On Mon, April 24, 2006 1:31 pm, Stephen John Smoogen said: > Do you have ideas on how this can be done cleanly? There are tons of > issues and they need to be evaluated in the order of getting them > fixed/changed. I mean I would love spatial sent back to the Windows95 > heck it belongs in... but that doesnt mean everyone else does. > Gathering the information, prioritizing it, and then championing it. As a matter of fact I do. First and foremost RHAT has to take a more proactive role leading the Linux desktop effort. It can not take the responisbility of pushing for every change to every upstream project, and it doesn't need to. However, there are a number of fundamental aspects of the desktop experience that users identify with the "distributions". These are things like: interactivity feel, spatial Nautilus, open window in backgroud, etc. Telling someone to go complain upstream about these is as if they were told to shut the fsck up. Correctly or not, these are the type of things people perceive as a distribution aspect, and the vendor needs to act on such feedback. Second, RHAT employs guys that are maintainers or at the very least have a very high status in almost all important desktop-related projects: kernel, dbus, hal, cairo, pango, gtk+, gnome, metacity, firefox, etc. For all important/highly debated issues, we need to hear RHAT's possition on the issue, so that we know what RHAT will push for. For example, if I hear that RHAT will push for changing the spatial experim^H^H^H^H^H^H^H feature to a more reasonable default, I would feel a lot happier. It will save a lot of useless flamewars in a number of lists, and I think it will really help everybody. Third, RHAT needs to listen and relay the opinion of people that use this stuff in production. It is a fundamentally different position. OSS develpers do stuff for fun, and a lot of them do it for the joy of playing with technology. They like to experiment, they like to play with stuff. See spatial, window-in-background, etc. That's fine, but a real production user is _conservative_. A lot of this changes are really trivial (they improve nothing really), but annoy the hell out of people that are just trying to get a job done. OSS developers don't work for these people, RHAT does. It needs to represent these concerns to the upstream. For example, in terms of UI, we have a model that is well known and works well: Windows. Don't deviate from the norm, unless you have _very_ good reason to do so. If you do, offer it as an option. If you think you can change the default, offer a simple way to revert the behaviour. Boring, I know, but effective. OSS developers don't like boring, this is why RHAT is making the big bucks. RHAT developers need to listen to fedora-devel. Like it or not, this is a forum for the future of fedora. Aggregate the input, and let us know where you stand. Tell us where you can push upstream, and where you can't. Don't just mindlessly send us on wild goosechases upstream. -- Dimi Paun <dimi@xxxxxxxxxxx> Lattica, Inc. -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
