Hey Richard!
Nice to see your reply here, and I didn't know that we are nearly same age (I'm turning 31 in approximately twenty four days).

What you are saying actually makes sense. And that's exactly why I did it this way. Present win32 subsystem is a mess consisting from own good code (mainly in win32k - GRE, and some other parts written by Timo/James/Giannis/etc), our own bad code (historical parts derived from ancient Wine, which was/is full of bugs), and just old Wine code which gets adapted into user32/gdi32 pretending to be "proper".

We don't have manpower to even maintain that (as updating that old Wine code involves quite a bit of work, to re-port new Wine code, to backport fixes, etc), and some parts are just unmaintainable (like ported ancient Wine code which just needs to be rewritten). Not to say about doing a proper, full blown win32 subsystem (which is even hard to define because win32ss in Windows is far from being as beautiful as NT kernel is).

Hence I decided to do exactly what you say. I took a Wine version which works, ported it preserving as much arch as possible (thankfully Wine code matured to be good enough for that) and created a new subsystem which is VERY simple to maintain (updating to new Wine versions is a usually few minutes of work and few hours of testing :-) ), and where we utilize rather big manpower behind Wine. And where we don't have hassle with Linux being the underlying layer, we just do it all natively, like Wine's dream.

Understanding all this, and actually seeing how good Arwinss performs already, it would be great to get more people to work on it, as the result of such work is substantially more noticeable than just a few fixes here and there in the present win32ss.

If Reactos continues to be in such unusable state, then it's going to remain in the history as "an os which never took off due to its complexity". I don't want that to happen. Too many people invested their efforts in it to fail so easily.

So let's think, it's just about time.

Regards,
Aleksey Bragin


On 31.12.2013 11:03, Richard Campbell wrote:
I'll go out on a limb here and agree with James Tabor to some extent.  I've been saying this for quite a while.  While ARWINSS is technically interesting, i feel that with limited developer choices, ReactOS should be focused on developing a usable, stable platform.  Focus on any target!  Windows XP, Windows 7 (7 is typically a good target as app support won't be going away for a while, however Windows XP is good as well), etc. and make it work..and look... great.  Make it usable for the every day user, even if it's still alpha quality.  You do that, you'll not only get developer backing, but have the satisfaction of having been a part of one of the biggest achievements an open source project has ever accomplished.

Note that the statement I have previously mentioned doesn't technically EXCLUDE ARWINSS...but rather I'm saying to zero in on a specific version of Wine, fork it, optimize it for ReactOS, plug it in, and focus on the bigger problems...like the fact that AMD Catalyst drivers won't install/run.  

As some of you well know i've been following (and in some instances in the long past, contributing to) ReactOS.  I just want to see it work.  I want to see it succeed sometime in my lifetime.  The journey has been incredible, but i'd like to see it really take off.  The only way it's going to get there is to draw in user interest.  Do any of you know exactly how long this project has been under development?  I do...i've been with it...and following it...for a VERY long time.  Hell, when i first started following this project, i was on dialup and a 386!  I was a teenage for crying out loud!  Now i am am 31 years old, I have a wife and two daughters and this project still continues forward.  Incredible stuff!

If ReactOS were to get to that 'usable' phase, it wouldn't be that hard to get REAL funding to tackle some of the hard stuff...like staying up to date with the latest versions of Windows.

Thanks for reading my rant!
Richard Campbell


On Mon, Dec 30, 2013 at 8:21 PM, BinSys <binsys@163.com> wrote:

I have always insisted that ReactOS should be fully compatible with WINDOWS and work towards, rather than the introduction of other things. ReactOS This is my concern and willingness to contribute rather than the sole reason of other open-source items like WINE.
Even I think that the introduction of WINE is not a right decision. (By Google Translate)

Regards,
BinSys
At 2013-12-31 01:05:07,"James Tabor" <jimtabor.rosdev@gmail.com> wrote:
>Things I like about arwinss...
>
>I call SWM winman.c window manager, SWiM, due to it reminds me of
>swimming through a list of windows. Studding this code for months,
>leaving no stone unturned.
>
>On 12/30/13, Aleksey Bragin <aleksey@reactos.org> wrote:
>> Thanks, I thought I would end up talking to myself :-)
>>
>> Well, I totally eliminated X11 out of Wine. It was not that hard really,
>> nearly all good code was in place already, the only thing which I had to
>> develop was a very small "window manager". Literally, which would manage
>> list of windows in the current window station and desktop, and issue
>> redraw messages when certain window gets focus.
>>
>> "A little flame never killed nobody"
>>
>> Regards,
>> Aleksey