New Open Source Windows-Compatible Operating System Released (github.com) 87
Red Hat product manager Pau Garcia Quiles (also long-time Slashdot reader paugq) spotted an interesting project on GitHub:
Free95, a new lean, Windows-compatible operating system is available from GitHub. In its current form, it can run very basic Win32 GUI and console applications, but its developer promises to keep working on it to reach DirectX and even game compatibility.
"Free95 is your friendly Windows Environment with an added trust of the open source community," according to its README file. (It's licensed under the GPL-3.0 license.) And in answer to the question "Why?" it responds "To remove Windows's bloat, and security problems. Being controlled by a large corporation is unsettling."
"It's still in-development of course," the developer post recently on Reddit, "and I'll appreciate anyone who'd like to contribute." In one comment they claim Free95 is "much more lightweight, simpler and faster than ReactOS." And looking to the future, they add "I might do DirectX stuff and make some games run. Or, what about DOOM?"
"Free95 is your friendly Windows Environment with an added trust of the open source community," according to its README file. (It's licensed under the GPL-3.0 license.) And in answer to the question "Why?" it responds "To remove Windows's bloat, and security problems. Being controlled by a large corporation is unsettling."
"It's still in-development of course," the developer post recently on Reddit, "and I'll appreciate anyone who'd like to contribute." In one comment they claim Free95 is "much more lightweight, simpler and faster than ReactOS." And looking to the future, they add "I might do DirectX stuff and make some games run. Or, what about DOOM?"
Free95 (Score:2, Funny)
Re: (Score:3)
In the end, it's always a catch-up game. Wine is doing well IMHO with regards to that. Anyway, interesting project nevertheless but I wouldn't bet on it to seriously replace the original.
Re: (Score:2)
I recall Microsoft Office and Internet Explorer being malware vectors more than Windows was alone. If this Windows compatible OS is used for games and non-Microsoft applications then I expect that the concerns of malware would be on par with most any other open source OS. I guess part of this depends on how closely the OS emulates the "real deal" Microsoft Windows OS. Perhaps I recall incorrectly but the malware got into Windows and spread because of features that were only peripherally related to making
Re:30 years going on 50 (Score:4, Insightful)
I think the problem with Win95 WRT to malware was that there was no user protections. Any program could do anything basically meaning anything you ran could completely compromise the system with no barriers.
Sure, Word would (at one point) run macros without asking putting Word documents at the same level as risk as executables (to a point), and other similar things.
But this was a completely 100% trusting environment with no user separation at its core. That was why malware thrived, no exploits needed to get root.
Re: (Score:2)
At the core, yes, you are correct. I recall doing all kinds of things with DOS including Terminate and Stay Resident programs. It was pretty cool but yes, you are correct, at the core any program could access everything and do how it seemed pleased on any resources. There were no concept of memory access control, no concept of running as root, no such like things essentially.
With regards to that, IIRC, the 386 was the first Intel chip to come out with a dedicated PIN allowing such things that were taken for
Re:30 years going on 50 (Score:4, Informative)
With regards to that, IIRC, the 386 was the first Intel chip to come out with a dedicated PIN allowing such things that were taken for granted on Unix system>
Intel 80286 had privilege levels, memory and hardware protection before the 386. There features were used by operating systems such as OS/2 1.x and Unix. I remember encountering an Unix System V Release 4 from AT&T for the 80286.
There was definitely an hardware limitation on Intel chip before the 386 to implement Unix on Intel chips
The main limitation of the 80286 was that memory was allocated as 64Kbyte segments and this made porting Unix software to it somewhat tricky. There was no 32-bit flat address space, although it could be emulated to a degree.
Re: (Score:2)
Hey! Thanks for that! I didn't recall the 286 thing but I recalled the 386 fully did what was needed. Thanks again, very informative!
Re: (Score:2)
I did a mistake, though. That Unix version I mentioned was actually for the 386. However there were multiple Unix versions for the 80286, including Xenix System III and System V, the SVR2-based Microport System V/AT (directly based on AT&T code) and others.
