Re: [Usability] HIG: Fixing the sort arrow direction guideline

[Braden McDaniel <braden endoframe com>]

On Sun, 2009-05-17 at 23:54 +1000, Andy Owen wrote:
> On Sun, 2009-05-17 at 08:58 -0400, Celeste Lyn Paul wrote: 
> > For those who want to know more about this proposed change, here is a common 
> > rational for the arrow position in Windows, Mac, and the web:
> > 
> > v users perceive arrow as bigger on top so the list should contain the 
> > "biggest" items first (decreasing, e.g. reverse alpha, 10-1)
> > 
> > ^ users perceive arrow as smaller on top so the list should be sorted by the 
> > "smallest" items first (increasing, e.g. alpha, 1-10)
> 
> Personally, I have always found the GNOME standard to make sense. I
> perceive the triangle as an arrow telling me a direction. If it is
> treated as a depiction of the actual items, then it feels un-natural
> (e.g. an earlier date corresponds to a shorter bar, or a word at the end
> of the alphabet corresponds to a longer bar). The idea of mapping the
> smaller and larger concepts onto non-numerical things seems like the
> sort of thing that makes complete sense to programmers, but is adding an
> extra level of abstraction.

No doubt the proponents of the current guideline can come up with all
manner of rationalizations for it; just as proponents of the opposite
convention could do the same. (And before you dismiss that opposite
convention as "unnatural", do take a step back for a moment and consider
the weight of those that have adopted this convention and the
implication of that weight on what is "natural".  You--and GNOME--are
most certainly in the minority.)

At this point, these rationalizations are practically irrelevant.

GNOME needs to justify *being different* on this point.  The current
guideline stands to be confusing to those coming to GNOME with
experience using other (more popular) GUIs.  It stands to contribute to
confusion among users who visit Web sites that have adopted the
convention used by popular desktop GUIs.  It stands to contribute to
confusion among users who typically use more than one operating system.

I think the real question that needs to be asked is: Is the current
guideline so valuable to GNOME that inviting all of this potential
confusion can be justified?

Does anyone want to make that argument?

-- 
Braden McDaniel <braden endoframe com>