make the switch.
There ain't no such thing as proprietary standards only proprietary formats. Standards are by definition open.
It is not necessary to catch up. The main argument for using Widows is backwards compatibility. If the new Windows version isn't backwards compatible then the user is free to choose another operating system.
Locking files all the time and not making it easy to see who or which process has locked them.
Also, why are most dialog boxes in windows not resizable?
You're a temporary arrangement of matter sliding towards oblivion in a cold, uncaring universe
How do I setup an NFS mount point under Linux without editing /etc/fstab by hand?
What you have said is totality meaningless unless the following is true:
/etc/fstab is the least of your problem and one way or an other the file must be changed since this file controls what file-systems (remote or local) are mounted.
If you want to use NFS to manually or automatically mount a file-system you need super user privilege to do this. If you as the person requiring an NFS mount has not arranged to have a remote file-system made available with the correct permissions (ie "trust") to the particular machine they are managing then they will not be able to mount the remote file-system. Editing
BTW even user MS Windows you still need permission from the server machine and must authenticate accordingly to mount a share.
There ain't no such thing as proprietary standards only proprietary formats. Standards are by definition open.
Never argue with a fool. He'll drag you down to his level and then beat you with experience.
Windows is the fool of operating systems. Doing things its way is incredibly expensive. It is light that defeats the dark. If you want to beat Windows, you can't follow its winded path, or you'll never come out ahead.
It's obsolete, due to virtualization.
You are assuming that everybody is willing to pay the Microsoft tax and/or pirate some Windows copy just to run some Windows application. Apart from which I seem to remember that consumer-level Windows copies were not licensed for running on virtual machines, anyway.
It is absolutely possible to have high performance userspace graphics, as was proven with some of the more up-to-date drivers. I think it was ATI that first did it, by the way.
Absolutely not. Read this and understand it (first link I found that was useful): http://www.cs.rochester.edu/u/cli/research/switch.pdf
From the summary: "In general, the indirect cost of context switch ranges from several microseconds to more than one thousand microseconds for our workload."
If you can skip that "several microseconds to more than one thousand microseconds" blip, you will get MUCH better performance. That is why graphics drivers are in the kernel and why you can NOT say that you can ever get high performance user space graphics drivers (under current architectures).
"Someone needs to talk to the tree of liberty about its ghoulish drinking problem." by ohnocitizen
Not to mention everyone is ignoring the elephant in the room, the reason why Linux has never stood a chance...Windows software.
There is ALWAYS some "must have" piece of software that Linux doesn't have any alternative to or the alternative is poor, and as long as that is the case you can just give it up. the FOSS community has acted like all you need is a browser and Office but even my little old lady customers have software that they simply won't do without, and its pretty damned obvious by now that Linux is NEVER gonna get enough marketshare in its current form to get all that software ported, its just not gonna happen.
So if the FOSS community really and truly wants the world using a FOSS OS? Then get behind ReactOS, because that is pretty much the only shot you have of dethroning Windows on the desktop. Hell you make it stable so "update foo broke my drivers" or even better if supporting windows drivers isn't a problem I'll be happy to stock my shelves with ReactOS boxes and laptops, and I bet I'm far from alone. I really get damned sick of the FOSSie faction saying I don't want Linux to succeed or that I'm working for MSFT...do you have ANY idea how badly they fuck us system builders over? The OEMs get Windows cheap enough they can pay for the cost in trialware, us system builders have to pay the same price the average Joe pays and it fucking sucks. Don't you think I'd rather sell systems without that high as hell cost of Windows dragging like a boat anchor behind it? I'd even be happy to pay say $20 a pop for each copy of ReactOS, sure beats the wallet raping i get from MSFT.
So c'mon FOSS community, get behind ReactOS. You get the code you are always wanting, we system builders get an OS that doesn't bleed us dry, the consumers get an OS that runs what they want to run, hell you could even have a dialog box pop up pointing out a FOSS alternative to this or that software when the user first goes to install if you want to wean them off of non FOSS, its a win/win for everybody.
ACs don't waste your time replying, your posts are never seen by me.
Um, I do not think people hate the centralized repository for config files, after all, they use an /etc directory and file structure without complaint. The problem with The Registry is that it is a binary file format that is impossible to decode or repair without a specialized tool that requires a fully functional environment in to utilize.
