Imagining better designs…

…and how to keep people safe at side-roads

In the UK there are now quite a few places where people are striving to get decent infrastructure in place to support cycling.

But the rules – that we’re expected to respect – aren’t up to the job. I’ve described the problem like this; We’re trying to copy Vermeer (the painter), but using only fingerpaint.

One of the situations where we’re lacking any standardised legal, effective design is where we want to keep people safe as they cycle across the end of a side-road.

I’m going to discuss this individual issue and ask what we would design if we were free to ignore the rules (but still wanted something safe and effective). What should we be demanding from an updated set of rules? I propose a new design to prompt discussion (Plenty of healthy discussion/criticism/comment is already happening in the comments at the end of this blog.)

The problem

(copying Vermeer, but using only fingerpaint)

In other countries we can see many examples of really usable, effective infrastructure – designed and adapted through decades of development and experiment. We can see that these would often work well here in the UK (and in other places too). But between us and this utopia are a set of seemingly immovable rules – rules about what you’re allowed to build here in the UK…

…and if there’s one thing that the UK doesn’t like to do it’s to change the rules around road design (and signage/marking).

So in the UK we find ourselves trying to copy refined foreign designs, but armed with unrefined tools – a bit like trying to copy a Vermeer painting armed only with fingerpaint.

Vermeer (regular readers will notice I chose a Dutch painter) is known for the delicacy of the effects he achieved – with a refined technique that meant his paintings can glow with light and shade.

Clearly we couldn’t copy his work without the tools to do so. With fingerpaint we’d just make a mess.

Shark’s teeth versus UK give-way symbols

My favourite example of a refined Dutch/Danish infrastructure tool is the ‘shark’s teeth’ markings used to indicate ‘give way’. I’m not the first to point out how good a tool these are.

This marking clearly indicates ‘give way’ – but it also indicates a direction of travel. In this case it is totally clear to road users that you can drive away from the photographer over the marking (but that coming towards the photographer is not allowed).

Shark’s teeth can be stretched along a wide line and remain clear, or can be used in a really small space, and still remain clear.

Shark’s teeth can be used to indicate a very complex set of priorities. Can you imagine clearly marking out the priorities indicated in the image below, but using standard, legally sized/placed, UK markings?

There are legal markings available – that’s not my point – the point is that they would be difficult to understand in such a complex environment, and that they would take up too much space. This environment is already very complex and intricate – and with our UK symbols I think it would become completely incomprehensible.

Here in the UK we have one standard symbol (actually legally a combined set of symbols) to indicate ‘give way’. This is really designed around marking a simple road junction (although is strictly also to be used for a few other very specific scenarios). It works well in this situation. We can see that the side road gives way to the main road. It’s just fine given this wide space and simple message.

But once you lose the space for the full thing – once you’re working in a more confined space – or once you’re trying to communicate something more difficult – it falls well short of requirements.

The Dutch symbol is used everywhere – emphasising priorities on cycle tracks, at zebra crossings, at road ends, and even to mark priority where a cycle track crosses a road, like here:

Priorities are totally clear here, it’s obvious that a bike track crosses, and there’s a very clear indication that each lane is one-way. All this is conveyed with these very simple paint markings.

The UK symbol is really just designed to say “this is the little road – you have to wait for people on the big road.” Once we need to say something more subtle it fails miserably. Even with best intentions the result is often a real mess – really just as silly as my fingerpaint ‘Vermeer’ above.

 

Who gives way here? Maybe it’s clear when leaving the side road, but what if you’re driving into the side road? And should someone driving wait behind the proper give-way lines, or at the road end?

 

The symbol combination is to inform us that one road gives way to another. It’s definitely not intended for (or legally to be used for) the complicated message someone is trying to convey here.

Wrong spacing. Wrong sizes. Missing symbols. Wrong location.

A riot of white (and green) paint. Triangles. Big dashes indicating a roundabout. Double dashes indicating give way. Zig zags indicating a zebra crossing or similar. Green markings indicating something non-standard about bikes.

Now this is getting really silly. An attempt at a legally sized give way symbol set – but it means the centreline looks like a stop line. The dashes are probably the correct size. But have you noticed… right and left lane are switched.  Now compare to the wet line which marks the actual path of bicycles.

And in case anyone is wondering – I don’t see how any of these road markings can even be legal (for a variety of reasons – the UK legislation is very restrictive and every symbol is for a very specific purpose and is to be used in no other situations).

