Currently Date variables in Displayr and Q are always sorted from oldest to newest (left to right) when used as columns. But this isn’t always how users want to represent the data, some want to view it from newest to oldest or alternative sorting methods. Regular variables can be manually re-sorted but dates can’t. Allowing users to manually sort date variables or even just allowing them to flip the sorting would add functionality both products currently lack.

When a filter is applied to a table with a variable, the empty periods don’t show:

!_name=image (f32cf033-c85a-4154-bf3f-7c34f6e90fe3).png|width=1074,height=286!

However, if we replace the format variable with a categorical or variable, the columns that have their sample filtered out will appear, like this (February and December are now both here), but showing as empty because of the filter:

!_name=image.png|width=100%!

There should be better control of data variables regarding how they show or hide to remove any inconsistencies.

Allow users to create a nominal variable from a date/time variable via the ready-made new variables menu, bypassing the need for them to use R code. They can then use this nominal variable in filters or anywhere else in their analyses where a date/time variable is not appropriate.

You could have them select the date/time period aggregation level in a prompt and then perform the conversion, rather than having them remember that they need to do it directly in the date/time variable upfront.

I’m tired of remembering how to do this/finding the help article with the info myself when I haven’t had to do it in a while. Yes, I know it’s simple R code.

This applies to both Displayr and Q.