Cycle Infrastructure Design (DfT 2008)

PLEASE NOTE: This article was first published in 2008, and infrastructure has evolved since then. Please see our Space for Cycling campaign's guide for decision makers for our current views on cycle-friendly infrastructure. 

Original article:

We hope the following helps steer you through Cycle Infrastructure Design (CID), highlighting its most useful and welcome guidance, but also alerting you to some areas that are less than perfect. This briefing is by no means exhaustive, however - it is not intended to be a substitute for reading the guidance itself!

Background

The Department for Transport (DfT) published CID in October 2008. It brought together guidance from a number of draft Local Transport Notes (including LTN 1/04, 'Policy, Planning and Design for Walking & Cycling' and draft LTN 2/04 'Adjacent and Shared Use Facilities for Pedestrians and Cyclists').

It also updated and revised 'Cycle Friendly Infrastructure' (CFI), published by CTC and endorsed by the Dept of Transport, the Bicycle Association and the Institution of Highways & Transportation in 1996.

CID was published jointly by the DfT, the Scottish Executive and the Welsh Assembly Government. It is primarily aimed at local authority transport practitioners.

1) CID and basic principles

• We are particularly pleased to see CID state that: ‘The road network is the most basic cycling facility’ (1.3.2, p10) and back this up with a restatement of the ‘Hierarchy of Provision’ (Table 1.2 & text, p10).
• The Hierarchy prioritises measures to reduce traffic volume and speed, i.e. tackling the major deterrents to cycling on-road at source. At the bottom (i.e. the measures to consider last) are cycle tracks away from roads and - right at the bottom - the conversion of footways/footpaths to shared use. Importantly, CID emphasises this approach in its chapter on off-road cycle routes (8.1.2, p41) too.
• It’s also good to see CID repeat the ‘Hierarchy of Users’, which is intended to assist in design, planning and development control decisions. It puts pedestrians first, followed by cyclists and public transport – and unaccompanied private car users at the bottom (1.3.4, p10).
• Also, we welcome the fact that CID doesn’t just outline raw design principles, but gives the reasons behind its recommendations. It sets out how cyclists tend to ride and why they like and benefit from certain features and conditions, and not others.
• Additionally, CID explains how drivers react to cycle facilities and cyclists, and how this should influence design. This helps ground the advice in road user experience, attitudes and behaviour, making it - and its more ‘counterintuitive’ stances - easier to appreciate. However, as you’ll see from below, we are disappointed that CID presents some (to our mind) safety critical requirements as optional.

2) CID and cycle lanes (Chapter 7. p35-40):

a) What’s welcome

    The road network is the most basic (and important) cycling facility available, and the preferred way of providing for cyclists is to create conditions on the carriageway where cyclists are content to use it, particularly in urban areas (1.3.2, p10)

    Creating space for cyclists by taking existing footway space from pedestrians is generally the least acceptable course of action (1.3.3, p10)

    In general, off-road cycle routes in urban areas tend to be the least desired option, and it is usually better to cater for urban cyclists on-road if this is practicable. (8.1.2, p41)

Key quotes from Cycle Infrastructure Design

• CID states prominently that: ‘Cycle lanes can benefit cyclists, but poorly designed lanes can make conditions worse for them. There is no legal obligation for cyclists to use cycle lanes (or any other type of cycle infrastructure provision)…’ (7.1.1. p35)

• CID explains the need to be aware of the link between official cycle training recommendations and the presence and layout of a cycle lane:

‘Cycle lanes are not always suitable and may encourage cyclists to adopt inappropriate positioning if the lanes are poorly designed. Designers need to decide whether a cycle lane is going to help or not. If so, its alignment should ideally reflect guidance and training on safe techniques (Franklin 2007) for manoeuvres undertaken by cyclists.’ (7.1.4, p35).

Unfortunately, it’s not uncommon for cycle lanes to steer cyclists into positions that make them most vulnerable, hard to see or be seen, likely to be overtaken too closely etc.