Regards
"Someone needs to talk to the tree of liberty about its ghoulish drinking problem." by ohnocitizen
I'm pretty sure the graphics drivers in Linux run in kernel space...
They don't. Some basic parts of the device switching/access run in kernel space, but the bulk of the graphics code doing "real work" is running in user space. That's the reason that switching between several X servers (partly with different accelerators and/or versions) on different graphic consoles works without much of a problem, and you can even debug/resynchronize systems with locked-up/crashed graphics drivers via network access.
Try that with Windows.
Personally I prefer config files and use Arch mostly. But anyway.
1) You own a computer with an exactly perfectly supported hardware configuration (unlikely)
Most recent install was ubuntu 12.04 on a server. Until I wanted to start doing crazy stuff, like setting up PXE booting on a private network on a bonded ethernet pair, it was a config-file free install. All the hardware worked fine.
My work laptop (a thinkpad) also worked fine under ubuntu with no config file hacking.
2) You never wish to install your own/non-package-managed programs
What? This is one of the things you don't have to edit config files for. Basically, uncompress them into /opt and go on your merry way. Or ./configure&&make&&make install to do it from the source. Either way, I don't believe I have ever had to edit config files to install out of distro software, even on Arch, for which config file editing is expected.
3) The default network configuration is perfect (and the NetworkManager/control panel hasn't screwed it up already)
This is a funny one. For default networking (nothing beyond simple connections) network manager works okayish but is a bit flakey. Actually one of the irritating things about it is it is much harder to customize via config files, to the point where I never bother.
2) Store all config in a locally run light SQL server. The access is a local XML socket. This makes it really easy to design all sorts of neat tools to manipulate the machine's configuration.
But I already have lots of neat tools, like vi, diff, rcs and git. They work fantastically well and I love it that way. If the penalty for Linux becoming popular on the desktop is that it sucks as a desktop for hackers, then I'd prefer it stay off the desktop.
SJW n. One who posts facts.
BSD is UNIX, and Linux is "Unix like"
Windows provides Notepad and a calculator out of the box. This should be enough for anybody.
Windows 95 OSR2 had USB device support,
Ah yes, I have vague memories of that.
Pretty much the best supported devices were USB blue screen buttons. i.e. you plug them in and it bluescreens. That worked quite reliably IIRC.
SJW n. One who posts facts.
Tell us about your hairy balls challenge.
I guess my experience over the last couple of years differs from you then. The Windows GUI tools aren't perfect for that matter, either. I've had to tinker with registry settings often enough which is morally equivalent to editing a config file due to various bugs. No it doesn't happen often - but it happens with Windows nonetheless, quite often with XP, quite rarely with Windows 7 (but I've still had to do it once or twice).
The amount of config file tweaking I do with my desktop Linux systems is no more than with Windows 7. The level of config file tinkering is pretty much at an all time low. My work workstation is currently running Debian 7 with Gnome 3 and before that it ran Debian 6. I've not touched a single configuration file by hand, and I've had the machine a year - it just works (including the proprietary ATi graphics driver, which has a GUI installer and also just works). Perhaps it's because I'm running Debian rather than the bleeding edge. I also have Debian 7 on a virtual machine on my Macbook so that I can run the Xilinx ISE software, again, I've not touched a single config file on that either, it also just works. Not only that, the Debian desktop auto resizes when I change the VM's window size or full screen it or plug into a monitor, all without having to even click an icon let alone a config file! And when I get a software update, all software gets updated, not just the OS but all the applications I've installed from the Debian repositories, so office tools, compilers, the lot.
Incidentally, I don't get all the hate over Gnome 3.x. Admittedly I never used the earliest versions of Gnome 3 (after all, I'm on the trailing edge with Debian) but I got used to it within a half hour and rather like it. Perhaps I'm strange.
The only place I edit configuration files on Linux is on servers. And on Windows servers, I also have to do the same (there's an awful lot you can't do with the GUI on a Windows server, or is just 100 times more efficient to do with a script).
The hardware I'm running for Debian desktops has been mostly HP (formerly Compaq) business systems, but my current work workstation is a custom machine kitted out to be able to run lots of VMs for testing and on none of this stuff have I had to tweak configs by hand. I ran CentOS before that which wasn't as good in that respect, but that was back in the CentOS 5 days, perhaps 6 is better.
Oolite: Elite-like game. For Mac, Linux and Windows
Darling could co-operate with puredarwin and Fiasco.
