Submitted by The second issue of the ReactOS status reports was published today by Aleksey Bragin (ReactOS’s project coordinator). It’s a general overview of the last 6 months of project’s progress, with some additional information and future outlook. The document is available in .doc, .rtf, Word 2003 .xml, .txt, .odt, and .pdf.
I’ve been following ReactOS dev since the console stage. When it supports all windows drivers, I’m going to switch.
:-/ Even Windows has problems with some Windows drivers. Mostly through no fault of Microsoft’s. Each vendor writes their own Windows drivers. Some are gems, and work wonderfully; some are turds.
ReactOS is an intersting project to follow, though. I just wonder if their relevance keeps slipping, as Microsoft provides a moving target with new Windows releases and updates.
i believe that reactos is also a moving target, first going atter 95 code, then nt4, now i think it is win2003sp1. Even if it isn’t true, a windows clone that is free is good. I think people will then build drivers for it and maybe become as popular as linux based distro. Everything will always have problems, since it is man made, but they are cutting out the crud and making it simpler. I for the most part, use nlite to modify xp and then i switch to the classic theme. What i am seeing with this is a windows that is already cut down and working, with the classic theme and with the potential that other wm can be used because it is free.
ReactOS might seem like a moving target, but since their current target is still backwards compatible, it’s not really. Binary compatibility remains.
It’s just me who finds weird that you can read the document in every document format available EXCEPT…..html? Even .txt is better than html?
Edited 2006-07-27 18:21
Yes. It’s text. It’s nice to read.
html is … very far from nice (under the hood)
HTML version is now available:
http://svn.reactos.org/svn/reactos/trunk/press-media/Status%20R…
…we are on the WWW, they publish it in the accepted document format of hypertext?
OSS advocates won’t leave Linux or *BSD just because of this (well, I certainly won’t and I don’t know anyone who is evenm remotely considering it). Windows users generally don’t care whether something is open or closed source. So who needs this? I mean it is a lot of work …
Like cloning closed source UNIX?
it’s a shame that out of the four alloted soc slots, two projects didn’t even make it past the midterm evaluation, seems like such a waste of opportunity. great though that the Remote Desktop Client project is doing well.
as for Reactos relevance slipping, I don’t really see that happening. since if you are all fired up about Vista, then you aren’t really Reactos’s target audience.
Maybe I’m a bit retro, but I prefer win2k over XP. Don’t need the bells and whistles. rather a stable, fast, low resource operating system that is compatible with windows software and drivers. also, given it being open source, it will allow for greater opportunities in tweaking according to your needs. atleast that’s where I hope Reactos is heading. like a win32 equivalent of ArchLinux
“So who needs this? I mean it is a lot of work …”
it’s _free_, and open source. out of the millions and millions of windows users there are likely those who find this appealing. I sure do, then again I’m also a Linux and Beos user so in your world I guess I shouldn’t exist.
One question I have always had about Reactos and WINE is about special cases. I remember reading (whether they are myths or not I am not sure) that Microsoft implemented many special cases in their apis to ensure backwards compatibility with applications that rely on bugs in previous versions etc. Given that they are somewhat reverse engineering win32 etc, will they need to implement the same backwards compatibility special cases?
Personally, given that the NT architecture was supposed to support many faces (NT, POSIX, win32, win16, etc) on top of the kernel, I guess I would prefer if the special cases were moved out of the core win32 implementation and into a special face (can’t really remember the right term for this sorry). Is this possible to do? From reverse engineering can they know what are special cases and what aren’t? What impact would this have on the size and stability of the implementation?
> Personally, given that the NT architecture was supposed
> to support many faces (NT, POSIX, win32, win16, etc) on
> top of the kernel, I guess I would prefer if the
> special cases were moved out of the core win32
> implementation and into a special face (can’t really
> remember the right term for this sorry). Is this
> possible to do? From reverse engineering can they know
> what are special cases and what aren’t? What impact
> would this have on the size and stability of the
> implementation?
AFAIK, the ‘faces’ you speak about only apply to userspace programs. For them, your idea might very well be possible. The interesting stuff however happens between the kernel and the userspace, in the so-called executive. This is also where drivers run. AFAIK, for executive modules, no different views exist. All modules see the interface presented to them by the lower-level unit (kernel and HAL) and the interfaces from other executive modules.