In an attempt to keep everything standard and safe what the UK has created is so many non-standard situations, which aren’t covered by standard designs, that we have a free for all – with everyone making it up as they go along. This is the worst of both worlds. Designers are neither sticking to the law, nor creating something which is properly fit for purpose.

It’s time for change.

I’m wondering if the time is right for some changes.

The Scottish Government, in particular, has a much more positive approach to supporting walking and cycling – in comparison to the UK (Westminster) Government.  And the Scotland Act 2016  recently devolved powers over traffic signs (which includes all the lines and symbols) to Scotland.

Would Scotland be willing to step ahead of the rest of the UK in this regard?

(Please note, I’m no legal expert – my understanding of the devolution of road signage powers is based on ordinary media reports, and a general familiarity with the UK signage legislation – I’d be pleased if someone could confirm my interpretation. Also any points in this article about the legality or otherwise of particular symbols are a lay-person’s understanding based only on a very general familiarity with the rules.)

A green stripe isn’t enough

With all of this in mind, what if we take one very specific problem and abandon the rules – but at the same time try to create something both safe and practical?

I was speaking to people recently about situations where it is NOT appropriate to use continuous footway to protect a cycleway across the end of a side road. What if we work on designs for crossing the end of side roads safely (on a bike)?

I’m a fan of continuous footway – and I’ve written several times to explain how this design feature should be used. But I’m a little worried that people here sometimes try to use continuous footway (and cycleway) in situations where it won’t work well. The Netherlands uses continuous footway (and cycleway) in very specific circumstances, and alternative designs elsewhere. (For lots more on this, see Design Details (1))

Here’s an animation illustrating our starting point (people on bikes sharing a wide road with fast traffic, and distracted drivers)…

We want something better than this.

In many places there have been experiments with the use of ‘light segregation’. Here’s a relatively well known example (Streetview link):

Rather than full separation from traffic, a combination of bollards and paint is used to keep people effectively separate. I’ve modified my first animation to show this kind of solution:

In this example (in the animation) I’m assuming we’re dealing with some kind of industrial side-road. I’m assuming that we’ve decided that the need to allow for longer/heavy vehicles – or limitations on budget  – mean that we are definitely not going to have a continuous footway/cycleway here.

The thing is, ‘light’ segregation typically becomes ‘no’ segregation (just paint) at side roads.

I’m particularly dissatisfied with the ‘strip of green paint’ style of solution. This design worries me because I see it starting to catch on. It’s not completely without value, particularly when new and bright. But embattled designers across the land are seizing on this as the new standard… whether or not it actually works …at least it can be copied from elsewhere so it’s not our fault…

Would you allow an unaccompanied 12 year old to cycle along this road? Would you feel that they would be safe passing in front of the car? What about in rush hour with people queuing across the junction?

I simply don’t think this ‘strip of green* paint’ works (well enough) in a UK environment – it’s not distinct enough to catch the attention of people unused to looking out for people cycling. And don’t forget, we typically allow our ‘paint’ to wear away to nothing before  replacing it – so even if this is a bit safer now, will it be safer in 10 years when the green has almost gone?

*(even if the paint is blue, or red – paint isn’t magic – and note I’m using ‘paint’ here as a generic term – this surface can be achieved with different materials)

And the law isn’t clear enough either. I’m certainly not going to trust that someone who kills me on this strip of green paint will be prosecuted for it.

Solutions somewhat like this might work relatively well in Copenhagen – but people driving in Copenhagen will encounter a small set of standard designs across the whole city. And of course if you look to the Netherlands – where people are the most used to looking out for bikes – designs like this are either absent are being removed for something safer.

All this got me thinking… it’s all very well to moan, but what would be better?

What solutions could we come up with in the UK for this purpose – for situations where we want people cycling to be safe, but where we don’t want to use continuous footway, or full segregation and traffic signals.

What if we were prepared to challenge the UK rules – what would we ask for? And what if we couldn’t be idealistic, but were forced to be realistic and practical? What would ‘good’ look like in these circumstances? What if we need something immediately understandable in current UK conditions?

An alternative for side road crossings

Here’s one idea – one design on the basis above. I offer it to prompt debate. I’m asking for opinions and ideas, and even alternative designs. Can you do better?

Design brief

For reference here’s the design brief I mentally wrote for myself – which others might like to work within.

