new dev branch stuff

[Jean-François Wauthy]

Le mardi 18 janvier 2005 à 22:57 -0800, Brian J. Tarricone a écrit :
> hi all,
> 
> so HEAD is open again for major b0rkage.  i'd like to discuss a couple 
> things.  let the flamewar begin.
> 
> 1) what's our new gtk target for xfce 4.4?  gtk 2.4?
> 
I think 2.4 is the minimum but if we keep our current release cycle gtk
2.6 would be better IMO.

> 2) are we going to officially depend on d-bus?  i'd like to add 
> command-line remote control support to xfdesktop, and i really don't 
> feel like doing something lame like a named pipe or unix socket.  yes, i 
> know the d-bus API isn't finalised, and i know the glib bindings aren't 
> completely finished.  benny has more expertise in this area, so i'd be 
> interested in hearing his opinion.  methinks that the only thing that 
> *really* matters is that the wire protocol is frozen.  if the API isn't 
> frozen, we may have to fudge with some #ifdefs, but hopefully it'll be 
> frozen by the time 4.4.0 goes gold.
> 
I don't know much about d-bus but i remember when i was recoding xfprint
i missed some functionalities in mcs and benny told me d-bus could do
all that. I also suppose the API will be frozen when we'll release 4.4

> 3) autogenerated files in CVS.  they are a pain.  i want to remove 
> them.  jasper already has for the panel.  jasper is smart.  jasper has a 
> ph.d.  therefore we should all do as jasper does.  (am i embarassing you 
> yet?)
> 
I agree but i know some PhD guys who i don't listen to ;)

> (when i say "we", i mean "xfce devs")
> pros for removing:
> a) we don't have to deal with making sure they're always updated in CVS.
> b) we don't have to deal with conflicts when we run autogen locally and 
> later cvs update only to find that someone has since run autogen and 
> committed changes to all the Makefile.ins.
> c) we don't have to deal with problematic libtool versions on random 
> people's machines causing libraries to lose their .so suffixes.
> 
> cons against removing:
> a) fewer users may use CVS, thus less testing as we go along.
> b) users need to have the autotools installed.
> c) users may have problems with their autotools, and complain when it's 
> not our fault.
> 
> now, for the cons, i don't think they're a big deal:
> a) well, ok.  this one might be a big deal.  but intuition seems to 
> suggest that the users that will jump the extra hurdle to keep using CVS 
> are the ones that give the most useful feedback anyway.  we can get more 
> people using CVS by either (or both) generating CVS snapshots every 
> night, or by doing a dev release every month or so.
> b) so what?  for most users that amounts to a few simple apt or rpm or 
> yum or emerge or lin or whatever commands.
> c) well, that already happens anyway, e.g., pro #3.
> 
> that's all for now.
> 
>     -brian
> _______________________________________________
> Xfce4-dev mailing list
> Xfce4-dev at xfce.org
> http://lunar-linux.org/mailman/listinfo/xfce4-dev
-- 
Jean-François Wauthy <pollux at xfce.org>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Ceci est une partie de message num?riquement sign?e
URL: <http://mail.xfce.org/pipermail/xfce4-dev/attachments/20050119/e6bcda05/attachment.pgp>