"The new row is created: Indented ..." preference needs some attention

Ward Clark ward at joyofmacs.com
Thu Sep 4 11:28:03 PDT 2003


My OmniOutliner 2.2.4 Keyboard Preferences include "The new row is 
created: Indented if there are subtopics", and I'm accustomed to this 
behavior.

When I noticed that this feature had stopped working, I investigated 
and found that there is an unfortunate interaction with the "When 
pressing return" Keyboard Preference:

	*  "Don't change current row" selected:  "Indented ..." works.
	*  "Split current row at insertion point" select:  "Indented ..." is 
ignored.

(I had recently changed my "pressing return" preference from "Don't 
change" to "Split".  I've now changed it back.)

I've pondered the interaction of these preferences, and I can see that 
this gets tricky.  For example, assume that prefs are set to "Split 
current row" and "Indented if there are subtopics", and you're working 
with an outline that looks like this:

	ABCD
		aaaa
		bbbb
	WXYZ

If the text cursor is between the "B" and "C", pressing Return could 
have two different results:

	AB
		CD
		aaaa
		bbbb
	WXYZ

	AB
		aaaa
		bbbb
	CD
	WXYZ

The latter probably makes more sense, and it's what OO does right now.

However, this row-splitting behavior doesn't seem to make as much sense 
when the cursor is placed at the end of the "ABCD" row.  In this 
situation, I'd prefer to have Return to this:

	ABCD
		(text cursor here)
		aaaa
		bbbb
	WXYZ

instead of the current behavior:

	ABCD
		aaaa
		bbbb
	(text cursor here)
	WXYZ

This started out as a simple bug report, and it's turned into a design 
dilemma.  One way of resolving this dilemma is to say that "Split 
current row" is ignored when the text cursor is at the end of a row 
because there's nothing on the right side of the split point.

If the current behavior is not changed, the Keyboard Preferences should 
at least be changed so that the "new row" preference is disabled when 
"Split current row" is chosen.

-- Ward (registered user)




More information about the OmniOutliner-Users mailing list