General comments from 2026 tutorials

[john-harrold]

Comments from @billdenney

Multiple datasets and merging

• You load SDTM data from .csv files. They are more typically .xpt or .sas7bdat files. You may want to mention that other file formats are possible to use, including .xpt, .sas7bdat, and Excel files (and make that true, if it isn't already 🙂)
• Something is cut off from the first box of the DM data. It's not clear if that is intentional, but it doesn't look intentional. If intentional, perhaps give it some type of gradient fade to the right on the column names.
• It's not clear from the video that you're renaming things visually (the voiceover does say that). It may be helpful to add something visual to indicate renaming (make a second row with a "New names" indicator, perhaps).
• It would be easier for the user if they don't have to create columns as placeholders for merging (like LLOQ for dosing). I'm guessing you want to ensure consistency, so adding a indicator like "these columns are expected not to be in this dataset" may make it easier for the user, and that could happen at the merge step.
• POSIXt columns are indicated as a number in the select pulldown. You may want to add a datetime and date classifier.
• Minor: It could be slightly easier for the user if they could rename multiple columns at the same time with a "+ add another column" option rather than repeated renames. (Though I see logic against that, too, since it adds another widget for the user to look at/use/be trained on.)
• ARD is not an abbreviation that I'm familiar with. If it's common within ruminate, defining it verbally could help just so that it sounds logical rather than arbitrary.
• You may want to consider using signif() on the numeric values to show on the pull-down. DUR has a lot of decimal places.
• For the renaming value, it could be helpful to make the values a suggested pull-down (with text entry override) so that the user can prevent typos. The suggestions could come from the names of all available datasets, other than the current one. (So that, when they type "ST", they are prompted for "STIME".)
• (though general within the transform module): Since the goal is to have columns match. It could help to have an option to "Compare columns with other dataset" and then do something like showing the output of janitor::compare_df_cols() (filtering out those that match).
  Around 9:20, you have something like this in the merge tool, already. It will help give a to-do list in the transform tool.
• When changing the columns for merging, if an equivalent column exists in the other dataset, then you could go ahead and update the column name pull-down on the right.
• It would be easier for the user to have an "add workflow to ruminate" option rather than the manual steps to add the workflow. (similar to this: ASM: Workflows formods#31)
• It'd be really handy to be able to load workflows, etc. from older versions of ruminate into newer versions. (That's hard, but will increase long-term use.)
• To reuse a workflow, it would help to have suggested data names rather than requiring the user to remember the data names.

NCA manual flagging

• It is hard to see the half-life info on the video. Consider making the half-life line thicker or some other more intensive indicator.
• You may want to remove the parentheses from the graph legend.
• It would help to give a brief description of why one would want to switch between report preview and interactive views.
• Your excel view would look cleaner if you removed the recovered workbooks and updated office.