Returning momentarily to Blue-Beyond-Kenmore. One thing that's interesting structurally about the Blue and Green Lines and how they fit into the overall network is that they are two half-lines that sorta form one disconnected whole. The spur to Lechmere notwithstanding, both lines basically just connect one series of suburbs each with downtown, as opposed to the Red and Orange Line, which have fully developed "legs" each on opposite sides of downtown.
Obviously GLX will put an end to all that, but for 100 years that's been the relationship, a reflection of their common heritage as streetcar tunnels.
Of course, early 20th century planners toyed with the idea of hooking the East Boston Tunnel into the Tremont Street Subway, which likely would have eventually resulted in the creation of a Kenmore-to-Maverick subway, likely with the current Green Line branches at one end, and probably an LRT-conversion of the Boston, Revere Beach & Lynn Railroad on the other.
But I think this is part of why a Blue-Beyond-Kenmore conversation is inherently a bit more "mysterious" or "confounding" than other stuff we talk about. Most of the other extensions we discuss are along well-established corridors that were outlined as long ago as 1945 -- Orange to Reading, Blue to Lynn, Orange to W Roxbury, Green to Needham, "something" to Arlington, and "something" along the Mass Pike, etc. Of those 1945 corridors, the obvious remaining gap which Blue could slot into is along the Mass Pike, but... we've kinda already figured out a good way to serve that corridor -- Indigo Line EMUs comingled with regional rail. No tunneling required (and worth noting that the 1945 proposal would not have entailed tunneling either, but rather partial ROW conversion). And the 1945 vision of through-run rapid transit in a loop out the B&A and back the Highland Branch obviously is never going to happen.
But with that "obvious" candidate suddenly looking like much less of a surefire thing... then suddenly we have the whole universe open to us again. You have everything from "Kenmore-Riverway-Forest Hills" to "Kenmore-Brighton-Riverside" to "Kenmore-Watertown-Waltham" to "Kenmore-Harvard" to "Waltham or Riverside, but via MIT instead of Kenmore" to "alt route to Harvard via Cambridge and then Watertown", even to crazier stuff like "Blue eats GLX" or even the occasional "Blue bends back to hit Charlestown and Chelsea".
Obviously some of these are more buildable than others -- for example, I don't think Watertown or Waltham are ever gonna be reached by rail via Kenmore -- but I don't think any one corridor is that much more obvious than any of the others. And I think that's why a consensus for Blue-Beyond-Kenmore -- hell, even Blue-Beyond-Charles -- is very hard to reach. The "natural" layout of the T would call for a Revere-Highland Branch line paired with a Woburn-B&A line. But since the B&A is unavailable now, it'll be much more "situationally dependent" as to where the Blue should go after Kenmore -- assuming it ever makes it that far, and whether it should even go further than that.
(Parenthetically -- I dislike the idea of sending the Blue Line down toward Riverway. That points it to eventually getting extended all the way down Forest Hills, and I see very little use for that. I'd rather see an extension out toward Allston, under the theory that you could eventually extend it out further from there, if you chose.)