I am sorry but you'd have to go out of your way to look for a Linux distribution that does not provide a better desktop user experience for ANY class of user than ReactOS.
I don't say that to knock the ReactOS project, lots of hard work has gone into that and I certain can see useful applications for it, someday. The fact is though that Linux only fails on the desktop now because the media cartels and device vendors want it that way. They don't make software and drivers available or if they do they don't make it available a package that is accessible to less apt end users. Then there is still Microsoft Office. XFCE provides a vastly superior desktop to Windows and OSX. KDE is very good too. Unless you need to be able to run your ten years of crufted Excel and Access documents with their 10KLOCs of VBA Open office is probably as good, and for users with simpler needs feature-wise AbiWord and Gnumeric are very easy to use and blazing fast on just about any hardware.
X.org pretty much configures itself now days and can give you a decent 3d accelerated desktop out of box with zero configuration in many cases.
No the reasons GNU/Linux is not on the desktop are
1) inertia
2) DRM
3) political
ReactOS is a long way from prime time. It for the most part can't run anything that won't also run on a Wine / X.org / Linux stack. The exception being drivers and the only real use case for that is if you have some really obscure special purpose hardware you need to run. I see that with lab and test equipment as well as industrial controllers but not much else. I expect ReactOS will be a very good way to keep some of that stuff going in the near future as most of it is also incompatible with Windows 7+ and XP is not getting patches for very much longer.
Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
Fine - change your video driver then without editing xorg.conf. Like, say, to change from a generic VESA driver to the nvidia driver.
"Ignorance more frequently begets confidence than does knowledge"
- Charles Darwin
I wouldn't say "broken by design," but I'd say it's very poorly designed. I have Win7 on this notebook and kubuntu on my tower. The notebook has twice the memory and processor speed as the tower, but takes twice as long to boot. It will be running Linux when I clear enough disk space, it's really getting slow. The tower is actually faster than it used to be, Linux upgrades speed things up rather than slow them down.
I don't mind powering down the tower. When I get up I hit the power button, make coffee, and it's just as it was when I shut it down; it enters the password for me (I live alone). When it gets updates I click once and keep working, no reboot required.
Windows, otoh, God but I hate Patch Tuesdays. The notebook is unusable for twenty minutes as it downloads the patches, then sits there trying to close programs when you do the mandatory reboot; that's just shitty programming. It should close the damned programs one at a time, giving the "save data?" dialog, THEN reboot. And a reboot should NOT be necessary.
To do anything in Windows takes twice as many steps as in KDE. Windows is a pain in the ass. IMO if Windows didn't come with every new computer few people would use it.
And I hate Windows astroturfers, spreading misinformation about Linux.
Windows is prettier than Linux, but I want my tools to be functional, not pretty.
Free Martian Whores!
The code we want has nothing to do with Windows. Wine would be far better for everything you are talking about.
It's not gonna work. No matter how much resources you pour into ReactOS, it's always going to be a copy of Windows and chasing taillights. Why go for the clone if you can get the original? It didn't work for OS2, Wine and Mono.
What would be the benefit of the customer to run a clone? Would you lower the cost of your machines in line with the cost for the installed ReactOS or would you use the bulk of the savings to up your margin?
# touch universe # chmod +rwx universe #
Claiming too?
I have 0 windows machines in my house. They are flat out not allowed. The fact that I do not need some of these applications makes that easier, but it is clearly not impossible to do.
There is a lot of cross-pollination between Wine and ReactOS.
How many digitally illiterate users? None. But how is that different from Windows, OS X, *DOS, *BSD, Haiku, etc?
For these people, there is only the option of paying someone else a nice sum of money and get a freshly installed machine back.
It doesn't matter what "borks" the machine, be it a technical problem or general user confusion over a slightly changed icon or location. These people are already up the creek when they turn the machine on.
# touch universe # chmod +rwx universe #
I would much rather have a myriad of text-based config files (all in a known place /etc) than have a monolithic nearly opaque binary database. For one, it separates concerns and makes it harder for a screw up of one config to bork something totally unrelated. And it makes it easier to be able to manage by hand (especially when the de/installer is brain damaged). In that regard, I think Apple's application bundles which keep everything needed for an application in a single spot is the best of all. Want to install, just drag the app folder to the disk. Want to deinstall, just drag the app folder to the trash. Simple, separate, and elegant.
