Now this is what I like to see. Smaller, alternative operating systems have been dying by the dozens these past 6 years or so, so in order for the remainder to survive, they need to work together. ReactOS and Haiku have been doing that for a while now, and the latest fruits of this collaboration is a much-improved USB stack for ReactOS.
Earlier collaboration between the two projects focussed on working together for conference attendances, both at FOSDEM and SCALE. As a successful Google Summer of Code participant, Haiku also aided ReactOS in applying for GSoC – ReactOS’ application was accepted as a result.
This time, though, the collaboration focusses on cold and hard code. The ReactOS USB stack needed to be improved, and by using the Haiku USB stack as a reference, Johannes Anderwald made significant strides in completing the work.
“Many of the definitions and data structures that represent USB protocols were borrowed directly from Haiku, though the differing operating system design necessitated a great deal of glue to be written to make use of the code,” the ReactOS team details, “Johannes also referenced Haiku’s USB stack to better understand the behavior of USB devices and subtleties that might not be entirely clear in the USB specifications, and also provided feedback to Haiku as he worked through the code.”
The end result is positive for all – not only did ReactOS get a more capable USB stack, I’m sure the feedback given to the Haiku project made the Haiku USB stack better as well. In an industry reality where geeks seem to care more about logo perspectives than low-level coding, this kind of collaboration is absolutely vital.
Great news these are two projects that I really want to proceed. I think just about everyone who ever used BeOS has a soft spot for it and to see it reborn would be great.
ReactOS might make a huge difference the ability to run Windows applications without Windows and Opensource, would be fantastic.
Edited 2012-02-21 15:49 UTC
If only smaller OSes could collaborate more we would have even more high quality alternatives.
Bravo and I am expecting to see other collaborations like AROS/osFree/Puredarwin/Plan9 (and derivatives)/Minix/Genode/Hurd/Syllable/ReactOs/Haiku.
We need more collaboration, more alternatives in order to finish with proprietary solutions.
Edited 2012-02-21 16:06 UTC
I’m not sure what they ended up doing in the end, but I know that AROS was curious about Haiku’s work on the WLAN (Wifi) stuff. I remember speaking with someone from AROS last year at SCALE – he was interested in Haiku’s stack. I’m not sure what strategy they ultimately went with, however.
Since Haiku copied most of the WLAN stack code (as well as other netcard drivers) from FreeBSD, perhaps that’s where they ended up going as well.
This is how FOSS is supposed to work
That’s a great idea in theory, but licenses get in the way.
This strategy makes sense though because no one is an expert on everything.
Windows is dead in the water, ReactOS will come pre-installed on all OEM machines before the year is out.
You are kidding, right? *
– Gilboa
* And trust me, I dislike Windows 7 and consider Windows 8 metro interface to be a bad joke, but ReactOS is -years- from being competitive with Windows XP, let alone Windows 7/8.
I have a feeling that the OP is joking.
Now regarding XP, I would be seriously OK with an XP replacement because I have NO issues with the XP interface especially compared with Windows 7.
There is always Classic Theme … if you must use it.
I dunno why you wouldn’t want to use Windows 7 over XP now, better dual core support, better performance with prefetch and your graphics card accelerates your graphics.
The Dock Style Windows take a few minutes to get used to an no different really than windows grouping under XP.
It took me less than ten minutes to get one with the 7 user interface … I am sorry … I am running XP 64bit and I find it limiting.
Edited 2012-02-21 20:31 UTC
It would be nice if windows 7’s classic theme supported 2d acceleration like windows xp. While aero is nice to look at for a little while, it’s clunky and big compared to the classic theme which is elegant and doesn’t waste of lot of monitor real-estate in comparison.
— edit —
Yes, I know the classic themes performance issues are due to the changes windows 7’s graphics stack. Just saying that the classic theme on windows 7 is less than ideal due to the lack of 2d acceleration.
Edited 2012-02-22 00:40 UTC
If it had square borders and wasn’t quite as “big”, it would be perfect. I agree.
Blame GPU makers, they didn’t do any 2D GPU accelerating development, all the money was in Games,
My Mate uses a Matrox Triple Head job with Windows XP and it is soo damn good when using it with AutoCAD.
I switched to Linux because I didn’t like what MS has done to Windows 2K excellent base when they released XP.
