It's all about re-thinking what rosapps is. Initially it was supposed
to be a place for various "app"lications. Right now it turned out to
become a kind of a trashcan, where stuff unapplicable for trunk goes.
Since this is wrong, it's perfectly fine to "force" devs to build
with rosapps (and rostests, if they are interested in thorough
testing) module.
Otherwise, with make/make all, we again end up with rudimentary rosapps.
Also there will be some kind of "unmaintained" module (undecided yet
- discussions are welcome) where some stuff from rosapps is going to
be moved (which is not currently maintained, and not interesting for
devs).
WBR,
Aleksey.
On Mar 9, 2007, at 2:38 AM, Ged Murphy wrote:
> Upon thinking about this, I'm not sure it's such a good idea with
> respect to usermode stability as it will get less testing now from
> non-included apps.
>
> This would allow us to have the best of both world.
> Many of the tests we run at application level rely on the apps which
> have just been moved being easily accessable.
> If these are gone, I suspect less feedback will be collected about the
> status of usermode.
>
> Ged.
Hi,
yes definately. Our project would certainly benefit from a good PR
and spread the word, because it eventually brings new developers and
thus development speed rises.
Thanks a lot, and feel free to email here or me.
With the best regards,
Aleksey Bragin.
On Mar 8, 2007, at 3:20 AM, Sam Banks wrote:
> Hi Guys
>
> I have recently discovered ReactOS through my research for
> Linuxforums.org, I have written an article on ReactOS, which should
> show up online in the near future. Basically I like your work and I
> would be keen to contribute some of my (unfortunately limited) time
> and skills to your project. I cant code but I can write ok.
>
> Could I be of any use to you guys?
>
> Cheers
>
> Sam
ion(a)svn.reactos.org wrote:
> Author: ion
> Date: Thu Mar 8 21:59:45 2007
> New Revision: 26032
>
> URL: http://svn.reactos.org/svn/reactos?rev=26032&view=rev
> Log:
> - Tree cleanups proposed on the mailing list. Move all non-Core OS modules to rosapps. Tests were already moved by Fireball to rostests.
>
This will break 'make bootcd'
Upon thinking about this, I'm not sure it's such a good idea with
respect to usermode stability as it will get less testing now from
non-included apps.
Another idea, instead of moving everything out of the main trunk would
be to give each component a new tag in rbuild.
This tag could be something like core="yes|no". Everything which has
been moved to rosapps could be labeled with a core="no"
2 new build commands would be available, make core and make corebootcd.
This would allow us to have the best of both world.
Many of the tests we run at application level rely on the apps which
have just been moved being easily accessable.
If these are gone, I suspect less feedback will be collected about the
status of usermode.
Ged.
I forgot to mention that I used Ged Murphy's .ppt as a base (design
mostly), and I used some parts of information about the kernel given
by Alex Ionescu.
WBR,
Aleksey Bragin.
On Mar 3, 2007, at 4:18 PM, fireball(a)svn.reactos.org wrote:
> Author: fireball
> Date: Sat Mar 3 16:17:47 2007
> New Revision: 25971
>
> URL: http://svn.reactos.org/svn/reactos?rev=25971&view=rev
> Log:
> ReactOS FOSDEM 2007 talk slides
Setting autoreply while being subscribed to a mailing list - that's
something not good to do.
Unsubscribed...
WBR,
Aleksey Bragin.
On Mar 2, 2007, at 2:02 PM, tim-sobolev wrote:
> В связи с седлительностью сервера ящик обьявлен закрытым
> Адресат переехал на tim тточчка sobolev сссоббаккка gmail ттточчка ком
> _______________________________________________
Hi,
thanks for your interest. Yes, Arty is indeed The Man to contact
regarding the PPC port, but anyway post your questions here if you
can't catch him on irc, and they will be answered sooner or later.
WBR,
Aleksey Bragin.
On Mar 2, 2007, at 12:15 PM, Tristan Miller wrote:
> Hey everyone,
> I'm interested in becoming a kernel developer, especially with
> the PowerPC port. Almost all of my experience comes from
> supervisor code (or embedded systems without MMU's), or from type-
> safe byte-code (and ironically enough, a weird combination of the
> two). However, this would be the first project that I've worked
> with others on. Any idea's where I should start. Those on IRC
> specifically pointed out Arty in regards to the PowerPC.
>
> ~Tristan Miller
Hey everyone,
I'm interested in becoming a kernel developer, especially with the
PowerPC port. Almost all of my experience comes from supervisor code (or
embedded systems without MMU's), or from type-safe byte-code (and ironically
enough, a weird combination of the two). However, this would be the first
project that I've worked with others on. Any idea's where I should start.
Those on IRC specifically pointed out Arty in regards to the PowerPC.
~Tristan Miller