I'm not aware of any SVR3 or SVR4 Unix for the 80286, though. These versions seem to require at least a 386. Seems that at some moment AT&T decided to stop supporting the 80286.
Re: (Score:1)
I'm not aware of any SVR3 or SVR4 Unix for the 80286, though. These versions seem to require at least a 386. Seems that at some moment AT&T decided to stop supporting the 80286.
I kinda thought SCO Xenix 2.3.2 for 286 was SVR3, but it was a long, long time ago that I ran it on my 286-6MHz with 1MB RAM and 40MB RLL disk. I knew the lead developer (Dave J. - SCO had a lot of Daves, I also knew a David B and a David V there) and he hooked me up with floppies. I ran a UUCP node on it. Supposedly, some code from Xenix went back into SVR4.
Once the 386 came out and it became clear that the 486 would look much like it, there was no good reason to keep up with the weird-ass 286.
Re: (Score:2)
Whee my troll is back.
I guess they are very offended by my (*checks*) reminiscing about Xenix, something that is ideally suited to this thread.
It's pretty amazing how I can offend this cowardly toolbag by just existing. That's why I post about the bad moderation, clearly my posts offend it and I want to do more of that.
Re: (Score:2)
Nope. Xenix (all versions) is either Unix V6 (Xenix 1.x, unreleased to the public) or Unix V7 (Xenix 2.x) or Unix System III (Xenix 3.x) or Unix System V Release 2 (SCO Xenix System V 2.x and MS/IBM PC/AT Xenix 2.0). The next version from SCO was called SCO Unix System V 3.x and it was SVR3 - based indeed, but it required a 386 or later. The SVR2 - based Xenix supported the 80286 and even the PC/XT (8088).
I believe that yes, Xenix code was used in SVR3.2 (basically SVR3 on 386) and SVR4 (which merged all Un
Re: (Score:2)
The 286 even supported Flat Real-Mode! [wikipedia.org]
One HUGE problem with the 286 is that you couldn't return to Real-Mode after entering Protected-Mode unless you reset the chip. That was fixed with the 386 IIRC.
Re: 30 years going on 50 (Score:2)
You're right. I only wanted to note that contrary to what Wikipedia says, "flat real mode" usually refers to a variant of the Unreal mode where one segment is set up to cover the entire 32-bit address space to enable the so-called flat memory model, and this is possible only on a 386 CPU or higher. The 80286 CPU's segments' maximum size is 64Kbytes; a lone segment on a 80286 can't even cover its smaller address space of 16Mbytes. With that said, 80286 supports its own Unreal mode, which is just like the rea
Re: (Score:2)
I believe that Win95 had some very basic level of memory protection.
A virus written for DOS or a TSR could not work in Win95 of my memory serves.
A don't think an application crashing meant you had to automatically restart the whole system for example (as a contrast to Mac OS at the time which did require such for memory integrity reasons).
Re: (Score:2)
Agreed. The disappointing thing was that this wasn't "a different time" when perhaps things like exploits, or separation of permissions/concerns weren't really known about - the industry absolutely did know the broad brush means to thwart obvious and basic attacks as were prevalent on Windows95. Sadly, Microsoft didn't implement any of them, and so it became the malware fest we all know and love.
One wonders how much further along we could all have been if either we didn't fall for the marketing spin, or tha
Re: (Score:2)
They didn't really put in proper mitigations for the home user until windows 7 (maybe vista?) with the UAC.
Technically you could run XP secure, but so much required admin privileges (this is as much software companies at fault as MS) that everyone ran on an admin account.
With Windows 7 and UAC you finally had a pretty reasonable home user security model.
Obviously on the corporate side things were better and by late NT4 they had things pretty well working.
Emulating Win32 not Win95 (Score:3)
Win95 was a malware vector and that was was 30 years ago. Please no more.
They are emulating the Win32 API, introduced by WinNT and shared by Win9x. You are reading too much into an unfortunate and misleading naming.
Re: (Score:2)
You are reading too much into an unfortunate and misleading naming.
