Filtering and Sorting a Table that has more than 1 row per user
Demir last edited by
Imagine if... you had 6 users. They all drove cars. The table shows which cars they drive and how many miles they have driven each car. It also shows a flag which is just a way to tag a user that is important.
The License column is how many days before their driver's license expires (has nothing to do with the cars, only the user. It's the user's license).
The Last Drove is how many days ago a user drove any of their cars.
OK, so... What if you want to filter by 2 cars types (Ferrari & Mercedes):
Now the problem - What if you wanted to sort the Miles Column from low to high:
- The Name column will have to repeat user names so that the Miles column can list numbers from low to high (users are no longer grouped)
- The Flag column will have to duplicate the flag next to each duplicated user Name
- The License and Last Drove columns will also have to duplicate its numbers for each repetition of a user Name (same as flag problem)
- The license belongs to the user not the car, so it shouldn't be duplicated per car, that's very redundant data. It should be attached to the user but it also needs to be sortable.
Question: Has anyone come across this before and solved it maybe by showing the table a little differently or adding a mechanic that stops having to repeat data (name, flag, license, last drove)?
Something I tried but doesn't work is:
- I added a 'totals' row above each user that has more than 1 car. That helps sort on just the user's total (keeping their group together at all times) but doesn't work logically because I don't want to compare a user's single car MILES with another user's group of car MILES. (If user1 has 100 miles on one car and 2 miles on another, I don't want the user's group of 2 cars to be near the bottom of the Miles sorted column as 102 miles, I want to know that one of the user's cars is only 2 miles and should be near the top of the column when sorted.)
I think the problems you raise in your questions are not really problems of ordering or classifying data within a table, but that you are seeing the issue only in two dimensions as a double-entry table when in reality interactivity allows you to use more than one x/y reading.
The duplicates problem in a table is not something new, there are font management applications that allow performing a search for duplicates, offering the result in a visible way. Or spreadsheet applications with duplicate rows...
First of all, I would highlight the repeated data with color, text indenting, semi-transparency...
Secondly, I would use the interactivity as a top layer, allowing the user to decide if they want to see the duplicate information:
- Alerting there are duplicate data. With an icon/button next to the most relevant item I think is enough
- Showing duplicated data by clicking on the icon
- Showing the matched items when hovering one of them