I think it is clear that this is an enhancement.  So then the question is: is there a good reason to make an exception here to the "no new features in maintenance releases" policy? David mentioned some considerations.  I would add testing and documentation.  Testing is a *big* concern with IDLE since IDLE is so dependent on Tk and we know from long experience that there can be differences among the various platform-specific Tk implementations (we currently support at least 4 different Tk varieties) and we have no standard automated tests for IDLE.  So that increases the risk that we could break something in a maintenance release which we try really hard not to do.  On the other hand, there is something to be said about maintaining compatibility as best as possible across IDLE version.

I'm not fundamentally opposed to including this enhancement in maintenance releases.  But I think it is important to recognize that it is an exception to policy and to have this discussion about it.  Personally, I'm +0 on the whole feature - it's not something I would use - so I don't really have a stake in it.  I do have a bit of a stake in the testing part and for that reason I'm -0 on the backports.  I'm glad to see Todd tested on Linux and OS X - presumably someone has tested on Windows.  If the backports remain, the feature should at least be smoke-tested at some point on all four of the branches and with all of the major Tk versions (Windows, X11, OS X Cocoa Tk 8.5, OS X Carbon Tk 8.4).  Since he has the most experience in this area, I'm willing to defer to Roger's judgement call on the impact of this change with regard to IDLE extensions.