It made me think of the South Park episode Free Hat [wikipedia.org] where the kids advertised "Free Hat" (accidentally leaving off the trailing "s") to encourage people to come to their meeting to protest edits made to films prior to being re-released (like changing rifles to flashlights in E.T.) but people actually showed up thinking it was about freeing Hat McCullough from prison, who was convicted of murdering twenty-three infants. Of course, I'm not saying Windows 95 killed any infants ... but I'm not not saying tha
"More lightweight than ReactOS" (Score:2)
Yea wait until you get fucking DirectX going you'll be just as bloated.
Re:"More lightweight than ReactOS" (Score:4, Interesting)
This thing is basically Win NT 3.1 which in total used about 70MB or so of disk space (but around 100MB on MIPS
Re: (Score:2)
What I remember from NT 3.1 was that it was slow as hell, despite not being bloated (yet).
Re: (Score:1)
Why doesn't the dev use the leaked XP code to build a slimmed down XP? Or maybe he is?
Re: (Score:2)
By and large NT 3.5x outperformed NT 3.1 on identical hardware. They optimized 3.5x a bunch and many people consider 3.51 to be the first "good" version.
As for using leaked XP code, that'd be an absolute nightmare from a licensing and legal perspective. I would expect Microsoft is constantly going over any project that claims to implement Windows features with fine toothed combs to make sure nobody has used leaked code. They've already plonked a couple projects like the kind of hilariously named OpenNT base
Re: (Score:2)
By and large NT 3.5x outperformed NT 3.1 on identical hardware. They optimized 3.5x a bunch and many people consider 3.51 to be the first "good" version.
And also the last good version :)
Re: (Score:2)
NT 4 and 2000 were pretty decent as far as Windows operating systems goes. 3.51 suffered from having the same UI as Windows 3.1 (unless you installed the beta, only limited availability, shell upgrade that gave it a Windows 95 interface, but at that point it was more or less indistinguishable from NT 4 except for the screen device drivers being isolated in 3.51, and not in 4, the latter because of performance issues.)
XP is where it started to go wrong, bloat, ugly Fisher Price UI, "Activation"...
Re: (Score:2)
People give the 'Fisher Price' interface a bad rap - but it was the modern styling trend back them - reference the Apple 'fruit' iMac, the, toy looking controls in GM cars (among others), etc. It was the 90s, and that was hip.
Re: "More lightweight than ReactOS" (Score:2)
Mac OS X looked great. It had candy drops where XP had not just Fisher Price, but something that looked even dumber. It was and still is a kind of terrible interface, but the look has always been good.
NT 4 deliberately compromised security and reliability for graphics performance, and Windows has been fragile ever since.
Re: (Score:2)
One of the more bizarre anecdotes about NT4: there was a font that came with the computer that, if used and the wrong character displayed, would cause a buffer overflow in the display driver, causing an instant blue screen.
I know this because someone sent me an F***ing email with some of the text in that font.
And because there was no such thing as "Patch Tuesday" back then, you had to hunt for the update yourself to download a patch from Microsoft to fix it. They didn't publicize the issue or send update no
Re: (Score:2)
Yeah I had Belkin equipment from that era, I'm well aware of the trend ;-)
I was surprised it was popular with some because it really didn't look good, I noticed our IT team disabled the FP look on most XP computers they rolled out because it was the biggest single source of complaints.
(Also yeah I found Apple computers from that period ugly, I had a Blue and White G3 for a bit but kept it under the table out of sight.)
Compare with Mac OS X: the latter, yes, used color, but sparingly and tastefully. Those gu
Re: (Score:2)
You could install the new interface on NT3.51 using files from Win95. The real problem 3.51 had wasn't the interface (progman was bad, but we were used to it and could get work done) but the hardware limitations. It only supported 2GB partitions, for example.
XP wasn't really bloated, it barely used more memory than Win2k and you could turn off the stupid interface as well. The activation, OTOH, was bad... though not unique to Windows. You had to deal with similar shit with some Unixes, like Digital Unix, wh
Re: (Score:2)
3.5 was significantly faster than 3.1 (maybe that's why it was codenamed "Daytona"), but had problems with hardware (including CPU) compatibility. Windows NT 3.51 improved on that and was about as good as Windows NT 4.0 .
