Proposals for REALLY pushing this project forward
Moderator: Moderator Team
Re: Proposals for REALLY pushing this project forward
Probably around $8-9K on the low end.
-
- Posts: 183
- Joined: Sun Sep 03, 2006 11:39 am
- Location: Paris, France
- Contact:
Re: Proposals for REALLY pushing this project forward
I'm not sure where you got such information, but our current Cc allows ext2 and ext3 working smoothly in ReactOS. The fact that ReactOS cannot be installed on it is just a matter of bootloader.dark wrote:Out of curiosity, how much would a contract cost for getting the new CC done (and installing to ext2 works)? That seems like it would be the main thing to make the system run far more stably.
I'd recommend you try ReactOS before writing such statements.
Pierre Schweitzer (aka Heis Spiter)
ReactOS kernel developer.
ReactOS systems administrator.
ReactOS kernel developer.
ReactOS systems administrator.
Re: Proposals for REALLY pushing this project forward
Does ReactOS record ACLs(for security attributes) in an ext2/3/4 FS?
-uses Ubuntu+GNOME 3 GNU/Linux
-likes Free (as in freedom) and Open Source Detergents
-favors open source of Windows 10 under GPL2
-likes Free (as in freedom) and Open Source Detergents
-favors open source of Windows 10 under GPL2
Re: Proposals for REALLY pushing this project forward
http://imgur.com/a/p7LscHeis Spiter wrote:I'm not sure where you got such information, but our current Cc allows ext2 and ext3 working smoothly in ReactOS. The fact that ReactOS cannot be installed on it is just a matter of bootloader.dark wrote:Out of curiosity, how much would a contract cost for getting the new CC done (and installing to ext2 works)? That seems like it would be the main thing to make the system run far more stably.
I'd recommend you try ReactOS before writing such statements.
Still doesn't work during install, 0.4.2 and r72404 (ext2)
Re: Proposals for REALLY pushing this project forward
Konata wrote:But no, the project doesn't need a bunch of loose cannons who wouldn't know what to do with themselves when they find they can't aimlessly mangle the project any way they want like they do with linux and it's API/ABI
What about people who develop typical desktop distributions like Ubuntu, Kubuntu, Xubuntu... ?Z98 wrote:The majority of Linux kernel developers are not working on Linux for the sake of creating a desktop competitor to Windows. Their primary motivation is to basically craft an operating system that suits their needs, or in the case of the paid developers, the needs of their employers. They really don't give a damn about how suitable Linux is as a platform to build a desktop OS that is usable by the general public.
Imagine two programmers. Both are knowledgeable and experienced C/C++ developers, but both of them are unfamiliar with Windows NT technology. However, both are eager to learn about it. Only difference is that second guy worked more or less on development of some other OS, not necessarily Linux.
Which of them would be more useful to ReactOS? Which of them would more easily grasp Windows NT?
My idea is to create short promotional movie, maybe no more than 1:30 long, where we would ask much needed programmers to join effort. Specifically, we should target other OS developers because there have to be some common things to all OS programming.
Which software would you recommend in order to create such movie? Blender?
-
- Posts: 183
- Joined: Sun Sep 03, 2006 11:39 am
- Location: Paris, France
- Contact:
Re: Proposals for REALLY pushing this project forward
I'm not sure how you get to that point. As I said, setup doesn't offer the option to install on ext2/3 due to the lack of support.dark wrote:http://imgur.com/a/p7LscHeis Spiter wrote:I'm not sure where you got such information, but our current Cc allows ext2 and ext3 working smoothly in ReactOS. The fact that ReactOS cannot be installed on it is just a matter of bootloader.dark wrote:Out of curiosity, how much would a contract cost for getting the new CC done (and installing to ext2 works)? That seems like it would be the main thing to make the system run far more stably.
I'd recommend you try ReactOS before writing such statements.
Still doesn't work during install, 0.4.2 and r72404 (ext2)
I repeat myself, out of that specific point, ext2 and ext3 are working smoothly on ReactOS. You should really try them out of setup and enjoy.