I mostly agree, with two exceptions. Well, one exception that has an interesting corollary.
There are places where Windows isn't run in order to run Windows, but because there's some kind of "appliance computer" (dental records, machine control, whatever) that needs to run a very specific set of software and nothing else, and that software happens to run under Windows.
This could let vendors who sell those "appliances" stop paying a Microsoft tax, lowering their costs.
The specific case of this that could be of most interest to Slashdot readers: imagine a ReactOS Steambox. Imagine if the Steam client knew about ReactOS, and in cases where a game wasn't compatible, made that clear. (So, you'd get less games than you do via Steam under "real" Windows, but more than you currently get via Steam under MacOS or Linux.)
It's sad, because like you said, the NT kernel itself is from what I understand a very capable kernel. It's just unfortunate that there will probably never be a truly good, modern, security-focused and multi-user-focused operating built on it..
You mean like ... GNU/NT ?
It's an oft cited reason for why people need to use the Windows Operating system, that can't be duplicated by current Gnu/Linux or BSD currently.
At this point, once they get the OS ready, it wouldn't be chasing anything. Its 32-bit version just has to target XP, and its 64-bit version just has to target Windows 7. No need to ever bother about 8, 8.1 or anything else.
A function of the price of macs. Had Macs been cheaper, Apple would have had a much larger market share. But they do have to fund all the work they do, and unlike Microsoft that does it by selling the OS, Apple does it by selling the box, and tying the OS to just that bos (the Hackintosh experiments notwithstanding). But that notwithstanding, OS-X does have reasonable software support, even if that's nothing near Windows
I'm assuming that this discussion is about running OS-X software on non-Apple hardware (since Macs would run OS-X software on OS-X by default). In which case, let's look at the choices.
If the user doesn't have software that is Windows only but is there on OS-X as well, s/he's in luck. Rather than Linux, I'd look at whether it could run on PC-BSD. After all, OS-X has FBSD underpinnings, which would be supported in PC-BSD, so one just has to check whether what Apple specific APIs or anything else that is needed in order to support this. There would also be the differences b/w Apple's XNU (essentially NEXTSTEP) and FBSD kernel. Build that compatibility layer into PC-BSD, and then you have an OS that can natively run Apple software (not even requiring OS-X jails) and yet run on standard PC hardware. Of course, one does have to check the hardware compatibility list.
This would be a lot more painless than trying to build a compatibility layer on Linux, where the differences b/w GNU and FBSD userland has to be factored in addition to the differences b/w Linux & XNU, as well as the APIs.
So, anyway, not ALL drivers are part of the kernel, more and more are moving out of it as time goes by. But yes, many drivers still are. Our Minix legacy.
I'm being a bit pedantic here, but Minix was Tanenbaum's microkernel system, so kernel-mode drivers are more like Linus' legacy...
An absolute necessity for performance reasons. They tried doing it in userspace in NT4 and it just couldn't keep up.
Actually NT4 was already the first version to move GDI into the kernel, because of (as you and others have already mentioned) the high cost of context switches and marshalling/unmarshalling in NT 3.5's microkernel-style display server architecture. See this TechNet article.
I believe the point was that to the average Joe user, not getting a GUI is just as good as not booting at all. Those of us who love the CLI wont have any problem debugging a broken X but my Mom would call me and tell me her computer wont boot if it barfed some errors and dumped her at a shell login prompt.
- No Bounce, No Play -
I worked in a callcenter where a Win16 program vastly improved the team's ability to do the job (it was connected to the phone system we used). Wine (ReactOS on Linux) ran the program much better than Windows 7 in any mode, so many people installed Linux machines for the improved legacy Windows compatibility when they found "Wine for Windows" didn't exist.
Science & open-source build trust from peer review. Learn systems you can trust.
Newer versions of desktop Windows you pedantic arse.
Microsoft is referring to Windows RT tablets as "PCs" on this page. This gives me a bad feeling about what Microsoft plans to do to versions of desktop Windows after 8.1.
What chasing? As another pointed out target the 32bit and 64bit at Win 7 and call it a day, nobody is buying Win 8.blah blah unless they are forced to with a new system and a lot of those folks end up coming to me going "Take that stupid thing off my system!".