Re: (Score:2)
In 4.0 they improved performance by moving drivers into kernel space and thereby lost their security rating.
CPU's today are plenty fast enough for userland drivers but we get a Crowdstrike planetary attack instead.
Re: (Score:2)
LOL +1 funny, I rarely laugh that much nowadays reading Slashdot.
This will be great (Score:2)
Said no one ever. I want to install an entire OS just to run Widows apps individually without all the luxury of a full OS? WTF?
Re: (Score:2)
I think Microsoft did this sort of thing with it cross compiled Mac OS apps in the 1990s. There was a library that was basically the MacOS API.
Re: (Score:2)
"In its current form, it can run very basic Win32 GUI and console applications, but its developer promises to keep working on it"
It runs only a dos box and notepad, but the developer pinky promises to keep up the work! Which is typical US: deceptive marketing, lies and promises that will be broken any moment.
Re: (Score:2)
The dos box is actually a very powerful and complex application. It's encouraging if the new OS has it. OTOH, it only uses a relatively small set of Win32 API functions.
Re: (Score:2)
Re: This will be great (Score:2)
Re: This will be great (Score:2)
Re: (Score:2)
Said no one ever. I want to install an entire OS just to run Widows apps individually without all the luxury of a full OS? WTF?
That's the premise behind ReactOS, and some people are still interested in that for some reason. It can only be nostalgia, since Windows has never been the best OS, only the best OS that your average person could afford and was exposed to, and that there were applications for. PC GEOS beat Windows 3.1 like a pinata, but the apps were on Windows. Windows 95 is in turn best summed up by what happened at the time change, when all the Windows 95 PCs at our garage LAN party at the geek house I lived in rebooted
Re: (Score:2)
If people want their lives to be needlessly complex and difficult, why not just run Wine in a full and proper Linux install?
src/remarks.txt (Score:5, Funny)
Remarks
Kap Petrov:
I'd say this version's mouse polling system is buggy, it can fuck up from the keyboard, and because
of the polling nature, processes will not run unless mouse is being actively moved. In cavemen terms,
this version is fucking retarded.
Vincent392 (RE TO KAP):
Fuck, someone is gotta fix that.
Kap Petrov (RE TO VINCENT):
Surprise surprise, that's fucking me, because none of yall are fucking contributing shit.
Re: (Score:2)
You should dig out the Linus rants from the early Linux' mailing lists.
DCOM? No thanks! (Score:2)
DCOM is a ready-made vector for intrusion and fragile components. If you're going to build a Windows "clone" that's more secure, skip DCOM support!
Re: (Score:2)
DCOM? Is it the thingy that constantly fills up my event log with weird errors and makes my system act strange when I disable it? One of the MS' buggiest excrements, that's for sure.
Kind of like meatless meat (Score:2, Interesting)
Once you get those veggie patties to taste just like the real thing, they're just as bad for you as the real thing.
By the time you copy enough of Windows functionality to make it run Windows software, you've also copied all the vulnerabilities to malware and tracking and everything else you didn't want in Windows in the first place.
Re: (Score:3)
Isn't it possible to implement the Windows APIs free of Windows telemetry?
By the time you copy enough of Windows functionality to make it run Windows software, you've also copied all the vulnerabilities to malware and tracking and everything else you didn't want in Windows in the first place.
That's because much of the "Windows software" you mention is itself full of vulns and tracking and everything else, in the first place. Run a curated set of Windows software, and forget about MS Office, to have no problems.
If you intend to run MS Office, you may as well use true Windows. They're made for one another.
P.S. Early versions of Office are actually quite good. An Excel 4.0 running on this new OS would be a great thing.
