The latest ReactOS newsletter has been published.
Timo Kreuzer (tkreuzer) worked hard on various parts of the kernel and HAL, fixing issues here and there. Structured Exception Handling (SEH) support for the amd64 architecture was finished, various bugs around the kernel are fixed. A major issue with interrupt handling in HAL was also fixed in May, which finally allowed a semi-stable boot in a virtual environment.
There’s also work being done on support for multiple monitors, improved support for SMP, and more.
Sometime in the 2050s, a year after we all have direct communication implants connected to our brains (thus rendering OSes obsolete), GNU/Hurd and ReactOS will announce their first functional release. I love these two projects.
I think when Windows eventually becomes obsolete or becomes something completely different and incompatible, ReactOS will have time to catch up. Maybe even Microsoft would help, like with Mono and .NET. Then, ReactOS would serve the back compat role, similar to, like, DosBox or something. But I have to wonder if something like Wine would be a better choice for that rather than a full OS.
I don’t know, maybe ReactOS would be the upgrade path for old Windows ATMs…
Nah, corporations need “bug for bug compatibility”. They will just keep using Windows until the machines clap out. Microsoft EOLed Windows Mobile (aka Windows Embedded Handheld), but corporations still use that OS until the handhelds clap out.
Phones are throwaway devices… once the OS is gone from being actively developed it can effectively disappear within months. And in all seriousness… why are you acting like phones are going to last 5-20 years like a desktop PC.
Windows Mobile/Embedded was used in all sorts of devices, and that is most likely what kurkosdr is referring to.
Windows based phones and PDAs are gone, but not things like handheld scanners.
Maybe it’s time to rekindle the idea of turning Win32 into an ISO standard? That gives everyone something to aim for. I think there’s really good reasons for it too. When an OS becomes the default standard it gives a single unelected and unaccountable company a lot of power and the power to print money. A standard could be used to bring Microsoft to heel and eventually force open source reference implementations in the long run.
https://xkcd.com/927/
We already have POSIX, which is reasonably well supported by most OSes. And if you need GUI, X11 is also available on nearly every OS too
What’s that got to do with Win32? Nothing as far as I can tell. No amount of huffing and puffing will magically allow Win32 apps legacy or new to run by shouting Posix and X11 at them. Nor will any xkcd links I never click on that anyone posts.
Pretty sure you just compared a mouse to an elephant. Win32 is an entire SDK, while POSIX/X11 is tiny. Also the comic you love so much only applies to open source software. Closed source software either sells, or dies. That’s why comparing open source to evolution is a false equivalence. Evolution is a two step process where natural selection removes the unsuccessful mutations, and open source software doesn’t have to deal with that second step.
ISO will not go ahead with approving a standard unless a “FRAND” declaration has been received by all entities with a reasonable claim on owning patents essential to the standard. This is what doomed any previous attempts to standardise the Windows API
https://en.wikipedia.org/wiki/Application_Programming_Interface_for_Windows#ISO_delays_the_standard
Microsoft will never allow the Windows API to become an ISO standard because then they would be bound to the standard, which means they could not add to it at will, like they do now (for example DirectCompute and DirectStorage).
It’s the same reason Microsoft abandoned RTF in favor of the binary Office formats, and why they refused to use ODF and instead pushed ahead with their own OOXML format, which they are free to make a moving target as the sole maintainer of the standard.
Really, the best we can hope for is Wine to reach a “good enough” level where most apps targeting Windows 7 run well and only things like games using the latest DirectX APIs are not working.
@kurkosdr
ISO dropped the process of making Win32 a standard for political reasons. There’s no legal or other reason I can see why ISO cannot move forward. Someone needs to ask ISO some searching questions and Microsoft too not just take their word for it.
“Microsoft claimed intellectual property over Windows APIs”. And that “intellectual property is? Copytright or a patent? Microsoft have lately switched sides and said copyright claims over API’s has no legal merit and holds back the industry. And exactly what aspect of Win32 is covered by patents ongoing or otherwise?
If Win32 “as is” becomes a standard Microsoft can still make whatever they want they just cannot claim it is ISO whatever compliant. That’s it. However what an ISO standard will do is allow every decision maker to mandate any OS they purchase must be ISO whatever compliant so they can run legacy applications etcetera. That is what scares Microsoft. It gives decision makers an easy specification demand and a fixed target anyone can work to.
As for DirectX API’s Microsoft still have questions to answer about market abuse. They absolutely did abuse their position and in some cases key members of staff openly bragged about it. The head of their DirectX (and initial X-Box development) was known for it and you can find their relevant comments quoted in interviews with the gaming press. Also on Windows OpenGL and Vulkan have the ICD mechanism to provide support. Microsoft utterly hated this but had to have it otherwise they would lose the workstation and professional market. Whether developers support Direct 3D or OpenGLVulkan is a matter of choice as is portability and scaleability. Developers who don’t abstract graphics and other API and who don’t build in portability and scaleability really don’t deserve being called developers. This work is trivial and there is no excuse other than ignorance and laziness for not doing it.
So yes I think it’s worth starting a discussion about making Win32 an ISO standard and also bringing developers on board with portability (and scaleability). You don’t need to supply entire frameworks to achieve portability especially for games. They almost all use a tiny subset of the OS anyway. Most of the critical none portable code is only because they chased a graphics API and features without thinking through the consequences.
Microsoft claims patents over the win32 API, just like they claimed patents over VFAT (and guess what, some of these patents were found to be valid by patent courts, because it should be obvious by now that software patents are a patent lawyer’s delight). ISO will never go ahead with ratifying a standard unless they receive at least a “FRAND” declaration from any company with a plausible claim.
It’s the same reason any attempts to create a royalty-free “baseline” of H.264 failed (and MPEG had to start over with EVC). Some companies asserted patents over it and only offered a “FRAND” declaration.
https://blog.chiariglione.org/a-future-without-mpeg/
Quoting: “ISO rules allow a company to prevent a no-licence/free licence standard from happening by simply making a cautionary “I may have patents that I am willing to licence” declaration”.
Microsoft is doing something similar for win32, only that they are not offering even a “FRAND” declaration, which makes any standardisation impossible.
My point is, don’t go to ISO unless:
1) Some government forces a company to make a standard (which is highly unlikely for OS APIs, I mean, governments dragged their feet to push an ISO standard for documents, despite the threat to national archives, and even then OOXML was the result)
2) The companies involved (aka the companies that have a plausible claim) are willing to play ball together. And even then you will probably get a “FRAND” declaration
This is what Google realised, and they don’t even deal with ISO. Instead, they release their own specification (WebM) and invite any company claiming to have patents on WebM to sue YouTube (the primary user of WebM).
So, the take-home message is: Don’t go to ISO for win32 standardisation. Governments most likely won’t push for it, Microsoft doesn’t want to play ball, and the standard (if by some miracle happens) probably won’t correspond to reality anyway. Wine’s documentation effort is the best we can hope for.