We have a lot of heavy rail in places that did not have it in the Streetcar era (Orange line, incl Assembly) and peripheral commuting is also less supported. The Everett bus study has proposed a great mix of changes and it's be good to take a similarly fresh look elsewhere
The first bridge (the "Great Bridge") across the Charles River was constructed in 1662 at the present site of the Anderson Bridge, where it connected northern and northwestern towns to the Highway to Roxbury - present day Harvard St - which ran through the site of modern-day Dudley Sq and through the narrow neck into the Shawmut along modern-day Washington St. Prior to the Revolution, the General Court only made basic inroads into the provisioning of infrastructure, which was the responsibility of the towns themselves and large-scale projects were discouraged by British authorities, in fact the "Great Bridge" was the first case of a substantial public investment in infrastructure in the State. With greater flexibility and economic expansion after the Rev. War, private corporations sought to tap into the need for improved infrastructure and birthed a complex system of tolled turnpikes and bridges to access the growing market in Boston. The Charlestown Bridge was the first, built in 1786, followed by the West Boston Bridge (the site of the current Longfellow) in 1797, the Canal/Craigie Bridge in 1809 (the site the current Dam), the South Boston Bridge (connecting to Dover St) in the early 1800s later followed by a bridge connecting to Federal St in Boston by the 1830s. Those are the historical access points to Boston.
Along which corridors does the core of our rapid transit network lie? The Red Line follows the West Boston Bridge and South Boston Bridge, the initial Orange Line follows the Charlestown and Washington St/Norfolk Turnpike access points, the Lechmere Viaduct lies along the old Canal Bridge access, even the Boylston Subway parallels (which wasn't really rapid transit) the Mill Dam Turnpike, built prior to the street-railway and connected Boston to the west and the southern Tremont St soaked up streetcar traffic running along streets that were designed to funnel Dorchester and South Boston traffic to narrow access points to Boston proper. The first time BERy deviated from this strategy was in the construction of Dorchester Extension, where the re-utilized the Shawmut Branch ROW, a decision made in part to avoid the costs of tunneling under the alignment they wanted (and served the greatest amount of neighborhood nodes) that runs/ran under the old Dorchester Turnpike (now Dot Ave) one of the earliest turnpikes.
Since then it's been re-utilizing old railroad ROWs either in spirit or as a literal replacement for Haymarket-North and the SWC, the Fitchburg/Central Mass for the RL-Northwest, the Old Colony for the South Shore, and the BRB&L for the Wonderland Ext. The the loss of nodal connections in Roxbury for the SWC realignment should bear testament to the extreme caution with which planners tinker with travel patterns that are, in many case, nigh-on 400 years old.
Somerville has become a tech bedroom community and also a destination for low wage service workers who have been priced out.
True and how are those people getting to where they need to go? Through the same process I described above. Davis-Kendall is major, major cohort (the largest origin-destination pairing that doesn't have a downtown station as the destination), but one that gloms off both a 17th century pattern (Mass Ave to Harvard Sq) and an 18th century one (Middlesex Turnpike, now Hampshire/Beacon) to West Boston Bridge. The major bus routes in Somerville, the ones that are hold-overs from the street-railway era, also travel along routes that predate even the street-car. Elm St, College Ave (note, the "College" in College Ave initially referred to "Harvard", not Tufts), Broadway, Milk St/Somerville Ave all predate street-railways. Hell, Washington St-Kirkland St-Mason-Brattle-Mt. Auburn to Watertown Sq not only predates the street-railway, it predates European settlement of Massachusetts - if you trace it's course on a contour map, it's age is clear in that skirts all major rises and formerly marshy depressions, connecting a seasonal aboriginal encampment at Charlestown to an important fishing weir and ford at Watertown Sq.
The development in the densest parts of Greater Boston followed these contours, while street-railway certainly did instigate development in streetcar suburbs, they traveled along well-established route through the more urban areas of Cambridge, Charlestown (when Somerville was part of Charlestown), and Boston. If anything, the increasing dominances of Boston and Cambridge as the "workshop of New England" is the historical pattern which was only substantially disrupted starting around 1930-1940, so more and more our current travel patterns reflect our former travel patterns.
What does a major redesign offer? I'm with insofar at improving ops, on-street infra to make bus service more reliable, but in terms of changing alignments, that's complete folly in my book in urban areas. It'd be nice to combine some alignments - for example, it's annoying to me personally that every bus west of Harvard Sq stops there without continuing east, making a trip to Inman or Union easier to stomach by foot, but that's more on an ops/streamlining question than anything. I'll yield you the parkways, though I'd say any busses running along the old MPC routes are more likely to be commuter/expresses linking distant major nodes than providing constant, neighborhood service. That was, after all, the impact of the parkways - they improved intra-town circulation, particularly for autos, but served more as circumferential routes and not last-mile connectors. Something, I'd argue is still their primary purpose in conducting traffic flow, with some exceptions for Longwood on the Riverway/Jamaicaway - if we're going to put busses on them, it's more likely to be an expansion of express/commuter service rather than a replacement for node-to-node routes.