Re: (Score:2)
Telemetry isn't the only issue. COM and DCOM, for example, were central to a lot of Win32 software. This technology turned out to be a major vector for malware, because all software on the system shared the same common components. So if some random software installed an updated version of mscomctl.ocx, everybody got the update, whether it was compatible with your particular software or not, and whether it was a legitimate version or not. There are many such components that contributed to the security diffic
Re: (Score:2)
The irony is that I remember reading the OLE/COM "bibles" from Microsoft, and they were very well aware of the version and DLL "hell", and they intended for COM to solve that problem, by versioning the COM interfaces and allowing providers of multiple versions of them to be installed simultaneously. Sadly, they never used that possibility fully (except maybe in DirectX) and their COM technology just created more DLL hell.
Re: (Score:2, Troll)
So WINE is full of vulnerabilities and tracking?
Your premise seems to be disproven by the fact that it's been done before without those issues. There is ReactOS as well.
Re: (Score:2)
You are missing the main point. Supporting Windows APIs isn't by itself an issue. I wouldn't be so worried about tracking or malware getting into Wine itself. But what COM/DCOM did, was force the sharing of third-party libraries. For example, if you used MSCOMCTL.OCX, and another app needed a newer version of it, the other app would install its own copy, forcing your app to use the new version as well. Tis is an attack vector because that third-party vendor could deploy a poisoned version of the OCX, infect
Doomed. (Score:5, Funny)
I'm sad to say that this project is doomed as an open source project.
The signs are obvious to anyone that know anything about open source projects:
* No offbeat acronym for a name
* No overly elaborate build system
* No absurd dependency list to compile it
* Not written in a niche language
* Doesn't depend on a specific compiler
* Aims to be compatible instead of insisting on incompatibilities to snub the original
The only thing they have going for them is having a terrible UI but I have no faith that they are going to keep it around as long as possible, continually adding more secret keyboard shortcuts that they expect every user to use.
At this point, I think the only way they can salvage this project is by rename it YAWAI (Yet Another Win32 API Implementation) which is both and acronym and potentially offensive and thus would ensure the survival of the project.
It's so tragic when you see open source projects that have such promise make such obvious mistakes.
Re: (Score:2)
April first isn't for another 3 weeks. (Score:2)
I mean it is astonishing that this prodigy has been able to re-implement in basically a weekend what it has taken the Wine project three decades to accomplish.
Thank you AI coding assistant (Score:2)
I mean it is astonishing that this prodigy has been able to re-implement in basically a weekend what it has taken the Wine project three decades to accomplish.
Thank you AI coding assistant. :-)
Re: April first isn't for another 3 weeks. (Score:2)
They are doing both more and less than Wine, so far mostly less. Wine isn't an OS, but it supports a lot more of the API than this project does so far.
Games (Score:2)
Re: (Score:2)
>"Unless it can play Windows games like a champ, it's dead in the water."
Actually, there are many use cases for a free/FOSS MS-Windows clone that don't involve running games. But the point is still there. Few people actually *want* to run MS-Windows, they just want to run some application that runs under MS-Windows. They might also want to have some familiarity with what they have used in the past.
It is unlikely something like will ever run most MS-Windows programs without issues. And even if it coul
Re: (Score:1)
My recollection with the early days of ReactOS was it was intended as much to enable using Windows drivers and underlying frameworks back when Linux and other options had much more limited hardware support, especially on the leading edge, as it was just for running Windows applications. Especially for stuff like graphics and wifi and DirectX where there was relatively limited support and being able to use vendor drivers and such would have been a big deal.
These days that's much less of a problem and stuff l
Re: (Score:2)
It is probable that zero of the use cases which don't involve gaming will ever be served by this toy, because in those cases you need at least as much reliability as is offered by Windows. It will always make more sense to run the old version of Windows needed in a VM for those purposes than to run them on this.
Re: (Score:2)
Solitaire, Minesweeper and Freecell are particularly mandatory. Maybe Hearts too.
This is some kind of joke (Score:3)
Right issues (Score:2)
These projects needs too much human work, but MAYBE that could be softened through AI in next future, both in terms of reverse engineering and programming an alternative.
But I think the biggest issue here is right issues.
