Re: Sorry State [Was: NLD10 and GNOME]

Em Qua, 2006-02-08 às 12:16 +1100, Jeff Waugh escreveu:
> <quote who="Dan Winship">
> > Two words: "bike shed"[1]. Or actually, "stop energy"[2] works too. Your
> > pick.
> This is a very sorry state of affairs for GNOME. But it is not only Novell
> and its employees who have adopted this commons-sapping, community-tearing,
> morally and intellectually lazy approach to open design and development in
> In contributing organisations, it is rationalised as a faster approach, a
> way to avoid massive discussions about inanities, and top of the list in
> these modern times, a way to avoid "design by committee" or "stop energy".
> How on Earth *do* we manage design out in the open? It is easier to avoid
> that question, in the name of getting things done.
> Outside the contributing organisations, it's appeased as something we have
> to accept to get the cool stuff, and a side-effect of our ability to involve
> contributing organisations, who have their own priorities. It sounds a lot
> like, "don't bite the hand that feeds you", whether that hand is delivering
> cool drops of code, or your pay packet.
> But ultimately, this is *killing our community*.
> And it must be fought.
> - Jeff

I think the process used by Novell is very common in the GNOME community
(and Free Software in general).

For example take metacity. Sawfish was the default window manager, so
Havoc could have started a discussion
"should-our-window-manager-be-like-this-instead". But he didn't; what he
did was write metacity following the design he had in mind in a window
manager. Metacity was included in GNOME because most people adopted it
and agreed that Havoc's design was better for the default window

The menu layout we use today is another example. If people had gone on
discussions about which is better - the foot or the "menu panel" -
perhaps things would have gone nowhere. But someone wrote the "menu
panel" and eventually it became the GNOME default.

Ubuntu has also done some changes in the panel, like the 'Add to Panel'
dialog. From what I remember this was first done in Ubuntu and after a
release using that configuration discussion started on the usability
list. Another example is the log out dialog on the right top corner of
the screen in Dapper, which wasn't proposed for discussion on, it was just implemented there when GNOME uses the window
selector for the top right corner.

There are some people posting mockups of "GNOME 3" and to be honest I
see very little discussion about them. People know (or learn) that
unless they code these mockups or convince someone to do it then most
likely nothing will happen. But if someone comes up with a different
concept for the panel and translates that into code then the community
will review and pick it up or reject it. Spending time discussing the
design first would IMO be a waste of time if the person has it clear in
their head what they want from this hypothetical new panel. The review
process will still happen, just not before the design. The design might
even have small changes after suggestions from the community, but the
basic idea of the original author is what makes this good or bad design.

I'm not sure I agree that "you can't do design by comittee" but I would
agree that a lot of the good design decisions we see in GNOME today came
from only a few coders doing their vision. I'd love to play with the
code as soon as possible but maybe there are other reasons for it not
being released yet. What GNOME can do is encourage the companies making
changes in their development branches to at least commit the patches in
a CVS branch. 

There's also the issue of who you target with the changes. Novell might
find in a usability test that the menu they designed is a lot better for
their target audience but most people in the GNOME community would
reject it in favor of the current panel layout (I'm one for example).
Should that stop Novell from doing what's best for their customers
(people used to Windows but now using GNOME because their company went
with NLD)?

My 2c.


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]