Slashdot Mirror
Will Linux For Windows Change The World?
An anonymous reader writes "A month ago, a trial version of a little-known Linux application called 'CoLinux' was released that is the first working free and open source method for optimally running Linux on Microsoft Windows natively. It's the work of a 21-year-old Israeli computer science student and some Japanese open source programmers; in Israel, analysts are already saying it could help transform the software world." (CoLinux is short for Cooperative Linux; we mentioned this project in January as well.)
The biggest benefit I see is that people could start running (and liking?) Linux applications without having to make "the big switch." Once they realize that they like Linux better and [hopefully] can do everything they need to under Linux, then the next computer they buy may run Linux alone. It's certainly more elegant and appealing to current Windows users than just telling them they're unsophisticated dolts for not using Linux.
Sigs are dangerous coy things
it would seem more productive to do this in reverse... that is to say, windows running under linux... not simply a compatability layer [wine] or an emulated system [vmware] -- it would be cool to see the NT kernel running as a process under linux (just as linux ran under mach in MkLinux, or OS9 runs under OS X)... it would probably be a lot faster to reboot that way... ;-)
-m
Perhaps if DirectX actually was inferior, and if it wasn't the primary or only API for 90% of the games out there, you'd have a point.
"If I need to use Linux Compiler while sitting on a Windows box, I would rather use vmWare. Also vmWare has made great progress in their GSX and ESX, to make all this very easy."
VMWare is like $300. CoLinux is free.
"Derp de derp."
A big part of what keeps many users from switching is fear of being in a totally new environment that don't understand. This provides a midpoint between the two worlds: get a taste of Linux, and if you start to panic just hit the good old "Windows" key on the keyboard and you're back to familiar territory. (Or CTRL-ESC, but chances are if you're running Windows you have a keyboard with a "Windows" key...)
.conf file that you have to edit by hand.
Of course, the second biggest part of the hurdle is customizing the system without having to learn all the nuts and bolts of operating system function. This is *almost* solved, but compared to the rather intuitive and standardized interface that Windows has nothing in the OSS community has been able to match it.
For example, tweaking options for a program should be done via an "options" menu of some kind there is a logical, visual organization to the settings with checkboxes and drop down lists, not a 30+ page
God help you if it's case sensitive or syntactically anal, too; you may never get it right unless you've done it several times before. Your average home user doesn't have the patience to deal with that kind of thing, and until this hurdle is taken down they'll stick with Windows for sure.
=Smidge=
How often do we, here at /., ask if a new software development is going to change the world? Constantly. And how often does it? Never.
This is no exception. It's just a sort of more native version of Cygwin. Sure, it could be kind of nifty, but it's not some major breakthrough which will leave the world shocked.
Could people please stop being so melodramatic with their subject lines?
definitely a good thing, because it might then encourage more people to take up Linux and have a look at it. It would give those people who are so 'married' to Windows a chance to look at what all the fuss is about, and to really evaluate Linux and see if it would be right for them. They wouldn't have to partition, re-format, re-jig their hard drive... and if things got too tough open up the appropriate Windows application to get their job done instead.
I also see it as a good thing in some corporate environments. Say you have a call center, and all the operatives have been trained to use some program for their task (let's say they're in a credit card environment) and their software is Unix based. Well, porting to Linux could be straightforward. Also for these operators they don't need to access the computer for anything much besides this application... and maybe the web and email to keep in contact with people. So these guys would have Linux desktops. Now there would also be some other administrative people who don't take calls, and who have other tasks. Like payroll, or some other fancy tasks. Maybe these programs were written for Windows, and there is no Linux port planned. Rather than trying to make these programs work through Wine or Crossover Office or something like that the obvious solution is to make Linux run on top of Windows. Then people have the best of both worlds for those kind of operations.
I also see advantages of running CoLinux in a dual boot environmemnt. That is, if you are short on disk space. I presume that CoLinux would run on the same filesystem as Windows. In a traditional dual boot system you might have a 20 gb disk, and split it up two ways - 10gb for Windows, and 10Gb for Linux. Let's suppose you are a Windows fan, and you easily eat up that 10Gb for Windows use, and hardly use Linux, except to 'play around with'. You then have 8Gb of disk space that Windows can't access natively (yes there are third party apps now that get around this) and as such you are short on space. So if Windows and Linux are sharing the same 20Gb partition, then Windows can use more than that smaller partition on those occasions it is deemed necessary (like downloading by broadband that 5Gb linux distribution on X # of CD's).
