The Official MBTA System Map

[Equilibria]

I have to say, I miss the “Long D” option you posted earlier as it addressed my biggest pet peeve of the current map design. Namely: the failure to depict the “Reservoir Re-convergence” of the B, C, and D Lines.

While I generally agree that readability should be prioritized over geographic accuracy, I do think there is some value in depicting in some way that:

a) after diverging at Kenmore B, C, and D do not keep diverging into the suburbs as one might assume, but actually re-converge to within walking distance of one another near Cleveland Circle before the B and D diverge again

b) the D has much wider stop spacing (and thus runs faster)

Granted, I don’t think many people will be using this as a true walking transfer (i.e. getting off one GL train and walking over to another branch to resume their trip), but for trips that start or end near Reservoir/Cleveland Circle, it would clue someone in that they could choose to make their trip via a less intuitive branch to avoid either a transfer at Kenmore or a gap in the schedule, if they’re willing to walk a bit extra.

I don’t think it makes sense to clutter up the map by depicting every single possible walking transfer in the system, but I do think that it’s worthwhile to highlight some that are partially useful and/or unexpected.

I agree with this completely. The failure to display the proximity of Reservoir and Cleveland Circle is a huge problem with the current map (derived from a map made by a guy from Moscow who had never seen the real-world layout of the system). I saw that the even spacing of the GL branches and OL was a priority for Ken Dumas, but honestly that's placing form above function.

 

[TheRatmeister]

did you ever get a confirmation beyond that reply for the map contest submissions that isn't waiverable?

Ken Dumas talks about it in his talk that's the first post on this thread. It's non-negotiable.

as well as the end of Red at Alewife.

The 77 gets harder and (IMO) uglier if you don't. I also just like the way it looks personally in both cases.

While I generally agree that readability should be prioritized over geographic accuracy, I do think there is some value in depicting in some way that:

a) after diverging at Kenmore B, C, and D do not keep diverging into the suburbs as one might assume, but actually re-converge to within walking distance of one another near Cleveland Circle before the B and D diverge again

b) the D has much wider stop spacing (and thus runs faster)

As far as I can tell you basically have to choose between having a lot of angled text and a Green Line that's somewhat geographically accurate, or almost no angled text and a diagramatic green line. For most people, most of the time, the latter option wins out. From what I understand the MPO did some testing of different designs, and people preferred the current one over the 2000s one that was more geographically accurate.

I saw that the even spacing of the GL branches and OL was a priority for Ken Dumas, but honestly that's placing form above function.

I would consider it to be the central design idea of the current map. And having made this map, it's definitely not just form. If you go back to my 'original' design, there are clear compromises made to legibility around Back Bay, Riverway, and of course many of the surface stops. (Almost) evenly spacing the branches out gives you from for unambiguous, non-overlapping labels for every station in a way that (IMO) my design can't. Realistically given the number of people who would actually use most walking transfers, it's better just to focus on a clear map. Not to mention that form is inseparable from function. Paris is a city that goes all-in on Function rather than form, and frankly I find that map almost unusable. Moscow is a great example of a diagrammatic map that turns a large, unwieldy system into one that is much more manageable through the use of regular shapes, straight lines, and yes, parallel lines representing the outer parts of the network. (god I wish I wouldn't literally be arrested upon visiting)

 

[FK4]

I also just like the way it looks personally in both cases.

De gustibus non disputandum!

 

[astroTrain27]

As far as I can tell you basically have to choose between having a lot of angled text and a Green Line that's somewhat geographically accurate, or almost no angled text and a diagramatic green line. For most people, most of the time, the latter option wins out. From what I understand the MPO did some testing of different designs, and people preferred the current one over the 2000s one that was more geographically accurate.

Fair enough. You’re certainly much more familiar with the trade-offs than I am. But the argument I’m making is that, especially for the street-level stops without bus connections, I don’t think angled text is the end of the world. Yes, a reader should be able to find Park St at a glance without tilting their head, but is the same treatment really necessary for Griggs St? I might even go so far as to say that the Griggs St label could be omitted altogether from the top level System Map if it helped make the map as a whole more readable and useful.

So if the trade-off for the “Long D” is more angled text for stops like Griggs St, I’d argue that it’s worth it. Although I can see why Ken Dumas and others disagree.

 

[Equilibria]

I would consider it to be the central design idea of the current map. And having made this map, it's definitely not just form. If you go back to my 'original' design, there are clear compromises made to legibility around Back Bay, Riverway, and of course many of the surface stops. (Almost) evenly spacing the branches out gives you from for unambiguous, non-overlapping labels for every station in a way that (IMO) my design can't. Realistically given the number of people who would actually use most walking transfers, it's better just to focus on a clear map. Not to mention that form is inseparable from function. Paris is a city that goes all-in on Function rather than form, and frankly I find that map almost unusable. Moscow is a great example of a diagrammatic map that turns a large, unwieldy system into one that is much more manageable through the use of regular shapes, straight lines, and yes, parallel lines representing the outer parts of the network. (god I wish I wouldn't literally be arrested upon visiting)

I understand the motivation, but legibility is still form to me - if you look at the map for a couple seconds more you can read diagonal text. It's not about walking transfers at Reservoir/Cleveland Circle (which is really one station). If you're a person from anywhere outside of Route 128 and you board, say, a D Branch train at Union Square to meet someone at a bar in Cleveland Circle, the current map would tell you to make a completely unnecessary transfer at one of the Central Subway stations to a C train when you should just ride to Reservoir. It's possible that the Transit app addresses this for the people who have it, but the printed map should assume that it's the only information you have.

I just don't see a disadvantage to the prior approach other than that it doesn't look clean to graphic designers. The current approach is fatally flawed if it shows two stations that are essentially co-located as being nowhere near each other. If the map is misinforming the reader, it is failing at being a map.

 

[astroTrain27]

If you're a person from anywhere outside of Route 128 and you board, say, a D Branch train at Union Square to meet someone at a bar in Cleveland Circle, the current map would tell you to make a completely unnecessary transfer at one of the Central Subway stations to a C train when you should just ride to Reservoir. It's possible that the Transit app addresses this for the people who have it, but the printed map should assume that it's the only information you have.

This is my point exactly. I’d like to add:
the benefit of showing the Cleveland Circle connection extends to the surrounding area as well. For example, the fact that the BC football stadium is accessible form the D as well as the B; or that Washington Sq is accessible from the B, C, and D would not occur to someone using the current map. But a map showing a walking connection at Cleveland Circle would clue a reader in that something weird is going on in that area and they might want to read the nearby stop names more carefully.

 

[Riverside]

I wonder if a geographic inset for Allston/Brookline/Longwood would be feasible.