Data Editing Guide

Who can edit data on Many Eyes?

Currently only the creator of a data set can edit it.

How editing data works

If you are the creator of a data set, you will see the icon. Clicking the icon will bring up the data editor view where you can edit individual cells as well as perform more complicated transformations on the data.

For example, the image below shows a person about to make a change to 5 rows of a data set. At this point, the user could delete the rows from the data set or reduce the data set to only the rows selected.

Versions list

Any time a data set is revised, a new entry appears in the list below the data set preview. The entry includes an auto-generated summary of the changes, the creator, creation date and a comment (if the creator chose to leave one).

Highlighting data changes and deletions

When you look at a particular revision, the data page will highlight the changes and deletions that went into making that revision. Changes and additions appear in a blueish-green and deletions appear in red. The image below is an example of a data set with 7 rows changed.

How visualizations use data sets revisions

A visualization always uses the most recent version of a data set. (If you would like to visualize a previous version of a data set, currently you have to make that revision the top version by opening the editor and saving right away.)

Data requirements

You can only edit tabluar data. Free text data editing is not currently supported.

All data editing capabilites

• Undo last edit: You can always go back a step.
• Revert to original data set: You can always go back to the original.
• Add a row: Extend this data set by a row.
• Paste rows from Excel: Copy rows or columns and paste from Excel.
• Delete selected rows: Remove certain rows from the data set.
• Reduce to selected rows: Only leave certain rows.
• Add a column: Add a blank column or a column based on an existing column.
• Delete a column: Remove an entire column from a data set.
• Sort data by a column: Sort all rows in the data set by a certain column.
• Highlight outliers: This is a special one. Highlight statistical numeric or text outliers.
• Highlight zero or null items: Often these are the most important ones to locate.
• Transform values in a column: Apply a function to a numeric column.