As for the benefits to the customer, lower prices and having an OS that a company isn't just gonna pull the plug on so they can sell them another copy? I still have WinXP P4 boxes in one corner of the shop for sale...do you think I WANT to have systems that I know have less than a year before they have a giant fucking bullseye painted on them? Think I wouldn't prefer to give them an OS that is gonna be updated past Apr of next year?
The simple fact is not everybody has the money for a new system in this shitty economy and I try to have systems at all price points just to make sure that anybody that walks into my shop can walk out with a computer. Can't run Linux on them thanks to the shitty driver situation and how often an update pisses on the drivers, not to mention the hardware roulette where you have no God damned clue what printers, wireless cards etc will run on the damned thing because the lists are ALWAYS out of date and require knowledge no average user is gonna know like which fucking kernel version they have (which just shows how shitty Linus Torvalds is at dealing with drivers, you shouldn't HAVE to know that shit just to get a functional device) so I literally have no other choice, its leave XP on them or have no systems cheaper than $250 in the shop.
And I make a grand total of $60 on a new build thanks to the boat anchor that is Windows licenses, if I didn't have to pay that wallet raping for a copy of Windows Home? I'd still make $60 a build and the price would lowered by the difference between the cost of Windows and the cost of ReactOS which like I said I wouldn't mind paying $20 to $25 a copy just to help support them, after all I was happy to pay $40 a copy for Xandros Home, only damned Linux I ever saw that could do two in place upgrades and have all the drivers keep working.
The simple fact is we system builders really don't have a choice in the matter, Linux is too damned fiddly, breaks too often, too damned many problems have as their ONLY solution "Tweak this pile of CLI crap (which no home user is qualified to do) because it was designed for hardware B,rev C and thanks to Torvalds sucking dick on drivers it won't work if the hardware is hardware B, rev D, and once you are done tweaking input this bullshit into bash and pray to St iGNUcious" and its pretty God damned obvious by now you can give it the fuck up when it comes to having either the consumer software or hardware companies support you, its NOT gonna happen. And NO Wine is NOT an option, its just as damned fiddly, fails as often as it works,supports too little, and requires too much of a consumer when it comes to skill to be a viable option for anything but a handful of software like old versions of MS Office which most consumers don't even use.
So we system builders would be happy to give the finger to MSFT, we just don't have that option because there isn't anything like reactOS to take its place. instead we'll get another 500 distros on distrowatch that can all be summed up as "Its (insert Linux base) with (insert one of the same 5 DEs) with the same old software, the same problems as always,same fucked up driver model,same breakage, same lack of software, same old same" and its not gonna work, if it was gonna work it would have by now. So Windows will keep the desktop, Apple will own the mobile, and we system builders will keep having to give money hand over fist to MSFT.
ACs don't waste your time replying, your posts are never seen by me.
Neither is officially 'Unix-certified'
Why do you want to dethrone Microsoft? Why not come up in the operating system that is innovative and creative and runs MS applications in a sandbox.
I agree w/ everything you've written above except one thing. The current ReactOS project targets XP, and is win32 based: they'd do well to stay focused on that one. However, once they get enough people in their dev team, they should have some of them work on a win64 version that targets Windows 7, but there'd be no need to target the win32 one at 7 as well. Ideally, given enough resources, they should have a modular OS that could vary its memory footprint depending on the resources available on the host computer. Also, for a win64 based system, they could specify the minimum requirements at 4GB, since anything below that could use a win32 based system.
Of course you can get high performance user space graphics when you minimize the number of context switches. Coalescing the requests and so on, it's all pretty well known. I didn't say that those costs aren't there, just that you need to be aware of them so that you keep them in check. It requires a certain kind of programming approach when you write, say, OpenGL code where the driver is a userspace process. You need to keep server roundtrips to the minimum.
A successful API design takes a mixture of software design and pedagogy.
> ReactOS is a project to build a free,
> open-source clone of Windows,
I think you just answered your own question.
> Why on earth hasn't this received more
> support from the OSS community?
Most of us in the open-source community don't *like* Windows all that much. We aren't particularly interested in a clone of it, no matter how it's licensed. I find myself vaguely curious about whether ReactOS will ever reach a point of actually being a meaningful competitor for Windows, but this is the same kind of purely theoretical curiosity that leads me to wonder whether stovepipe hats might someday make a fashion comeback. It's a purely abstract curiosity. I don't actually *care*.
> Linux is fine for servers, portable devices, and
> embedded systems, but trying to stick it on the
> desktop is a foolish dream that has failed for
> over 10 years.
