Re: [Usability] Double-click in notification area?

[John Kodis <kodis comcast net>]

On Fri, Aug 27, 2004 at 12:02:13PM +0200, Daniele Levorato wrote:

> All the other Desktop out there works with double click... there's
> no reason to make GNOME different, breaking a standard.

This line of reasoning seems like an admission that Gnome can never do
anything any better than any other desktop, since doing things better
would require doing them differently, and could cause confusion.

> You said that nobody has shown you a good argument against single
> click... but in my opinion you haven't argued too anything really
> valid against double-click.

There are at least two fairly strong arguments: double-click
activation is used inconsistently, and requires an action that can be
difficult to perform for people with limited mobility or in limited
environments such as on kiosks or palm computers.

In addition to the inconsistency and difficulty, double-click
activation seems to violate another tenant of good UI design: that of
optimizing for the common cases.  At least in my own use of the
desktop, there are several common actions: activating an object,
dragging an object, and performing a menu-selected action on an
object; and one relatively uncommon action: selecting one or more
object to perform some action on at a later time.  Yet, wherever
double-click activation is in effect these priorities are reversed:
selecting an object is the easiest thing to do, while the more
frequently performed actions require an additional, albeit minor,
effort.

How about this as a possible approach:

- Left click activates (everywhere!)

- Left press begins a drag-n-drop operation

- Right click pops up a menu like now, but with "Select" (or perhaps
  "Toggle selection" or "Deselect") rather than "Open" as the first
  menu item.

- Selection of one or more items continues to be available through
  control-left-click, for the experienced users who seem to be the
  target audience for this type of feature.

- Double-click activation is relegated to the dustbin of computer
  history, except for its continued use in legacy desktops where it
  serves as a textbook example of optimizing for uncommon cases in UI
  design classes.

-- John Kodis.