London Underground Map Myths


Henry Beck's rules constitute a mapping 'gold standard' so that using these rules will always result in a reasonable design

Read, see or listen to various commentaries about maps, and you will find that a large number of people regard Beck's rules as some sort of gold standard. In other words, those backward designers who have failed to follow them have created inferior maps. Those who have adopted them have had a head start in creating clear, attractive, simple-to-use designs.This myth really needs a book in order to argue the various points and strands, so you will have to bear with me, and wait for other pages to be put up on this site to get the full story. For the purpose of this article, Henry Beck's rules (whether or not he was the first to use them is a separate issue) are simply that only horizontal, vertical, and 45-degree diagonal lines are permitted. There is nothing known about human psychology to suggest that these rules are the optimum from a human perception or cognitive point of view. Absolutely nothing.

As I hope you have gathered by reading my other pages (or even my book), I belive that aspects of the design of the current Underground map leave a lot to be desired. Even within the confined space of the card folder, it could be done better. But then if you look on discussion web sites and listen to the radio or watch television, you will find people, non-expert and supposed expert alike, waxing lyrical. The current Underground map is still clear, attractve, does its job well, isn't it marvelous that it can still be like this after all these years. In many cases, I suspect that such people are not really taking a close look at the map, or not thinking about what makes a good or bad design. Either they are lazy, or they know the network so well that they are taking for granted how a novice user tries to make sense of things.

Effectively, shallow critics are looking to see whether the map obeys Beck's rules and/or it looks familiar. If it does, it must be good, after all, London Transport has a long-standing reputation for design excellence. If not, well it's only fit for one place!

Map design is much more complicated than this. Which rules are chosen is a separate issue from the quality of implementation within a set of rules. Different systems have different qualities that will make them more or less suited to a particular rule set. A good designer might create an effective map with a poorly chosen set of rules, a bad designer might make a complete hash even with a well-suited set. Time for some shock tactics! For some reason all of my map projects have names beginning with the letter 'C', so take a look at 'Cranky Map' if you dare.


It follows Beck's rules, so it must be good! [Click on the image to see the full London map.]

This map follows Beck's rules to the letter, and all station names are unambiguously placed. None overlap lines or the river, so what has gone wrong? First, there are lots of kinks. The whole map has been planned so that not only is it full of these, but also so that they cannot easily be removed (not as easy as it looks to do this). This map has been designed so that it is inept but cannot easily be fixed. KINKS ARE BAD, they hide the underlying structure of the network, and increase the 'cognitive load' (how much mental effort is required to navigate a map). Beck's early designs were clever because he turned twisty curves into straight lines that were easy to follow. Turning curves into corners is considerably less helpful. On Cranky Map there are 22 kinks inside the Circle Line (more than even on the current official map, which has 14). The map is also rather unbalanced, with cramped stations in some places and gaping holes in others. This is deliberate.

I've also ruined this map by falling for the complaints by the 'geographical brigade'. Look at Queensway and Bayswater, no one is going to travel between these two stations by Underground now, and at last, Lancaster gate is nearish to Paddington - a rather awkward interchange in my experience, but some people seem to like it (the weird interchange I've shown at the Paddington/Lancaster gate stations combines all the ambiguities of the current Underground map conventions in one convenient place). I've added an extra crank to the Northern Line to get Moorgate closer to Liverpool Street, likewise Shoreditch to Liverpool Street. Of course, the geography is wrong in other places. That is also deliberate, and increases the 'cost' of the geographical shenanigans elsewhere. Oh, and let's not forget trying to cut down on interchange circles. Charing Cross to Waterloo looks interesting to say the least. The red station names on the central map give a hint of the awfulness getting the colour of writing wrong. Just because something is corporate, it doesn't mean that it is a good colour for text.

So why did I do this, apart from to unsettle you?

First, I hope I have shown that the quality of implementation matters. The choice of rules is irrelevant if the implementation is poor, so don't evaluate the quality of a map on the basis of the rule set that it uses, evaluate a map on its own merits, a combination of whether the most appropriate rule-set has been used, and whether the rule-set has been competently implemented.

Second, learn by parody. This design highlights how a map can go wrong, with excessive kinks and lack of balance. It has numerous dismal features all over the place. Now go back to the official Underground map and see echos of these poor design features there. Can they be solved? Yes. Is it worth the effort of trying? Yes, all those millions of passengers deserve better.

Third, establish the credentials of the critics who think that the current official map is great. Show them this map, and see whether they like it (it does follow Beck's rules after all). If they don't, ask why they like the official map so much when it shares many features, albeit less pronounced. If they do like it, ignore everything else they say, they don't know what they are talking about!


Back to the myths homepage
Last updated 11/01/08