While I did get used to it (but XP was relegated to development VM) – I do agree that Windows 7 interface (especially the annoying I-know-it-all-network-configuration) is, in my view, horrible.
… Though, I would imagine that the vast majority of the Windows userbase will disagree with me strongly.
– Gilboa
Lol WAT? Win2K was terrible on release and didn’t get good until Service Pack 3 or 4.
In everyway except for the shitty Luna theme (which takes about two minutes to turn off) is an improvement.
Sure, but Windows 2K/SP3 was stable, fast and didn’t get in the way.
XP, at least in pre-SP, was slow (relatively speaking), unstable (relatively) and got in my way.
… Well, you get the idea.
– Gilboa
Most of that was the alpha transparency and clear type that slowed it down.
I neglected to say that I was a part of the XP beta via MSDN and didn’t really see the benefit of switching from 2K to XP.
In many ways, I saw the beginning of the Vista disaster in the 2K to XP switch.
– Gilboa
There was nothing wrong with VISTA tbh. People complain about it but all the issues with it (which weren’t many) were fixed around about Service Pack 1 time … VISTA Service pack 2 is pretty much Windows 7 underneath.
If you think Windows 2k was the pinnacle of Microsoft’s Tech you are quite clearly deluded.
Guess we’ll have to agree not to agree …
– Gilboa
If you go back far enough, plenty of people were whining about win2k (virtually the same way you do with XP), while at the same time praising what came before 2k (which too often meant not the earlier NT versions, but – shudder – 98SE …that’s what such ~nostalgia does, you get the idea)
That’s a dumb reason to switch, XP can made nearly-identical to 2k by changing 4 maybe 5 settings… unless you’re just one of those people who thinks that preferring 2k to XP makes you more “l33t”.
Actually, XP was quite a bit more bloated than W2k… unless you ended up running one of the “tiny” versions that came out during the last decade.
I preferred w2k as well, and only jumped ship because of the lack of driver support and the desire for built-in RDP (which only w2k server ever had prior to XP).
Troll much?
– Gilboa
ReactOS may offer, once it reaches 1.0 level, a viable alternative to Windows 8 for Win32 legacy applications. However, this may also be only for older hardware unless there is a reversal in the boot-lock BIOS code being considered as part of the Windows 8 deployment.
Anyways, this is what I hope for.
I admit I shed a little tear. This is so adorable.
I’m stucked right now just trying to develop an USB stack/drivers for SGI Irix. And that’s one of the most difficult tasks in software programming.
The USB protocol/stacks/controllers are unnecesary difficult.
Oh wow. For any good reason, or just for (some value of) fun?
just for fun, I would like to have mi VIA VT6212 PCI card running in Irix.
Hi!
USB support is really difficult to implement, leave alone all the different standards flavors, if you have any questions, just join our irc channel #reactos-dev, my nick is janderwald
regards
Edited 2012-02-21 23:27 UTC
Thanks!! I will do, I really need help with this
Could this be by design?
Only a commercial OS, or a open source OS project with highly dedicated and skilled contributors, could through the necessary resources to generate efficient code for such hard thing to do?
And I presume that USB 3.0 adds an extra level of complexity over the well established USB 2.0!?
Collaborating on these difficult components is a good thing for the open source OS projects. This leaves more resources for the stuff that really differentiate these projects – the user experience.
I’m curious to know of the dozens of alternative OSes that have died in the past few years… I can only think of SkyOS.
SkyOS wasn’t open-source though. It’s next-to-impossible to kill a FOSS operating system. Look at AtheOS, that got abandoned, but got forked by the (albeit floundering) Syllable OS
Although, Cosmoe was also forked from AtheOS and is basically dead now…
But yes, it’s hard to kill FOSS, it tends to just fizzle into obscurity, but the sources are likely still out there somewhere!
But Cosmoe was a radical project. The developer (Bill was it?) was trying to get Atheos running on a Linux kernel to begin with… Then he decided to converge with BeOS. It was never very linear nor was it very organised.
Funny, that’s where Syllable eventually landed
Well, yes – and IIRC, he did even commit some code to Haiku.
And this is probably the leading cause of alt-os death… which isn’t necessarily a bad thing.
Sometimes people just want to scratch an itch and try something new – if it works out, great! if it doesn’t, oh well, at least they (and others following the progress) may have learned something, right?
