I think it has a bright future, but it's not competing with the present or future Microsoft and will take a while before it's fully cooked.
I think their future will be legacy support in a plethora of use cases. -Servers: There's some legacy Windows Server 2003 stuff out there where they have custom in-house stuff that doesn't work in NT6.x. (Vista and above) -Business Appliances: There are businesses that invested like $50,000 in a vinyl printer and the company that made it doesn't make printers anymore and never updated the drivers past Windows XP. -POS Machines: There's a lot of machines that run Windows POS Ready 2009 (Windows XP) and business will use the same cash register until the end of time. If you ever been to a Boston Supermarket, you'll find their Registers still run OS/2. -Legacy Games: It just got DOS support and it might be compatible with 3D games once it reaches 1.0, but mostly legacy with something like Manhunt only working in Windows XP and it's in the ReactOS roadmap to make drivers designed for ReactOS to also be compatible with Windows XP.
Who knows, maybe post-1.0, they might band-aid some things to make it compatible with Windows Server 2012 R2 stuff so people can get a cheap NT Multi-CPU Workstation with something like KernelEx, but I'm setting my expectations back. OSes are hard to engineer and let alone reverse-engineer with the same ABIs and APIs, so it will take a while before it's usable. But, We really don't need much from OSes these days, just a few missing features. If Vista had DX12, security patches, Drivers, (U)EFI support and a tweak to how it handles multithreads and power, it would still be a viable OS after 10 years. So, if we can ever hope to have an open source Windows 7 for your legacy Windows Applications, you need to start with Server 2003 first. So, I think Windows 7 was the plateau of needed features for most power users, the only thing Windows 7 needs to be modern in my eyes after 7 years is just DX12 and CPU power management and there's really nothing more I need from an OS and I don't even need DX12, it's kinda a shitty API or it's not shitty and devs just recompile and have it be unoptimized, but that might create a stigma that DX12 is shitty and everybody will port to Vulkan.
ReactOS? Its still in alpha state...... I see too many times this name passing by. Its a project that exists for 18 years now and they almost got the same goals/features as Wine. That is already available in you're daily Linux.
When i first discovered it 8 years ago i was excited too. look where we are now.......
Like The_Cable said ReactOS can be done through wine.
I could see reactOS being aimed towards older hardware where something like linux would be taxing on the hardware but people still want decent security for there devices.
I'm not a fan of Wine because it only sorta works. Granted ReactOS doesn't work, I just think you shouldn't use "you can run windows programs that don't use an API newer than 10 years old" to sell Linux. For years, Wine has been sold as a "Windows killer", but devs weren't going to support DX9 forever and DX10 and above is more of a black box, so it's harder to reverse-engineer.
Once Mr. Neckbeard McBasement has finished making Linux compatible with OSX binaries, most of these problems will go away unless you're using Windows-specific programs, and hopefully by then, Wine will have gotten a little less shitty.
ReactOS is a nice idea but they've been at it in some form or another for over 20 years, and it's still in Alpha.
OSX is more of a moving target than Windows. Most OSX Binaries are sold through the Apps Store and the binaries are updating with OSX. If you want to shoot for OSX Compatibility, the highest you can aim for is Snow Leopard or Lion.
Well, the whole "alpha" and "beta" thing is arbitrary. Alpha used to mean "not feature defined" and beta meant "feature define, but not complete" and "release candidate" was "feature complete, but buggy". ReactOS has known what features they wanted for 15 years and I think it will take less time than 20 years. (I say 10) The only thing on their 0.5.0 roadmap is NTFS Read & Write Support, WDM Driver support, Printing and DirectX Core and most of those algorithms are already implemented in other projects.
Apple is just annoying and they know that. They do it on purpose switch stuff. I heard lot of modding community that wanted to create awesome stuff with apple, but gave up because apple kept changing stuff without saying why.
They still bitching because Bill Gates used to copy them in the 80s. They just make it impossible for a purpose, but we all know its just Unix, Just super expensive Unix.
About the Wine is sold as Windowskiller that was just a method to give Windows the finger. It was a way of for people to change OS and they needed a selling point to have people switch over to Linux. It failed....
But you address the biggest issue in open source. Instead of focusing what we got already we always get super excited about new stuff and ideas instead of fixing that what we got already and improve it.
I right now think bytheway that pass-trough and Exo-Kernell are the way to solve these issues. I think if ReactOS will be ready then it's probably already too late for it.
I think by the time Windows XP becomes legacy rather than just old, ReactOS should be mature enough to be a replacement where wine on Linux just isn't an option due to low-power hardware. At the moment, XP POSReady is still on it's extended support and for anything older than that, there's FreeDOS which works extremely well. ReactOS looks to me like preparation for the future of legacy applications.
It all depends on the expectations and if you can work with what you got. There's a project to reverse-engineer the Hitachi Super-H CPUI called "J-Core". (the CPU Architecture used in the Sega Saturn and Dreamcast) Now could people use that to make a Dreamcast clone or an awesome Xeon-Class CPU evolved from it like how x86 evolved from a 70's traffic chip? Probably :/ I wouldn't hold my breath, but it makes more sense to start as a patent-free CPU for people that would rather spend what they would normally spend on an ARM License into hiring engineers to build on something that's already there and get a few ASICs made from 15-year-old lithography machines that would cost almost nothing to make so you could make something Arduino class or something for IOT.
So, could ReactOS evolve to Windows 7 compatibility? Probably, but it's still best to take it one step at a time, once it's stable, start off small in legacy to get it's foot in the door and see where it goes from there.
Windows XP has been Legacy for 7 years. But Legacy stays around for a while, that's why it's called a "legacy". There are parts industrial tech that runs the world that has multiple layers of legacy going back 50 years.
POSReady which is basically XP SP3 with a couple of registry tweaks will be receiving critical patches until 2019 and I don't think that it can quite be considered 'Legacy' while it's still being technically supported and even today there's a fairly large user base all things considered. When even POSReady stops receiving critical updates, then I think it should be considered legacy like 2000 and NT. At the moment, it's just old and a bad choice for new systems; like Redhat 4.
Good point. Also, there is a difference between RHEL4 and Redhat 4. I know of a guy that patched a version of Redhat from the 90's to have a fix for ShellShock.