Pierre Schweitzer (aka Heis Spiter)
ReactOS kernel developer.
ReactOS systems administrator.
ReactOS kernel developer.
ReactOS systems administrator.
Re: Proposals for REALLY pushing this project forward
That is quite unlikely, because Unix ACLs and Windows ACLs are as incompatible as Unix Access Rights and Windows Access Rights.erkinalp wrote:Does ReactOS record ACLs(for security attributes) in an ext2/3/4 FS?
But you are free to implement some mapping code...
-
- Posts: 282
- Joined: Thu Jan 04, 2007 12:43 am
- Location: Italy
Re: Proposals for REALLY pushing this project forward
"Proposals for REALLY pushing this project forward?"
I have two proposals:
1) trash the limit of 2K3/XP API compatibility and support also newer functions. The look and feel could stay on the classic user interface, but since I use XP every day I can tell you that lot of software refuse to work and the amount of those programs will always increase. If you really want, you can limit the available functions at compile time and leave this choice to the final user. This will also simplify the alignment with the components shared with the WINE project. I do not see it as a so big trouble... after all, this project started initially for cloning Windows 95, no?
2) Do something against ONLINE-614: no offense, but really, if there won't be a solution for it, you won't go anywhere.
Sincerely.
I have two proposals:
1) trash the limit of 2K3/XP API compatibility and support also newer functions. The look and feel could stay on the classic user interface, but since I use XP every day I can tell you that lot of software refuse to work and the amount of those programs will always increase. If you really want, you can limit the available functions at compile time and leave this choice to the final user. This will also simplify the alignment with the components shared with the WINE project. I do not see it as a so big trouble... after all, this project started initially for cloning Windows 95, no?
2) Do something against ONLINE-614: no offense, but really, if there won't be a solution for it, you won't go anywhere.
Sincerely.
-
- Posts: 1790
- Joined: Fri Aug 07, 2009 5:11 am
- Location: USA
Re: Proposals for REALLY pushing this project forward
On compatibility, I see the point. We do have some Vista/7 functions since they are needed to get some of the Wine code to work. However, the thinking is, AFAIK, that we should wait until we have a solid base underneath before we go changing compatibility targets. Some of the necessary APIs are there, but others are still necessary before we do switch targets.
As for Free95, that went nowhere in terms of code, so it was no trouble to switch to NT compatibility, since no coding had begun.
As for Free95, that went nowhere in terms of code, so it was no trouble to switch to NT compatibility, since no coding had begun.
Re: Proposals for REALLY pushing this project forward
This would actively stagnate the project. I highly doubt there's anyone who looked at the project and said "hm, I'd love to share my intricate knowledge of NT with this project, but it only targets Server 2003! No deal." The developers need to focus on 2003 right now just to get a stable system. After ReactOS is "done enough" then they can progress to supporting newer versions. ReactOS' kernel already has some features from newer versions of Windows, they're prepared to support later versions, but not now. Targeting Server 2003 is a strict discipline and they can't afford to chase a moving target right now.Carlo Bramix wrote:trash the limit of 2K3/XP API compatibility and support also newer functions. The look and feel could stay on the classic user interface, but since I use XP every day I can tell you that lot of software refuse to work and the amount of those programs will always increase.
I asked the developers about this, they said doing this would be incredibly complex and nearly impossible. So targeting 2003 and 2003 alone is their only option right now.Carlo Bramix wrote:If you really want, you can limit the available functions at compile time and leave this choice to the final user.
-
- Posts: 282
- Joined: Thu Jan 04, 2007 12:43 am
- Location: Italy
Re: Proposals for REALLY pushing this project forward
Well, if there is a philosophical reason, like to say that "we took a way, we must continue until we won't reach the goal", then ok, there is nothing to do...Konata wrote:This would actively stagnate the project. I highly doubt there's anyone who looked at the project and said "hm, I'd love to share my intricate knowledge of NT with this project, but it only targets Server 2003! No deal." The developers need to focus on 2003 right now just to get a stable system. After ReactOS is "done enough" then they can progress to supporting newer versions. ReactOS' kernel already has some features from newer versions of Windows, they're prepared to support later versions, but not now. Targeting Server 2003 is a strict discipline and they can't afford to chase a moving target right now.