Edited 2012-02-21 23:28 UTC
Not quite. I did have a quick hack up of the appserver and AEdit running on Linux under SDL, and then subsequently produced https://bitbucket.org/Vanders/libs/overview, but the plan was never to move from the Syllable OS as a whole to Linux: the Linux stack was going to be a complement to Syllable Desktop.
Killing it by abandonment is still a kill, I suppose…
I think one can find plenty of small abortive attempts, FOSS licensed, scattered throughout the web. And, lets be honest, no one will ever really pick up vast majority of them (of course, there’s probably not much to pick up in many or most cases, not much code – so OTOH, should we count that which was never alive?)
I remember that OS. SkyOS supporters proudly embraced the fact that it was closed source as a sign of strength and they didn’t learn from history. oh well
And then all of the open source fanboys threw fits over the fact that SkyOS wasn’t open source, and accused SkyOS’ author of “stealing” from Linux… based on nothing other than the arrogant assumption that it was the only possible way he could accomplished what he did.
Ah, the good old days.
I give these guys alot of credit for keeping at it so long, but has been in the works since 1998 and hasn’t made it out of Alpha yet. I would love to see something 100% WinXP compatible for older hardware but by the time this project gets to that point nobody will remember how to use XP. I am not bashing these guys. I just think they took on a project bigger than they could really handle.
By the time reactos actually makes it to 1.0 (about the same time windows 28 gets released), microsoft will probably be giving away the code for to windows 2000/XP etc. with copies of windows 28 in some kind of virtual machine for nostalgic reasons.
I’m sure that when using windows 28, which will no doubt come as some kind of optical implant/HUD system for your eyes, people will be still be hoping for an open source version of windows 2000…
Edited 2012-02-21 18:59 UTC
No it won’t … what f–king fantasy land are you living in?
Windows Apps for business are carefully checked for compatibility with new Windows. This includes undocumented features. It takes years (the company until recently I worked for has only just started their Windows 7 migration).
I love these “All businesses will move if this … ” … without taking in that their IT employees have to move 1000s of users over very very slowly.
If anything is going to be run that is legacy to “Windows 28” it will be run in a VM by Quantum Computers that is managed by skynet most likely 😉
Edited 2012-02-21 20:42 UTC
Has anyone here ever used ReactOS in anger, I still have one machine stuck on W2K doing it’s thing.
In anger? I can’t say i have, but i wiped a PII box and installed ReactOS and BeOS R5 PE to use as a backup PC
XP UI with most of the effects turned off is Win2k … might as well use something that works and is well supported rather than this project that is chasing a target that is accelerating faster then them.
As someone else mentioned above, this is no small feat.
Writing a USB stack is *really* hard. I helped a colleague who was working on a client (not host!) USB stack for the MSP430. Even with sample code and some support from TI, it took him almost a month to implement a subset of the ACM functions that we needed. I don’t even want to think about what writing the USB host stack involves.
I think host support is easier.
Since in USB all communication is initiated by the host, the driver author gets to decide what to do and when.
For clients, you’ll have to cope with whatever the host throws at you. (but to be fair: some devices are rather weird in what they answer)
But I only wrote a bunch of host drivers (ohci, uhci, ehci, some xhci stub, various class drivers) so far, and no client drivers.
In all fairness, coping with weird input was indeed difficult, but we still only had to support a single class and we could cut some portability corners due to the low resources (8K of RAM and 48K of flash is enough but isn’t *that* much).
Although it is certainly taking a long time to bake ReactOS, there is an impressive amount of development going on. You can track the progress online:
http://cia.vc/stats/project/ReactOS
As you can see, the USB stack is still be heavily worked on.
@Janderwald – Thank you for your work on ReactOS. Some day it will be very important.
ReactOS has changed what version of Windows they are targeting several times. If I am not mistaken, they currently target Windows 2003.
It is true that it will be years before Microsoft needs to worry about competition from ReactOS. That said, it may not be that much longer until it is useful for a number of people. USB and wireless networking were probably the biggest hurdles for use on real hardware and they are both basically there now.
Windows is a moving target but the core architecture does not change as much as people think. Windows 7 is not really all that different at heart than Windows NT4. The user interface is a red herring for now. What matters is the API that the OS exposes to applications.
Implementing the guts of an OS as complex as Windows is a huge task. Once they get most of it done, people may be surprised at how rapidly they can start to add the bells and whistles.