Ver 3 Weaker graphic design distracts

Curtis Clifton curt.clifton at mac.com
Fri Jan 7 13:27:22 PST 2005


Dave,

I think you might be missing a couple of subtleties in what I was 
agreeing with and suggesting, as was I...

On Jan 7, 2005, at 2:49 PM, David Emme wrote:

> --- On 2005-01-07 12:12 PM (-0600), Curtis Clifton wrote:
>
>> On Jan 7, 2005, at 10:26 AM, sophos at mailcan.com wrote:
>>
>>> I think at least some of the cognitive interference problems would
>>> go away if the highlighting was used only for mouse selections and
>>> not for the current text insertion point.  We are all familiar
>>> with keeping track of the cursor. Removing the extra highlighting
>>> (or making it optional) would lessen the visual distinction
>>> between what I am typing now and what I might be reading for
>>> reference while I am typing.
>>
>> I'm among those that prefer OO3 to OO2, but I would agree with
>> sophos's suggestion.
>
> Just to be different, I like it (highlighting the current row) just
> the way it is. Some feel it's distracting. To me it helps me focus on
> where I am in the outline. Since I use OO for a to-do list for
> software development, I'm "constantly" referring to the outline, then
> doing something else. When I return to the outline, the highlighting
> makes it easier for my eye to find "where I left off". Perhaps if I
> were using the outliner to actually do "writing", rather than as a
> check list, I might feel differently.

I too like the selection rectangle when selecting a whole row, for 
precisely the reason you said.  But when I am _editing_ the text of a 
cell in the row, then perhaps the rectangle border is not necessary.  
OTOH, how do we know when a non-text cell has editing focus?  Hmm, 
maybe we need the  border after all.

>
>> Also, in a multi-row selection would it look better if contiguous
>> rows at the same level were highlighted with a single rounded
>> rectangle, instead of one rounded rectangle per row.

I think you missed the "at the same level" in what I wrote.

> OTOH, I can see (implementation aside) where the current behavior
> makes sense. Suppose I have a topic with several subtopics
>
>     A
>         B
>         C
>         D
>
> Selecting A draws a single rounded rectangle around the group A,B,C,D.
> Now if I want to operate on A, B, and C only (change a font or text
> color, perhaps), I click in the gutter of A and drag down to select B
> and C. I still have the large rectangle around the group, and
> independently have the selected topics (A, B, C but not D) separately
> highlighted. I suppose the smaller (A, B, C) selection could be a
> single rectangle instead of 3, but I actually think that might be
> visually more confusing, since it would look much like the large
> rectangle around the entire group.

My suggestion applied to your example would have two rectangle borders 
and two filled areas.  One would border would surround A and all its 
children; the other border would surround B and C.  One filled area 
would just be row A; the other filled area would be B and C.

This also generalizes, if I selected A-D, then a single border and 
single filled area could be used.

Below is a lame, ASCII art attempt at showing what I mean.  This will 
only make sense in a fixed width font, and even then it might not.  
(I'm probably wasting time now, but procrastination is a powerful 
motivator.  :-)

Current Behavior         Proposed Behavior

Select A, B, and C:
________________         ________________
|***A***********|        |***A***********|
|   ________    |        |   ________    |
|   |***B***|   |        |   |***B***|   |
|   _________   |        |   |*******|   |
|   |***C***|   |        |   |***C***|   |
|               |        |               |
|_______D_______|        |_______D_______|

Select A-D:
________________         ________________
|***A***********|        |***A***********|
|   ________    |        |***************|
|   |***B***|   |        |*******B*******|
|   _________   |        |***************|
|   |***C***|   |        |*******C*******|
|   _________   |        |***************|
|___|***D***|___|        |*******D*******|


Best,

Curt


More information about the OmniOutliner-Users mailing list