I don't see it as a "real major" security problem, because I perceive its main target is the desktop, and not for running security-critical applications which could get hacked to shreds. Also that these Windows boxes would be firewalled anyway for Internet access - behind native Linux firewalls on native Linux machines.
Mark.
Here is how i think this could have a great impact.
Very importantly, Linux running on Windows can be used to train Windows users on migrating to Linux. No messy dual-booting setups required. Just copy Linux and "click to start".
Secondly, this opens up Linux for sampling to many more interested users who are wondering what the hype is all about. I am not talking about the typical Slashdot geek here. Instead normal people with techie inclinations who want to try out things.
Thirdly, it is an easier way for running pilot trials of Linux deployments in a corporate environment. As no extra servers are required, no extra money needs to be sent. Although administration effort will obviously increase, it won't be to the extent of twice the administration effort of the original Windows server on which Linux is running.
One huge barrier to Linux adoption is that management does not want to do a trial deployment at most times due to the cost involved. This will certainly mitigate that.
DirectX is great for PC Games - but for real scientific/commercial work it *SUCKS*.
Much more money in PC games though I'm afraid. And as always, money talks.
cleartype fonts
"ClearType" is a Microsoft branding term. The generalized term, subpixel rendering, is definitely supported under XFree86 and x.org. LCDs with multiple layout order of the RGB elements are even supported.
2.) automatic directX compatibility for games
WINE does this, but honestly, while there are a lot of game developers out there that know DirectX, there's nothing particularly magical about DirectX, and it'd be pretty hard to "just" do DirectX without supporting the other chunks of the Windows API (though I guess you could do a "DirectX-like" API). OpenGL is a truly open standard that's widely supported (and preferred by videocard developers), and SDL and its child libraries provide a more modular system than DirectX does.
one solid universal gui
I see why you want it, but it's not going to happen. Too many KDE people like KDE (which is, while not unbreakably, still strongly tied to Qt) and too many people have legal issues with Qt or prefer GNOME for technical reasons.
Honestly, I don't think it's all that necessary, either. Windows users have been using non-Windows widget sets for a long time in major apps -- Lotus Notes or Mozilla or any of the standard Win32 variants, which operate differently over the Win95-WinXP lifetime. People adapt pretty well. Both Qt and GTK are pretty snappy. Both interoperate pretty well today. Two widget sets is hardly a reason for a platform to fall apart.
May we never see th
Fonts are fonts. I use Windows fonts in Linux. They look great. Big deal.
What do you mean by "the window manager have them?" My fonts look fine. In fact, try any recent distribution like SuSE, Fedora Core, Mandrake, etc., and I think you'll get the same impression.
Without DirectX, few games ever make it to Linux. Thats because DirectX is much more than just a 3-D gaming API. It has other features that make games easier to develop for.
OpenGL+SDL does as well.
Without a standard window manager and a standard API to program for (thanks GNOME vs KDE war), there is hardly any incentive for an application developer to go to linux. Sorry, its just too complicated to make it run correctly (across window managers).
Umm, are you implying that an app compiled against Gnome libraries will suddenly break if you try to run it in KDE? Actually, you can just choose the one you like best and develop for it. Copy and pasting will take care of themselves, and with good themes, they can look nearly identical.
What do you mean running properly "across window managers?" Window managers almost certainly could never prevent a program from working properly, unless they draw a border and buttons when they're not supposed to, for example.
So basically, you can't decide if you would want to program for Gnome or KDE, and you don't like the fonts that distros ship by default (even though haven't been an embarrasing smidgen on the Linux desktop for years), so you don't really think it's worth your time to develop for Linux.
I think it's more than fine to just say "hey, I'm doing fine developing for Windows, I don't have any problems with it, so I don't need to switch." So often zealots convince people on Slashdot that you ought to be ashamed of yourself if you run Windows, and while I disagree with your post and reasons for not choosing Linux as a development platform, I think it's totally fine to not choose Linux for no reason other than you're content with what you have :)
Slashdot: Where people pretend to be twice as smart as they really are by behaving like children.
No, really? DirectX was designed explicitly for games. That means that early in its life, it sacrificed accuracy for speed (compared to OpenGL, which took the opposite approach and didn't really gain speed on consumer hardware until 3D accelerators took off). Even now, DirectX is driven by games and multimedia, not CAD and scientific/engineering requirements. There's absolutely nothing wrong with that, and in fact it's better for games that it's focused on games and multimedia rather than engineering applications, because the requirements for games are different.
