When a new version of Celestia becomes available, they need to first save their modified data files, install the new version, and then manually go through their old files, one line at a time, to add their previous changes to the new data files. For a user who has a LOT of add-ons, this process can be painful and time-consuming.
Suggestion / Request for Enhancement ...
Define a second set of data file names for each of the user-modifiable data files. This would allow the user to enter their changes ONCE, into these files. The Celestia distribution would include these text files, and they would contain only the necessary identifier information, but no data. The user would add data as required / necessary instead of changing the base data files.
When installing an update, the install process would detect if any of these files already exist, and if so, NOT copy the skeleton secondary files from the distribution. This provides for an instant update, even when multiple add-ons are installed!

When run, Celestia would first read the base data files, then read these secondary *user* data files, and REPLACE or APPEND the information in them, to the information it already read from the base data files.
When a line in a base data file is to be commented out, the user would copy the existing line to the secondary data file and place the comment marker in front of that line. When Celestia sees a comment marker in a secondary data file, it checks the data it already got from the base data file and if it exists, removes / clears it.
Or, Celestia would read both, the base and secondary files, merging them before it actually sets data values.
The secondary data file names could simply have a "u" appended to them or have a new file extension, such as solarsysu.ssc or solarsys.usr.
Thoughts? Comments?
-Don G.