However, I still think that this fossilization is not the right way.
I started to contribute to development of ReactOS core in 2007 if I'm not mistaken, nowadays we are in 2016 and, although it is progressing every day, ReactOS is not near to the features of Windows 98. Please do not take what I have written as an offense, but we should be quite careful when we say "done enough"...
I think that it is not true.Konata wrote:I asked the developers about this, they said doing this would be incredibly complex and nearly impossible. So targeting 2003 and 2003 alone is their only option right now.
ReactOS relies on SPEC files https://www.winehq.org/docs/winelib-guide/spec-file
These files have been designed to accept also some options.
Nothing blocks us to add an option "-version" to be applied only to some exports, for making them available only if they have the version equal or not greater than supported platform. No changes in the source code would be done. In reality, it would be better that some functions are also excluded with some conditional compilation, but hopefully GCC could do this for us automatically if we want, by eliminating dead code during linking.
It's easy.
-
- Posts: 10
- Joined: Thu Jul 19, 2012 2:39 pm
Re: Proposals for REALLY pushing this project forward
"ReactOS Dev Team accept patches" (tm)Carlo Bramix wrote:It's easy.
-
- Posts: 282
- Joined: Thu Jan 04, 2007 12:43 am
- Location: Italy
Re: Proposals for REALLY pushing this project forward
See also point (2) of my original post...raijinzrael wrote:"ReactOS Dev Team accept patches" (tm)Carlo Bramix wrote:It's easy.
viewtopic.php?f=2&t=15734&start=15#p122927
Re: Proposals for REALLY pushing this project forward
#1: Development is already being pushed forward in the last year or so. a lot of the major blockers have been dealt with which allowed ROS has moved to a rapid release schedule. They have also acknowledged that they needed to fix their patch process and by that more patches have been accepted & deployed.
Progress: http://imgur.com/0elCatl
Patches: https://www.reactos.org/node/926
#2: "Convert developers" - We don't need to "Convert" developers. You can't just shake a stick and developers will flood over. One way of thinking is that we already have converted linux developers because of WINE. We use some WINE components for ROS and both projects have had benefits from their partnership.
https://www.reactos.org/wiki/WINE
#3: "trash the limit of 2K3/XP API compatibility and support also newer functions" - This won't push forward development than just stall it a lot longer. Having a stable target means that they can have a functional OS sooner and add functions around it. Once a whole stable system in place, we can start fuddling with targets and won't have everything fall apart.
Progress: http://imgur.com/0elCatl
Patches: https://www.reactos.org/node/926
#2: "Convert developers" - We don't need to "Convert" developers. You can't just shake a stick and developers will flood over. One way of thinking is that we already have converted linux developers because of WINE. We use some WINE components for ROS and both projects have had benefits from their partnership.
https://www.reactos.org/wiki/WINE
#3: "trash the limit of 2K3/XP API compatibility and support also newer functions" - This won't push forward development than just stall it a lot longer. Having a stable target means that they can have a functional OS sooner and add functions around it. Once a whole stable system in place, we can start fuddling with targets and won't have everything fall apart.
Re: Proposals for REALLY pushing this project forward
I've thought I've one thing that might secure a study source of funding, approach the antivirus vendors and write up a contract that says ReactOS will come pre-installed with the trail version of your antivirus for a year if you donate x dollars/euros/etc. to funding the project (or the user is annoyed by default to get their antivirus and can then dismiss it forever). I guess you could do similar deals for other software besides antivirus, like make a search bar similar to WIndows 10 and Ubuntu that defaults to Google/Yahoo/Bing, etc. Some people will probably hate this suggestion, but money is money.
Who is online
Users browsing this forum: Bing [Bot] and 43 guests