Whatever, dude.
Linux was *designed* for the desktop. The fact that it's good on servers and embedded systems is just a nice bonus. I've been using Linux on the desktop since the late nineties. My computer does what I want it to do, and the OS stays out of my way and lets me do whatever I'm doing, and I don't have to jump through a bunch of stupid hoops all the time. This is partly because I have my desktop significantly customized -- it wouldn't be so good on an out-of-the-box install; but said customization is *possible* because I'm using an inherently customizable system. Windows allows you to customize the color scheme and mouse pointers and stuff, but anything that would actually have a major impact on how the software operates, forget it.
Now, granted, a lot of people prefer Windows. But most of those people aren't open-source developers. This is not a coincidence. Windows was *designed* to appeal to people who are NOT computer geeks. That was kind of the whole point, actually: regular people wanted to type up papers and stuff but didn't want to learn technical stuff (e.g., the command line -- which is significantly simpler than programming but still overwhelmingly more technical than anything most Windows users will touch with a ten-foot pole). Windows was made for regular people.
But most programmers, it turns out, don't really think that way, and Windows tends not to appeal so much to most of them.
There are, of course, exceptions.
Cut that out, or I will ship you to Norilsk in a box.
"graphic routines" in this case, doesn't just refer to drivers.
Glad to know you have a vulnerable desktop.
Most recent install was ubuntu 12.04 on a server.
Well yeah. It's a server install. There are no high-end graphics cards, OpenGL issues, sound cards, etc etc. Linux totally rules the server world and I will never even ask for a GUI tool for a server install. Give me vi and a VT220 and it's over.
All I'm saying, is that for me, "desktop linux" installs have been similar to what I am used to on the server end of things. Not a problem for me, but definitely a stopping point for some folks.
Your laptop experience is nice -- I haven't had one go that well. Thinkpads are popular hardware with linux desktop folks though, so I would say you got as close to "supported" as can be.
I'll respond to this based on my most recent desktop linux box.
This is a Dell with two Nvidia graphics cards. The cards are not exactly the same, one is a '4500' model and the other is a '5000'. Or something. I run Debian -- almost always do these days -- and yeah, lots of things worked out of the box. Printing, Networking, hard disk label problems (/etc/fstab stuff), and definitely graphics -- these all had to be manually adjusted. Graphics still have issues, multi-head with two different nvidia cards is hard.
And then there's this, since I'm on 64 bit, apparently, there are some funny GTK bugs that slip up every now and then. Printer dialogue boxes come up empty. People's flesh on youtube videos is blue.... I could go on.
Please don't misunderstand me, I am a linux admin on the server side through and through (+15 years unix experience, nearly all flavors). But on the desktop side of things, I'm all Mac or, occasionally now, Debian Linux. I don't mind the occasional bug in the linux desktop experience, and I even enjoy tracking them down and fixing them. But I think there are many users that wouldn't.
To continue in the pedantic mode, Minix was not microkernel in versions 1 & 2, but is now, in version 3. So what Linus wrote from scratch to emulate was pretty much monolithic, and he's on record as having rejected a microkernel as the way to go.
Windows NT 3.1 and 3.5 were close to microkernel, which helped at the time, since they also lived on CPUs that could handle microkernels better than x86, such as Alpha & MIPS. But as NT evolved, more and more things moved from userland to the kernel, making it less of a hybrid. That trend has been reversed now in Windows 8, since portability (to ARM) is again important to Microsoft, but honestly, I think Microsoft has lost the opportunity by not using the potential head-start they had w/ MIPS & Alpha years ago. Even though we had nothing close to 2GB RAM, had Microsoft made their current Windows 8 (I'm talking architecture here, not the metro UI) on the Alpha & MIPS, Windows would have been as portable and ported as anything out there.
ReactOS could do this for their 64-bit OS, whenever they do that. Have a microkernel below, maybe L4 or something, and build the win64 subsystem on top of it. Then just port the microkernel to different OSs. If they can do this, porting applications to various CPUs, such as MIPS, PPC, ARM would just be a recompile.
To continue in the pedantic mode, Minix was not microkernel in versions 1 & 2, but is now, in version 3.
Well I mean Minix was already microkernel at the point of the Tanenbaum/Torvalds discussion. (Kind of funny actually how the later replies to that thread talk about how legendary that same thread is.) Don't know what version it was though.
