Re: [Usability] GNOME 2.6+ usability: points of critique

[Alan Horkan <horkana maths tcd ie>]

On Mon, 11 Oct 2004, Brian Skahan wrote:

> Date: Mon, 11 Oct 2004 08:41:05 -0400
> From: Brian Skahan <bskahan etria com>
> To: usability gnome org
> Subject: Re: [Usability] GNOME 2.6+ usability: points of critique
>
> On Mon, 2004-10-11 at 13:36 +0200, Robert Fendt wrote:
>
> > It mainly boils down to two essential points: flexibility and arrogance.
> > GNOME has been "simplified" more and more for some time now, almost
> > always at the expense of flexibility and configurability. And when
> > someone complains, the reactions often are along the lines of "we know
> > better than you, we will not change it back, so get used to it or get
> > lost". Want examples? Sure.
>
>
> I would like an example of someone being told "we know better than you"
> without evidence to support the claim.  I've heard this complaint

This is not a debate where facts need to be proven, it is his opinion,

Even if a user cannot prove something the fact that the sentiment exists
and you yourself admit you have heard it means that their is a
communication failure going on somewhere and we need to do better.

All feedback is a valuable opportunity to improve things, even when it
comes in the form of a mild rant (in fact I think criticism can be
especially illumination).

> regularly from people "about to switch", but I haven't really seen any
> evidence of it.  I've complained about quite a few things in gnome on
> mailing lists and IRC, I've never been told "its because we know
> better".

thankfully most developers are smart enough not to shoot people point
blank with that level of arrogance but decisions like the switch to
Spatial mode Nautilus by default was explained and justified and people
got tired of explaining it in detail.

To those not closely involved it did seem a bit rude to be told, it has
been decided and it is going to happen and the marketing of some of the
changes was not a smooth as it might have been.

> > 1) I use a focus scheme usually known as "focus on mouse contact", and
> > am used to being able to trigger mouse events in a window (i.e.,
> > 'click') _without_ raising it. While Metacity fortunately still can do
> > this (so the wm itself is not to blame here), why on earth do I have to
> > set this in a lousy 'regedit' rip-off (which of all things pops up a
> > window telling me I am not supposed to use it anyway)?

> How is a tree-node GUI interface to a complicated XML file a "regedit"
> ripoff?  There's quite a few configuration values stored in gconf that
> no-one wants to see - like x y coords for window placement or window
> size values.  How do you propose to decide which options are in the
> "front-end" configuration and which are in gconf, or what would be a
> better interface to gconf?  I'm sure there's a better interface to
> gconf, but it looks basically like an XML editor.

in some ways gconf-editor is like regedti, it is more like regedit than
any other application, and users should be forgiven for such comparisions.
(if you can direct me to more information about gconf and gconf-editor
so I can better explain this to users in future I'd appreciate it)

the underlying implementation is certainly unlike how the windows registry
works (the windows 9x registry was a disaster, the NT registry slightly
less so).

> It probably would have been better to have the GUI config for "browse by
> default" in 2.6, rather than waiting till 2.8.  That said, I don't think
> a one time move to spacial navigation constitutes "force me to re-learn
> all the time".

though you have to admit that he should have been able to upgrade from 2.4
to 2.6 and retain Nautilus in browser mode, so while he may be overstating
the problem for emphasis it is not an unreasonable comment

I'm a little suprise at the need everyone feels to be so defensive, even
if Gnome was perfect we'd still get criticism so it is best to just go
try to make the most of it.

Sincerely

Alan Horkan

http://advogato.org/person/AlanHorkan/
Inkscape, Draw Freely http://inkscape.org
Free SVG Clip Art http://OpenClipArt.org