Here's a lovely example of how non-median international boundaries can
have counterintuitive results. The attached JPG is a map of the EEZ
boundaries between Italy, Tunisia, Malta and Libya, based on one in
Prescott.
The sequence of events goes like this:
1) In 1971 Italy and Tunisia agree a median boundary, discounting the
Isole Pelagie - these islands are not used to generate median lines, but
they keep territorial sea. This Italian-Tunisian border is shown in blue
on the map.
2) In 1982, the International Court of Justice delivers a judgement on
the Tunisian-Libyan EEZ boundary. They draw the line plotted in red on
the map, which lies west of the median (dashed black line). The outer
terminus of this border is undefined, since it would depend on a (so far
non-existent) treaty with Malta.
3) No other treaties exist in the area. Future EEZ negotiations would
presumably use the various median lines (dashed black on the map) as a
starting point.

Now, Italy waived their claim to equidistance based on the Isole Pelagie
*only* with respect to Tunisia - they are still perfectly entitled to
make equidistance claims from these islands in negotiations with Libya
or Malta. The hatched quadrilateral on the map marks an area of sea east
of the Tunisian-Libyan boundary which is closer to the Isole Pelagie
than it is to either the Libyan or Maltese coast. In theory, Italy has
every right to claim this sea as its own. But if it weren't for the
non-median ICJ line, the area would fall within the Tunisian area, and
Italy would have no rights there, by the terms of their treaty with Tunisia!

Grant