I would have ZERO problem with that as long as their 32bit XP version can support later editions of DirectX, which since its probably doing DirectX to OpenGL conversion using shims or some such that shouldn't be too big a problem.
A lot of the problems I have with WinXP boil down to the fact that its 13 years old, has had thousands of patches, and that makes it a hell of a lot more prone to breakage and less stable than Windows 7. If they can make a version of XP 32bit that is as stable as Win 7 32bit? Sounds good, i wouldn't have any problem with having that on my shelves.
But if they don't come up with a 64bit version the only systems it'll be useful for will be legacy as I have circa 2006 Pentium D systems in the shop that will hold 4GB of RAM no problem and pretty much every system since that time, with the exception of Intel Atom as Intel crippled those to make them less attractive, has supported 4GB of RAM or more. while I haven't looked to see how ReactOS supports PAE (or if it even does) I can tell you on Windows its really not hard to slam into the 3.2GB RAM limit of XP 32bit, not when 1GB graphics cards can be had for as little as $15.
One final thing though, making a hard set 4GB minimum frankly wouldn't be smart, as not only would you be forcing those that didn't have 4GB now but were planning on going 4GB later to use 32bit but as someone who has run windows 64 bit OSes like XP X64 on systems with as low as 2GB of RAM I can tell you that even without 4GB you will see benefits when going 64bit, for example having 64bit registers can make some tasks such as video transcoding faster. Ultimately it should be targeting the same system requirements of the OS they are wanting to replace, in the case of 64bit Windows its a 1GHz 64bit CPU, 2GB of RAM and a 20GB or better HDD. I'd say those system requirements are pretty reasonable and then the user should choose what is right for them.
ACs don't waste your time replying, your posts are never seen by me.
I live in a tower you insensitive clod!
Anyway VirtualBox image of ReactOS downloaded, I even have a potential use for it!
Linux on the desktops/laptops/notebooks/dongle-PCs (plural!) is successful for me and has been for up to ten years and that's what I care the most about; me :)
Android is for lolpenises.
Apart from using a paradigm of a desktop with overlapping bits of paper (folders, projects, tasks in hand ; whatever) where you pick up tools from the desktop .... yeah ; nothing like dissimilar.
I returned a previously-dead X86-box to the world of the working computing machine yesterday, by the not-magic of Freegle (you may know it as "Freecycle" ; I don't care about trivial branding issues) and the new owner asserted that he was visually impaired, so graphics performance was not an issue.
This reminded me that the surface "floaters" are not now, and never have been, the important parts of any OS. Different people do have different needs. Stephen Hawkins needs a better stereo microphone at about the same time that Evelyn Glennie needs a new set of loudspeakers and David Blunkett needs to replace his 7x3 pixel monitor with a much bigger 7x3 monitor.
My recently-deceased buddy, Charlie, spent years trying to persuade me that I really needed a VooDoo or better graphics card. Eventually he loaned me his own, and then was somewhat upset when I continued playing games in glorious EGA-equivalent graphics. I still have no real interest in advanced graphics, but it's hard to get a system which doesn't waste ridiculous amounts of copper, silicon and attention on drawing pretty pictures.
The horses (or their riders) do really determine the courses that are going to be taken.
Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
They needn't come up w/ a 64-bit version of XP, since that one never really caught on. In fact, this is the first place I've noticed breakage in the compatibility - like not being able to run older versions of Acrobat. Which is why I suggested one XP based OS, and one 7 based OS. The minimum for the latter could be set at 1GB, but no less.
The current project of XP would be useful for just legacy, which is why I suggested that they grow the team, split it and come up w/ a team that does a win64 based Windows 7, and then they'll have most of that spectrum covered (except the really ancient Windows 95 stuff)
Many Linux versions include Gui methods of doing this.
Linux on the desktops/laptops/notebooks/dongle-PCs (plural!) is successful for me
My point exactly.
But I honestly haven't really seen any compatibility issues between 32bit XP and 32bit Win 7, in fact when I switched my customers to Win 7 I had a grand total of 2 pieces of hardware that didn't work, both of which worked in 32bit but the customers wanted x64, and a few stupid programs that came down to DRM bullshit which of course probably wouldn't work in ReactOS anyway.
