☐ cool new feature
☐ could help many users!
☑ removes something that feels like a „bug“
☐ nice to have
☐ nobody needs it
What inspired this wish list request?
The right click new formula column menu is very helpful.
Depending on the type of the selected column, the list of entries is restricted: *)
for numeric columns: for character:
At fist sight, the restriction seems to be along the line:
this command is not possible for this data type - so remove it.
But, unfortunately, the removal of entries goes beyond that.
e.g. If just numeric columns are selected, the character/concatenate options are missing. Why?
One can concatenate numeric entries ... if they are converted to strings before the concatenation.
This sounds far-fetched - but actually it is implemented exactly like this if the user selects an column of type character in addition.
Then the numeric values are wrapped with a char() and all values are concatenated.
So, actually no reason to disable this option for the special case of just numeric columns.
A respective wish can be found here:
vice versa: If a column with type character is selected, some functions are disabled as well.
What is the improvement you would like to see?
Some of the removed functions actually work for columns with type character.
Therefore, disabling these options just restricts the possibilities of Jmp and it's users - without actual need.
Examples are:
- Count from the Aggregate menu - which just counts the non-missing entries, independent if the column type is numeric or strings.
In combination with GroupBy, this is an easy way to count entries per subgroup - also for columns of type character:
https://community.jmp.com/t5/Discussions/Row-column/m-p/594830/highlight/true#M79874
so, please, don't disable this option for columns with type character. Distributional/Rank provides an easy way to give entries a well defined label. This also works for columns with type character.In combination with the groupBy option, one can easily "anonymize" data in a synchronous fashion.This also works for columns of type character. so, please, don't disable this option for columns with type character.
Rank (reverse) and Cumulative Probability also work for columns with type character. Please don't disable them.
Why is this idea important?
Application case for the count function: count non-missing entries.
Application case for the Rank function for columns with type character :
With GroupX and GroupY enabled, often it happens that not every GroupX value shows up for every GroupY variable. Therefore there is plenty of white, wasted space. An easy trick to get rid of the space: synchronize the Y labels via rank.
before:
after:
edit:
After finding Rank in the Character submenu, it's mainly Count and Cumulative Probability which are missing.
Hm, I could live without them ...
Open( "$SAMPLE_STIPS/Wafer Stacked Small.jmp" );
Data Table( "Wafer Stacked Small" ) << Delete Columns( :Wafer );
New Column( "Rank[Lot_Wafer_Label]",
Formula( Col Rank( :Lot_Wafer Label, :Lot, <<tie( "minimum" ) ) )
);
Graph Builder(
Variables(
X( :X_Die ),
Y( :Y_Die ),
Group X( :Lot_Wafer Label ),
Group Y( :Lot ),
Color( :Defects )
),
Elements( Points( X, Y, Legend( 16 ) ) )
);
Graph Builder(
Variables(
X( :X_Die ),
Y( :Y_Die ),
Group X( :"Rank[Lot_Wafer_Label]"n ),
Group Y( :Lot ),
Color( :Defects )
),
Elements( Points( X, Y, Legend( 24 ) ) )
)