→Why doesn't GRAMPS just use a .gz extension?
The second problem is, if we looked at the contents, we would not be able to tell the difference between a gzip'd GRAMPS XML file or any other gzip'd file. If we looked at uncompressed data, we would not be able to tell the difference between a GRAMPS XML file and other XML files. So again, the Shared Mime system could not tell us that it is a gzip'd GRAMPS XML file.
For these reasons, we must rely on the <code>.gramps</code> extension. If we don't, we would not be able to tell if this was a valid file. Even worse, the mime type of <code>application/x-gzip</code> would be associated with another application (such as File Roller or Ark) instead of GRAMPS.
GRAMPS is not unique in this problem. For example, the OpenDocument format used by OpenOffice, KWord and AbiWord is actually a collection of files in a <tt>zip</tt> archive. If you run <tt>unzip</tt> on a OpenDocument file, you will see something like: