Call Monday 24 Jan 2005

[Daniel Stone]

On Mon, Jan 24, 2005 at 06:35:42PM +0100, Roland Mainz wrote:
> Daniel Stone wrote:
> > On Mon, Jan 24, 2005 at 08:11:59AM -0800, Alan Coopersmith wrote:
> > > Adam Jackson wrote:
> > > >Are you seriously suggesting disciplinary action for a change that appears
> > > >to have majority community support and that we have no established policy
> > > >for doing any other way?
> > >
> > > I don't know about what action should be taken, but I don't believe there is
> > > any sort of consensus or majority support for dismantling the monolithic
> > > tree, unless and until a plan is in place to get a working modular release
> > > going.   Destroying what we have without a replacement is just foolish.
> > 
> > Again, I would like to reiterate that, despite my intense dislike for the
> > current situation on more of a level than even just the security nightmare this
> > creates, *no* code was removed from the tree.
> 
> No, you just changed the default build options of parts of the tree
> which you do...
> a) ... NOT OWN!!

No-one owns config/cf.

> b) ... were part of X11R6.3, 6.4, 6.5, 6.5.1, 6.6. Unfortuntely the
> X11R6.7 release introduced a glitch from the Xfree86 repository which I
> fixed (I did this work as I try to care about all the (reasonable (not
> things like https://bugs.freedesktop.org/show_bug.cgi?id=2363)) bug
> reports if I can handle them or forward them to a matching owner). And
> that glitch does NOT give you the right to UNDO that without going to
> the normal procedure of depreciating things (e.g. via xorg_arch etc.).
> There was no EOL notice or anything else which inidcates that the XRX
> technology is depreciated.

I would dare say that code doesn't just disappear entirely from the tree by
accident -- misguided as you may believe or not, I doubt this was anything but
intentional.

I have yet to see the xorg_arch model used for anything.

One of the blockers for modularisation is that we do not have a plan, and
obviously cannot move forward without one; if we're going to insist that
xorg_arch EOL is a formal process (is it?), then can we at least please have it
documented?  Getting slapped over the wrist for not following conventions that
don't exist is a little odd to me.

> > In fact, thanks to XRX, today has
> > actually been a net gain in terms of the amount of code in the monolithic tree:
> > the beast just got bigger.
> 
> Bigger where ? You have turned XRX OFF in the default build for _all_
> platforms!

The number of lines of code in the tree is quite a lot bigger.

> Daniel: What would you say when I would (theoretically!! (I am not
> planning this, the question is only of theoretical nature!)) start to
> rip off all the Debian-specific parts of the Xorg tree ? Wouldn't you
> scream and complain - [YES] or [NO] ?

Actually, no, because I don't believe distribution-specific packaging concerns
have a place in the upstream tree.  This is why I haven't attempted to work the
debian/ dir into X.Org CVS; similarly, why I have avoided doing so for D-BUS
(which I co-maintain), and have in fact actively resisted many calls to do so
for the packages I have maintained during the years.

(Short answer: no, I would not complain.)

Again, this has not removed any support for XRX.  All it has done is require
people who want it, to enable it.

As an aside, if it's a Mozilla plugin, would it not be better being built in the
Mozilla tree?  The standard X implementation building browser plugins seems
absolutely perverse to me.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://lists.freedesktop.org/archives/release-wranglers/attachments/20050125/3002c34e/attachment.pgp