I want something which:

• would keep people safe when cycling across the end of a side road;
• would work where ‘light segregation’ is being used;
• is a practical and effective design in the current UK environment (which would immediately be understood by road users);
• is not a continuous footway/cycleway*;
• would offer a clear enough indication of priority that someone ignoring it would be prosecuted.

And – as noted above – the design does not need to follow current UK rules. My design would definitely not currently be legal in the UK.

*To be clear, this is not intended as a design to be put in place where continuous footway is the answer. To understand where continuous footway is definitely the right answer, even in the short term, have a look at the articles linked at at the bottom of this article.

Design summary

I decided that this design could not use Dutch shark’s teeth because in the UK triangles have appeared all over the roads – sometimes tangled up with junction markings – to indicate speed humps or other similar changes in road surface. This proliferation of triangles (with them sometimes even being used to pretend that a hump exists in the road when it doesn’t really) has devalued them as a road marking. I’d really like to see the introduction of shark’s teeth in the UK, but we have to pick our battles. Here I want to create a design which would work even without shark’s teeth.

As many will know, a new design has become legal recently in the UK – for something called a ‘parallel crossing’. This is basically a zebra crossing, but with two lines of squares (‘elephant’s footprints’) parallel to the zebra markings – indicating the route for cycling.

There is not space in this situation for a full parallel crossing – with the full set of legal markings. However in the UK the zebra stripes are very well recognised, so I’ve based my design around a modified/simplified parallel crossing.

So would this work? In comparison to the previous situations, would the person driving be more likely (or not) to notice?

 

---

Design details

If you’d like to understand my reasoning for the various features of this particular design then read on – I’ll now list/explain them.  If you’re only interested in the bigger point then you might choose to skip to the conclusion…

Details – a non-flashing short ‘Belisha beacon’

In this design I’ve used unlit yellow globes on low black and white posts.

Is this a good idea? Or a silly one? Is it any sillier than the existing Belisha beacon? Belisha beacons are rarely bright enough to be seen in a modern context – yet they demand an electrical supply, and maintenance. Would this short post be simple to understand, even if the globe didn’t flash? Would it block the view of a child if the post was too short? Would it be a bit difficult to manufacture? It would be visible from all sides, but would a simple sign be better?

Details – elephant’s footprints

I’ve included the normal zebra stripes, and the increasingly standard elephant’s footprints, but also a green stripe to emphasise continuity.

Zebra crossings are well recognised and widely understood. Including the zebra stripes makes it much more likely that people driving assume they need to give way. I’ve simplified the markings here – omitting various additional symbols.

Elephant’s footprints are a standard marking in the Netherlands – immediately recognisable as indicating a place to look out for bikes. We should seize the opportunity to use them here too.

Is this simplified marking too simple? There’s a boundary line (indicating where to give way) at the back of the picture (above), but for clarity I’ve omitted the matching line at the closer side of the crossing. Does the loss of these symbols mean the design would be more or less effective?

Details – track at road level

The cycle track is at the same level as the road at all times.

Does this introduce any problems? Is this simplicity enough to offset the inconvenience caused by the wiggle the design introduces?

(My wife wants me to call this the ‘Weetman Wiggle’ in a bid for fame and glory…)

I’m working on the principle that this also makes the design cheaper. None of the original footway is (necessarily) needing to be removed – the new infrastructure is built where previously there was roadway.

Details – give-way marking

The design leaves a little space for the standard UK ‘give way’ marking – but with a shortened centreline.

Is this adequate? Or is this a mistake? Could we do without this marking completely? Could we come up with something new?

I should be clear – the deviation in the cycle track should be the minimum required to allow for the infrastructure which tightens the corner – this small space is intentionally not wide enough to allow a car to pull entirely off the main road.

Details – green (red?) stripe

I have used a green stripe.

It could be any other colour, but it would be good if the UK could make up its mind what colour should be used to indicate cycling infrastructure – I’d favour the nice Dutch brown-red colour. But that’s an argument for a different blog article (see Ranty Highwayman here).

I’ve started and ended the stripe a good distance from the junction. It needs to be visible to people (driving or cycling) well before they reach the junction. Being able to see the wiggle in the green stripe makes it clear from this angle that the path continues across the road end.

Details – side-road narrowing

For people coming from the side-road – driving towards the bigger road – I’ve added a narrowing of the road.