I could see them putting out the XP version for legacy, just as you say, but they really need to have some kind of 64bit OS because lets face it, systems that won't run more than the 3.2GB RAM limit are gonna be becoming more and more rare as time goes on. Hell the old junker I'm using at the shop as a netbox until I can find an AM2 board for that ULV Athlon x2 I have in the CPU drawer is a circa 2006 Pentium D and even it will hold 4GB of RAM if I actually cared enough to replace the RAM.
ACs don't waste your time replying, your posts are never seen by me.
I had 2 problems, which I solved w/ XP mode. One was Adobe Acrobat 6, which I use to make pdfs, which runs under XP but not 7. So I installed it in XP mode. The other was an old printer, whose driver worked in XP, but not 7. That too was installed in XP mode, but once it was, it can be accessed from 7 as well. However, Acrobat can't (I'm not talking about the Reader here, which is free for download.
For the 64-bit version, wouldn't a Windows 7 equivalent be adequate? That's what I was suggesting - a Windows 7 equivalent of ReactOS for win64 and an XP equivalent for win32.
most games are still Windows-only
Then buy a Linux box and a game console. Play indie games on the Linux box, and play major-label games on the game console.
Yes Win 7 X64 would be a good target but then you run into the problem of having to keep two lines running, the 32bit XP and the 64bit Win 7 and what I was trying to say is Win 7 would be a better target all around. As you noted XP X64 had serious lack of drivers and its pretty obvious from looking at the numbers Win 7 appears to be the last widely adopted Windows we are gonna see, the new direction being seriously off-putting to most consumers and businesses.
This way they could keep all the work they had done on XP as an "XP Mode" for Win 7 to increase compatibility and with only having the single target more of the work could be shared. I think continuing to focus efforts on a 32 bit only OS that is not only nearly 14 years old but which will be abandoned by MSFT next year really isn't the right way to go, as i pointed out systems half a decade old can already use more RAM than 32bit can take and once XP is abandoned software support will drop off sharply, just as we saw with Win98 and Win2K.
But no matter which route they take they really need to get on the ball, there is tens of millions or more XP boxes out there that is gonna lose support in just 10 months, if they want to be a contender they need to have a usable OS by the time XP hits EOL or all that work will end up being for naught as people have no choice but buy a Win 7 or Win 8.1 PC to replace it.
ACs don't waste your time replying, your posts are never seen by me.
So I guess it's two of them then, even less than I thought.
1. Run Synaptic (or whatever GUI package manager).
2. Install nvidia driver package.
3. Log out.
4. Log in.
No xorg.conf editing required. That hasn't been necessary for years.
"Those who consume the bulk of goods are those who make them. We must never forget this secret of our prosperity."
No no no.
I've been using Linux for a long time. Editing config files is a must for everyone.
Unless:
1) You own a computer with an exactly perfectly supported hardware configuration (unlikely)
Actually it's very likely.
2) You never wish to install your own/non-package-managed programs
Most users probably won't. If they do, WINE doesn't require editing config files, and neither does unzipping a .tgz of a binary build and double-clicking the executable, just like in Windows.
3) The default network configuration is perfect (and the NetworkManager/control panel hasn't screwed it up already)
This is the closest you get to truth, but even then, I haven't had to edit a conf file for a NetworkManager problem in a very long time. This is basically a solved problem. It's certainly no worse than Windows.
4) There are no bugs in your distro to work around
This is so broad and vague that it's meaningless. As if Windows has no bugs to work around!
5) Gnome never poops on your configuration files
Why are you using GNOME? "Well there's your problem!"
1) Big distros need to get together and spend a lot of time on configuration management and configuration tools.
Basically done. e.g. KDE's system-settings is as good as OS X's System Configuration.
2) Store all config in a locally run light SQL server. The access is a local XML socket. This makes it really easy to design all sorts of neat tools to manipulate the machine's configuration.
Yeah, let's store simple configuration in a binary file only accessible through special software, and while we're at it let's put XML in the way! That'll fix it! Dude, you destroyed any credibility you had by advocating this.
"Those who consume the bulk of goods are those who make them. We must never forget this secret of our prosperity."
I don't think Windows is in any way prettier than Linux. Uncustomizable UI, single-pixel fonts everywhere, white, white, white windows and dialogs...I could go on. Linux, on the other hand, can be themed to be as pretty or ugly as you desire. Mine looks better than Windows or OSX.
"Those who consume the bulk of goods are those who make them. We must never forget this secret of our prosperity."