I would agree it would be a nice idea to choose which at install-time,
then it would be easy to test both, and would allow any arwinss-haters
to get what they want.
Sent from my iPod
Andrew Faulds (andrewweb)
On 20 Jan 2010, at 04:42, Joshua Bailey <raptoremperor(a)yahoo.com> wrote:
I'm not a developer, so I have no say in the
matter, but if you want
an outsider's opinion, I
would suggest implementing Arwinss for the near-term, so we have
better software
compatibility. Software compatibility brings publicity, since
people will actually be able to
do things with ReactOS that they couldn't do before, like run Win32
programs. A lot of
the software I test has issues with Win32, and if Arwinss gets them
running in the near-
term it'll show people the real potential of ReactOS. Some of these
people may be
programmers, who can then funnel their interest in ReactOS towards
contributing to
development.
Arwinss is a stopgap measure. Regardless of your personal
differences in the matter, use
Arwinss to gather publicity and near-term compatibility, so
development can be focused on
other issues, such as DirectX, USB support, or better filesystem
support. When the other
components are up and running, when we have a functioning ReactOS in
beta, hopefully the
publicity that will follow ReactOS's progress will have added to the
pool of developers. This
larger pool of developers will then have the manpower to re-
implement a more "native" Win32
subsystem. While everyone in an open-source project is technically
free to go as they please,
perhaps the current pool of developers could have a gentleman's
agreement to fix the native
Win32 when the rest of the Windows components are up and running.
Everybody seems to
agree that the current Win32 is a monster, so use Arwinss to rally
more troops to battle the beast.
Even when the native Win32 is fixed, there may be a place for
Arwinss where the optional
features it offers can be used. Specialized deployments of ReactOS
could take advantage
of Arwinss, though the main branch of ReactOS would re-implement the
native Win32. A
corporation that likes the X Server features of Arwinss might decide
to sponsor ReactOS,
and that assistance could be used to progress the rest of the
project as a whole. Don't forget
native Win32, but let's see what opportunities Arwinss opens up for
ReactOS.
Perhaps the choice between Arwinss and native Win32 could be made at
install. Arwinss
would be the recommended package until native Win32 is fixed.
For what it's worth, I'm looking forward to testing my personal list
of benchmark software
against Arwinss. :)
-Joshua Bailey
_______________________________________________
Ros-dev mailing list
Ros-dev(a)reactos.org
http://www.reactos.org/mailman/listinfo/ros-dev