This is a standard Dutch treatment at a road end. Is this sufficient to highlight the crossing? Is there anything else (beyond the globe-posts) which would help?

Details – dropped kerbs

I didn’t get around to drawing dropped kerbs for pedestrians – nor tactile paving. But I’m assuming they would fit here – simply taking the footway to the road level.

I’ve intentionally not raised the road on a hump here.

Would the combination of infrastructure here cause any confusion for someone who is blind or partially sighted? There would be a full kerb edge separating the cycleway from the footway. Is this simple or does this introduce any engineering issues?

(While we’re talking about kerbs, I should be clear – safer angled kerbs should be being used in this design where the kerb edge is along the cycleway. I didn’t get around to drawing these).

Details – kerb line radius

The kerb edge I have drawn (detail in the image below) shows a tight corner. I want to slow people down, not keep them moving. There is a small (tiny) kerb line radius here.

I’ve included some substantial bollards to help emphasise the presence of this kerb/corner.

If this road is to be accessed by larger vehicles it will be necessary for these to be driven onto the ‘wrong’ side of the road – both when joining and when leaving the side road. This is an intentional feature of the design. I’m assuming a relatively low number of such vehicles. I’m assuming that the minor risks and inconveniences that this introduces are more than made up for by the improvements in safety for those cycling and on foot.

Is this reasonable?

What if the kerb was swept further in? How big a sweep – how large a radius – could be used before safety was significantly compromised?

I’ll run the animation again, this time showing a design with a much more swept kerb line. But remember that in these videos the person driving only behaves in the way they do because that’s how I’ve animated it. Feel free to question this animation.

 

---

To summarise…

So to return to the main questions…

Would this design work?

Is this design better than the green stripe solution? Is it practical? Would it be relatively cheap to build? Is it close enough to UK standards so that it is easy to interpret? Do you have a better idea? What are its limitations? Where would it work well? Where would it fail?

What other re-designs should we be imagining – if we free ourselves from the rules (but not the requirement for safety)?

Which rules and conventions need to change if we’re to be able to design infrastructure to properly support cycling and walking?

Surely in the UK it is time for us to step ahead of 20th century (1960s?) infrastructure tools? If we’re to build a new 2020 infrastructure, fit for the 21st century, we need new tools. We need the law to catch up… not to become ever more professional at trying to work within it.

---

Addenda

An image of this junction from Twitter jumped out at me just after publishing this article (thanks bostonian_abroad / @walking_boston)

Streetview link

Superficially it has similarities to the design I created above. It’s interesting to ask what’s different. What’s good about it? Will it work? What’s problematic?

Compare this image to mine. Is the cycle track as obvious? Why not?

How about from the side road? What’s lost by relying only on the white paint and the hump, and by leaving everything else the same?

I’d love to hear from locals about whether it works.

The following Streetview links might also be interesting – there are some similarities in these designs (to my suggestions) but also important differences.

• Shaftmoor Lane, Birmingham (4 crossings) – thanks to Phil Jones for the heads up
• Edgbaston Road, Birmingham (just zebras)

---

See also…

• Design Details (1) on continuous footway
• Amsterdam vs Copenhagen (part 2) discussing junction design
• Amsterdam vs Copenhagen (part 3) some discussion of issues with UK copies
• Read everything – New here? This is a suggested reading order.

Comments…

• As I’ve said above, my intention is to prompt debate. Please use the comments below. Feel free to criticise. Or say nice things… the choice is yours. There are already useful comments and lots to be learned in what’s been contributed.

If you landed here from a Twitter link on a mobile device you may need to press ‘Leave a comment’ below to see the comments on this article or you can reload the page to see them.

Some highlights to be found in the comments below:

• meltdblog contributed an image of a location in Melbourne Australia, where a design with some similarities does not work well.
• Sam W and bohrsatom give me an opportunity to explain more about why I’ve not suggested a raised hump/table, and why I’ve argued for the crossing to be close to the road end.
• hanneke28 contributes a Dutch perspective (directly from the Netherlands) and points out some possible modifications to the design, including methods to increase the space for large vehicles (only) to corner.
• Andy R asks/advises about UK legalities, and my suggestion not to allow vehicles to pull entirely off the main carriageway.
• Catriona Swanson contributes local knowledge around the success or otherwise of the location (with the green stripe) which I used as an example of a problematic design.

---