Datasets coming from a database server have a configuration option. This allows a user to toggle columns on or off, so that fewer variables are loaded in the analysis page. Or delete or copy a column; or pre-configure facets of the data.
The current configuration page shows a box for each variable, which for a dataset with a large amount of columns (which would be common when using SPOT for multi-dimensional analysis & filtering) is a bit bewildering. A tabular overview is likely to be nicer. In addition, this would allow something like "toggle all" options, in case a user likes to load just a few variables (toggle all off, then switch on the few the user is interested in).
This loses the (minimal) preview of the data inside the column. Potentially, this could be done with a hover effect over the name, which pops up a small box showing the first few variables in that column.
(I hope my understanding is correct that this preview contains really only the first few rows, and the actual dataset is not yet loaded yet.)
Datasets coming from a database server have a configuration option. This allows a user to toggle columns on or off, so that fewer variables are loaded in the analysis page. Or delete or copy a column; or pre-configure facets of the data.
The current configuration page shows a box for each variable, which for a dataset with a large amount of columns (which would be common when using SPOT for multi-dimensional analysis & filtering) is a bit bewildering. A tabular overview is likely to be nicer. In addition, this would allow something like "toggle all" options, in case a user likes to load just a few variables (toggle all off, then switch on the few the user is interested in).
Such a table would look a bit as follows:
etc.
This loses the (minimal) preview of the data inside the column. Potentially, this could be done with a hover effect over the name, which pops up a small box showing the first few variables in that column. (I hope my understanding is correct that this preview contains really only the first few rows, and the actual dataset is not yet loaded yet.)