If you're writing scientific software, use OpenGL. If you're writing a (Windows- or XBox-targetted) game, use DirectX.
Oh, yeah, it's also possible to use DirectX and OpenGL together. Like SDL, DirectX is an entire framework, not just a 3D rendering interface. Id and theCarmack use DirectX for input and sound while rendering their 3D visuals in OpenGL.
2.) DirectX is a MICROSOFT ONLY format. It will never, ever, be in any linux distro except in emulation form. And for second, why should it be? OpenGL is fine and great, and with 2.0 coming out you can stuff DirectX where the sun don't shine.
At its very core, DirectX is just a set of APIs. Yes, it's a Microsoft API, but the exposed interfaces are well documented, and ignoring any possible legal issues, it is entirely possible to write a DirectX implementation on another platform. Okay, some of you may disagree on whether or not DirectX is well documented, but it's documented well enough for emulation purposes.
There are wrappers available that translate Direct3D calls into OpenGL calls (similar to Glide wrappers from the 3dfx days), and I don't see any technical problems with removing the OpenGL layer and having the new Direct3D implementation call the graphics card directly. However, and correct me if I'm wrong, I think Linux 3D graphics drivers are currently all proprietary, so nVidia and ATI would have to provide the Direct3D layer.
Still, even with an emulation layer, why SHOULDN'T DirectX run on Linux? Ignore legal issues and Microsoft's desires. Believe it or not, there are some developers who've only used DirectX and not OpenGL+SDL. It's worth having DirectX on Linux even if only a tiny fraction of those developers decide to port to Linux. That fraction may grow, and after familiarizing themselves with Linux they may switch to other APIs that are better supported on Linux, such as OpenGL and SDL.
Distributions.
Distributions put these things together for end-users to enjoy, and any recently updated distro worth it's beans has either Gnome 2.4 or the Bitstream Vera fonts. In my not-so-humble opinion, they are far superiour to the fonts in Windows. Unfortunately, however, they look pretty horrid in Windows, if you ask me.
Don't you think you're being pretty unreasonable saying Linux w/KDE or Gnome is unsuitable for anything like this (fonts) when it's already been addressed? You can lead a horse to water but you can't make it drink, download a better distro plz.
"We invented personal computing." - Bill Gates
How do you know he doesn't use a GUI? Is that how you deal people deal with criticism - mod everybody who has no problems down as commandline geeks?
Ever since XFT2/fontconfig and the Bitstream Vera fonts have been released, I've been enjoying high-quality, subpixel antialiased fonts on my Linux desktop computer. I suggest you to upgrade to a modern distribution and use the Vera fonts.
Alternatives are not going away. What's wrong with making both GNOME and KDE so userfriendly that the user can find it's way no matter which desktop he's using?
Some people prefer simplicity while other people prefer power and bells & whistles. What's wrong with being able to choose what desktop you want based on your *preference*?
In case you don't want to choose - fine, use whatever default desktop is chosen by your distribution. You don't have to choose if you really don't want to.
As for standardizes interfaces: even Windows doesn't have standardized interfaces. Installers all look a little different from each other - fullscreen blue InstallShield, MSI, Win2k-style InstallShield, fullscreen Inno Setup, Win2k-style Inno Setup, WinSFX, WISE Installer, etc. etc.
An installation system that handles dependancies with no user interfaction is being worked on - see my sig. We're close to 1.0.
One of the strengths of Linux is options
Well, funny that you think this is a strength. This is IMO the main weakness of all Unix based systems. Too many options means you can be familiar 100% with your system, and yet you won't be able to operate (well) another Linux, because things are so different. Hence the difficulty to debug your mom's Linux on the phone because your freaking brother installed it and he's on vacation right now, so you have no clue how to drive your mom through the command line stuff. With Windows, if she has a Win98 and you too, you're on the same page. It basically boils down to:
If you want a huge userbase, and a lot of knowledge of your system spread around, present a homogen system. Heterogen system will look (from Joes SixPack's point of view) as different systems, and he will be - rightfully - scared. Joe Sixpack wants a system that works. Not a tetrazillion of options and choices. Joe is scared by choices by nature.
Write boring code, not shiny code!