A strong case can be made for building Finch West as LRT from the beginning; to avoid disrupting the street twice in 20 years.
But is that worth having LRT on 1/4 of Finch and leaving the other 3/4 with buses in mixed traffic? The ridership on the Finch East bus is actually higher, but if LRT is built on the western half of Finch West, that's going to leave the Finch East riders crammed.
I'm not saying that LRT is necessarily a BAD option, I'm just saying that BRT along more of Finch is a much more equitable option. Run BRT on the entire corridor first, and then look at what sections need upgrading. It's not like the BRT investment is wasted, because:
-The right-of-way will already have been expanded
-The intersections will already be wide enough (only minor reconfigurations necessary)
-The sidewalks and everything won't have to be replaced
And realistically, once every 20 years is about right for a roadworks project. Maybe not ripping up the entire street, but usually some sort of reconfiguration is done in that timeframe.
A case can be made for LRT on Jane (between Eglinton and Steeles). Although the projected ridership can be handled by BRT, the advantage of LRT is that the vehicles can share the Eglinton tunnel and reach Eglinton West subway, or go to Yonge. Buses cannot use that tunnel, and would have to either turn back at Eglinton and create an extra transfer, or run to Bloor in mixed traffic.
Moreover, Jane has width issues between Eglinton and Wilson; it means that a properly built BRT might not be that much cheaper than LRT.
The interlining is definitely a plus to the LRT, but in my mind that isn't worth the extra billion that LRT would cost.
With respect to running to Bloor in mixed traffic, I suggest you take a look at Montreal Road in Ottawa's Vanier neighbourhood. What they've done is turned the outside lanes into peak period buses only lanes, to be used for general traffic and/or parking outside of that time. I would venture to say that that would be the best solution for Jane between Eglinton and Bloor.
Between Eglinton and Wilson, what I would suggest is queue jump lanes at intersections to begin with, and then gradual road works projects to expropriate/widen at selected areas where there's a bottleneck. North of Wilson, go with full curb side BRT lanes. It may not be the best solution south of Wilson, but it'll do, and will still generate significant time savings. The road works projects can be spread out over a period of about 5 years, to minimize their impact on the capital budget. Maybe coordinate them with some other infrastructure projects (pipe replacements, etc).
On the other hand, for Don Mills - assuming that DRL subway is extended to Eglinton, but its extension beyond Eglinton is deemed impractical - BRT might be a pretty good choice. From Eglinton and all the way up to Finch, Don Mills is already a 6-lane street and the outer lanes are already designated as HOV. So, introducing a BRT is mostly a matter of repainting the lanes and changing the signage.
Obviously, the above 5,000 pphpd projection is for the section south of Eglinton (that should be served by DRL). North of Eglinton, the demand on Don Mills is likely to be similar to Jane.
In my opinion, implementing a full BRT on Don Mills from Eglinton to Sheppard (at least) would be a great way of demonstrating how well it can work. The most expensive part of the project would be the upgraded shelters. A few cans of paint, a few new signs, and putting those new artics to good use. If the City wanted to, they could have a full BRT up and running on that corridor in less than a year. Use it as a pilot project for how/if it should be implemented on other suburban corridors.
For Sheppard East, BRT could be a viable option; but there are two issues. First is the connection to the subway terminus. LRT can dip under the 404 to reach Don Mills subway, but the buses cannot do that easily, and will have to go with mixed traffic. I don't see an easy solution for that, short of extending the subway further east.
The second issue is access to Conlins Yard if Sheppard has no LRT. Though this one is easier to resolve: either by combining an in-median BRT with non-revenue LRT tracks between Progress and Conlins, or by running a revenue-service branch of Scarborough LRT along the eastern part of Sheppard.
The subway terminus is definitely an issue. What I would try first is rush hour buses only lanes from Consumers to Don Mills. The area is going to be backed up for cars anyways, but at least that would get the buses moving. In the medium term, it would push a Sheppard Subway extension to Victoria Park up the priority list. Another option of course is to widen the overpass (I think it's actually 2 structures, 1 for each direction).
With Conlins Yard, as was mentioned earlier, there is the possibility of keeping McCowan open as a secondary yard, and use the Kodak yard as the primary yard for both the ECLRT and the SLRT. If Conlins needs to be used, yes the two options that you mentioned would work well.
You've raised some great points Rainforest. Naturally, with any plan you're going to get some snags, but I don't think that any of the concerns that have been raised here are really 'show-stoppers'. I'm liking this discussion though!