Whoa! Thank you for taking such a detailed look! I appreciate it, and am glad it was interesting.
I'll respond to some specific points below. I do want to be upfront that, you know, I do this for fun, and so some of the nitpicks you've rightly pointed out aren't ones that I'm going to spend time fixing -- at a certain point, that's just not something that's gonna give me my jollies. (Though, if I ever do make maps professionally, I would absolutely want to hold myself to the standard you have laid out!) But, responding to some specific items that might be interesting to discuss:
Things I find too disagreeable to ignore:
- This map makes no distinction between the Fairmount Line and the Rapid Transit Lines, which I think is more than a little misleading. A 30 minute wait and a 5 minute wait are significantly different and the map should reflect that.
I've tried to be pretty up front that this map has intentionally been designed to be "future-proofed" for the eventual rapid-transit-ification of the Fairmount Line. Perhaps that means I am drawing a map for a system that will never actually exist. But designing around a "fully sized" rapid transit line through Dorchester seemed like a more difficult challenge than doing so around a commuter rail line, so I opted for the former.
- There are a couple places where geographic accuracy is prioritized over clearly showing information. The worst for me is at Union Sq/East Somerville where it looks like there is a bus/CR interchange, which could be avoided if the Lowell/Fitchburg Line split happened between East Somerville and Gilman Sq, for example.
Yes, I was actually just looking at that part of the map this morning, and noticed that the commuter rail lines could've been simplified.
(Unrelated, but Union Sq/East Somerville is always going to be awkward to map -- the 96 and 109
almost have a transfer to Union Sq, and the 96 and 47 will
almost have a transfer to East Somerville, and yet in both cases will require a ~5 minute walk.)
- Okay this one is super minor, but while I get that you're doing all the labels so they're on a white background, Harbor Point looks really silly being that far inland.
Yeah, this hadn't crossed my mind, but I do agree it could be less pronounced.
- Again a minor one, but Charles/MGH is a lot further inland than it needs to be.
There actually was a lot more trial and error that went into this than might be obvious. Visualizing the Charles River actually was a big challenge in this design -- it was very easy to make it look distractingly large ("Lake Charles"), and likewise easy to make it look too small. The positioning of Charles/MGH was part of an effort at an overall gestalt effect to make the river as a whole be less distracting on the map. I definitely agree that its current position looks somewhat odd, but it also definitely was the compromise among a number of (IMO) unsatisfactory options.
- Giving the 57 detailed stops but not giving any bus route the same treatment still feels very silly to me. Again, the only reason we consider the 57 to be any different from routes like 116/117 or 22 is that it had streetcar service for 10-15 years more than they did
Yeah, I don't have particular justification for this, other than "I like it like that". Which I grant is wholly arbitrary.
- Not showing Lynn on the map is a very significant tradeoff
Agreed.
- I'm not sure Bicknell Sq and Oak Island are relevant enough to warrant 'via' text on their respective bus routes
The "via" stuff is definitely a personal idiosyncrasy -- I do just like it. Oak Island is harder to justify, but Bicknell Sq is worth marking explicitly because it'll be a significant "waypoint" in that the frequencies drop beyond it.
- There are no arrows showing which way the Silver Line goes at Logan. Omitting them for the shuttle might be okay since the patterns there are all over the place but they are necessary for the Silver Line.
This was not intentional, and I may do a minor revision to add them.
- The CR stops at North Station, Ruggles, and Forest Hills aren't placed correctly. I'm guessing you moved the CR at some point but forgot to move the stops.
So, this was intentional, albeit not necessarily consistent in an obvious way. The "center-running tab" was used in locations where the CR line has close parallel lines on both sides. For example, using the "side tab" at Ruggles would've mostly obliterated the walking transfer indicator, and abutted the station directly against the Green Line. Forest Hills arguably could've accommodated the side tab, though.
- The 9 stops at E. Berkeley St going inbound, and Herald St going outbound
It also stops at Arlington going outbound but I didn't show that either. I really did not want to deal with the one-way pairs on the 9 (or the 111, for that matter), so I simplified.
- Through-running onto the Fairmount Line at Readville isn't shown
I definitely do admire your map's accuracy/precision on this point. You'll notice that the alignments on mine are such that it would be trivial to add in some sort of connecting line from the Fairmount to the Franklin, if needed. It seems to me, though, that the necessity of showing that connection varies widely, depending on the eccentricities of any given timetable revision. (Sometimes it matters a lot, and sometimes it barely matters.) Given that, I editorially chose to show a simpler configuration.
- Interchange bar/line things: Why is the one at Tufts thicker than all the other ones? Also the ones at Bellingham Sq and Airport are off-center
Out and out oversight there, and I'm not even quite sure how I did that at Tufts -- I don't think I had thicker lines in an earlier draft.
- Font issues: It seems like there are multiple fonts in use, and something is weird with the lowercase 'r' for the labels.
Multiple fonts, yes, and intentionally. Yeah, the r's do look weird, I wonder if something happened during the export process -- they seem to look more normal in Illustrator.
- Copley/Back Bay should have a walking transfer, that's a pretty important CR/Green Line one.
Someone recently suggested that, based on their own experience doing this walk, that it's actually not a particularly fast nor friendly transfer. I haven't had a chance to look more carefully into the question, but that was why I excluded it.
- Off-center stops at Green St, Back of the Hill, Melnea Cass Blvd, Chelsea, and probably others
This was something I struggled with. I feel like there must be an easy way in Illustrator to align the center of a polygon to the centerline of a path? But when I've tried it, it'll align it to the center of the overall path (e.g. if the path is C shaped, the polygon will get aligned into the center of the C). Any suggestions?
- I think more care should be put into the dashed lines for the walking connections. If you look at the ones on my map you'll see that I've tried to avoid the weird half-length dashes and overlap with the lines where I can
I felt the half-length dashes were needed to more visually "anchor" the connection to its endpoints (which I think may be otherwise unclear in some places). Agreed that the overlap with lines would've been better to avoid as much as possible.
- If you were to swap the positions of the 1 and 12 around Nubian you could keep the 12 completely straight rather than the little bend it does now between Melnea Cass and Lenox St
Yes, the challenge here was actually about aligning the bus stop indicators at Nubian. In the current layout, there's an implicit dogleg created by the line of "east-west" stops, combined with the pair of "north-south" stops on the SL4 and 1:
It's a little awkward that the 1 has a stop but the 12 appears to go by without stopping, but overall the stops contiguously connect all the way to the label. Swapping the 1 and the 12 would orphan the 1's stop off to the side (forgive the crude Paint.NET mockups):
What I might eventually do (if I come back to revise this map further) would be to have the 1 hook around and line up its stop with all the others:
The downside here is that the SL4 stop indicator is moving problematically close to the Roxbury Crossing label. So maybe that would need to move up, or maybe the Silver Line would need to extend all the way to the
bottom of the Nubian cluster, etc etc etc.