Comment 678 for bug 532633

scholli: sadly this issue has still suffered from the silent treatment, even during the second adjustment, so there hasn't been any substantial information about /why/ the change could have been made. That's not fixable quickly, and it's not something that I can directly fix personally---disparate ideologies, little-c corporate mentalities and strong artistic opinions will all take time to adjust but it's something that can be worked upon.

It is however possible to make _purely technical_ arguments for the lastest close-min-max order and why it might be considered superior. This is based on the combinations of buttons that can be simultaneously present:

  close (only)
  close-minimise
  close-minimise-maximise

the latest ordering is such that the positioning of the buttons that /are/ present will remain constant (relative to the edge of the window). Both of the previously-trialled orderings meant that the button positions would have altered when the complete set was not present, reducing the potential for establishing muscle memory.