The only reason Microsoft is not actively prosecuting these efforts is because the low level of adoption of these projects. They are too lacking to be useful in most scenarios so they are not a problem for MS.
But in case they advanced to a point they become a real alternative to Windows, I
Re:Right issues (Score:4)
They have lots of patents
Pretty much any patent that was in use when Windows Vista was first released has expired, so there's no risk in implementing anything from XP or ealier.
and it's expected lots of code coincidences, even if there isn't stolen code or too deep reverse engineering ends generating a decompiled code that's near a plagiary.
That's not how copyright works. In order to be copyright infringement, you have to copy the original work. That's why it uses the term "copy".
Re: (Score:2)
> Pretty much any patent that was in use when Windows Vista was first released has expired
Best not to assume that, for example MP3 was documented and released in 1991, but the last patents on it expired in 2017, 26 years later.
A major problem is that the clock starts running when the patent is approved, not when it's filed, so it's relatively common for patents to expire 25-30 years after the technology is in common use. Applying that logic to Vista, you could expect some of its technologies to remain pa
Re: (Score:2)
Best not to assume that, for example MP3 was documented and released in 1991, but the last patents on it expired in 2017, 26 years later.
Patent law changed in 1995. Old patents were grandfathered in. But in general your comment remains correct. Vista is not long enough ago.
Anachronistic. (Score:2)
I remember getting excited about Wine and other "let's run Windows stuff on FOSS" attempts. Like, for instance, the official Corel Draw 9 for Linux, which was quite impressive and even came with the Corel Font Manager, a thing which Linux was desperately lacking at the time. (I still have the CDs, licence and all btw.). Anyway that was more than 20 years ago(!) and the need for this sort of thing has loooong since past.
These days there are more than enough games for Linux and some special case software asid
Re: (Score:2)
These days there are more than enough games for Linux
Linux versions of games are often trash, and harder to get working than the Windows versions.
Remember those Loki games for Linux? None of them work now. IME, not even with Loki_Compat. But moreover, you want to run the Windows versions anyway. Even back then there were practical differences. For example, Alpha Centauri for Linux doesn't have automate nearby for formers, only general automate. It's got an old version of the former AI! And let's keep to the genre and compare Civ VI* for Linux to the Windows v
Re: (Score:2)
I personally fundamentally don't get how Windows still has such a market over macOS or Linux. It's that bizarre, even today. That's my impression anyway.
Bottom line: No FOSS Windows clone needed in 2025. Really not.
Assuming this is a sincere question...there are a good number of reasons for this.
For starters, it seems that there is an assumption that as long as a particular category of software is available, that it is interchangeable with industry standards. KDenLive is indeed great software, but exceedingly few third party plug-ins support it. I recently purchased Beauty Box [digitalanarchy.com] to help smooth out some overly hard face lighting; I don't think KDenLive supports this, or most other plug-ins that are used by many creatives
ReactOS (Score:3)
ReactOS attempts to accomplish the same thing as well, except it tries to go for full kernel compatibility so you can use drivers as well, because a big issue is device support.
Otherwise your best bet is doing Linux + Proton/WINE as thanks to Valve's involvement it's compatibility is rather remarkable.
GitHub (Score:2)
Re: GitHub (Score:2)
Why would they put an end to this?
Re: (Score:2)
Re: GitHub (Score:2)
Linux? Nope. But I get what you say. If Microsoft was Nintendo, then perhaps they would try to crush this project. But Microsoft isn't Nintendo, and they don't care because honestly, they have no reason to care.
Re: (Score:2)
Isn't it too early? (Score:1)
For April Fools day.
I wrote a full Vista FOSS clone, 100% compatible (Score:2)
Windows 7 open-source clone (Score:2)
Re: (Score:2)
Looonnnng Ways to Go (Score:2)
Open source projects need UI / UX folks. (Score:2)
Umm.. Did someone forget ReactOS??? (Score:2)
It's a given that there needs to be a non-MS work-alike OS, but why not ReactOS, its MUCH MUCH farther along and for various cases of "working", it actually does work and many windows programs work fine on it.
https://reactos.org/ [reactos.org]