• CID says: ‘Drivers do not always realise that cyclists need to move away from the kerb to avoid surface hazards and may expect cyclists to stay in lane regardless of its width. A narrow cycle lane may therefore give motorists (misplaced) confidence to provide less clearance while overtaking than they would in the absence of a cycle lane.’

• CID goes on to advise on what to do where localised width restrictions might tempt practitioners to reduce a cycle lane to an unacceptable width: ‘At localised carriageway width restrictions, designers can continue a full width advisory cycle lane alongside a sub-standard all-purpose lane, or the cycle lane can simply be discontinued. A narrow cycle lane should not be used here.’ (7.4.3, p37)

• CID says that two-way cycle lanes should be designed carefully, noting the potential for cyclist vs driver conflict at the mouths of side roads, where a driver waiting to emerge may not expect to come across cyclists approaching from two directions. (7.9, p40)

• CID notes that ‘A cycle lane downhill can make conditions worse for cyclists’ (because the speed differential with motor traffic reduces or disappears and the cyclist has to take up a more prominent position further out from the kerb). (7.1.6, p35)

b) What’s missing

• CID fails to clarify that a wide advisory lane is preferable to a narrow mandatory lane (although it does say that a 1.5 minimum advisory lane through a pinch-point should be considered where there isn’t enough space to introduce a mandatory lane of this width, 5.7.4, p30).

• CID says that where mandatory cycle lanes encounter side roads, they should be replaced by short sections of advisory lane to allow motor vehicles to cross them (7.2, p36). We feel that coloured surfacing, plus cycle logo only (i.e. no lane markings) is another option here. This helps accommodate cyclists riding out wider than a lane might prescribe; and stops motorists thinking that a lane’s space is all cyclists need at these junctures.

• Similarly, coloured surfacing is not mentioned as a way of maintaining the path of a cycle lane through the controlled area of a pedestrian crossing, where no lane markings are allowed (7.5.1, p37), or through a marked bus stop area. (6.4.1, p33)

c) What’s weak – and wrong

• CTC believes that cycle lanes must always be of best practice width for the circumstances and never compromised. This dictates a recommended width of 2m, and a minimum of 1.5m. If this is not possible, then introducing a cycle lane at all will almost undoubtedly cause more problems than it solves. We feel that CID does not state this nearly strongly enough, although it details the space required by cyclists so thoroughly in ‘General design parameters’ (Chapter 1, p16).

• CID should rule out cycle lanes of less than 1.5m (although we agree that they might be acceptable for short lengths in a few very limited circumstances). It is dangerous to suggest that drivers can overtake cyclists safely with less than this to spare.

Where road-space is limited, it is better to narrow the general traffic lane and mark a cycle lane of adequate width as ‘advisory’ rather than 'mandatory' (i.e. dashed rather than solid lines, advising drivers against entering it, but not making it illegal). If that is not practicable (e.g. on narrow roads with significant bus or goods vehicle flows), then a cycle lane should not be entertained. Instead, traffic volume and speed should be reduced, i.e. a solution from higher up the ‘Hierarchy of Provision’ must be found.

Specifically:

• In ‘Pedestrian refuges, traffic islands and central hatching’ (5.7.4, p31), CID says ‘There is evidence that overtaking motorists refer to the cycle lane marking rather than the cyclist when overtaking, and cars may pass too closely if the lane is narrower than 1.5 metres.’ This good, width-specific advice should be quoted again in ‘Cycle lane widths’, (7.4, p37).

• CID says: ‘For cycle feeder lanes to advanced stop line arrangements, a minimum width of 1.2m may be acceptable. Cycle lanes less than 1.2 metres wide cannot easily accommodate tricycles or child-carrying cycle trailers wholly within the lane’ (7.4.2, p37). 1.2m for an ASL feeder lane may be tolerated as an ‘exceptional minimum’ (e.g. where the general traffic lane really can’t be reduced to allow for a 1.5m cycle lane), but it should not become the norm.

• CID says: ‘In general, a cycle lane located between two all-purpose traffic lanes should have a minimum width of 2 metres’ (7.1.4, p35). We believe that cycle lanes in these locations should ALL have a minimum width of 2m (i.e. the wording ‘in general’ weakens this recommendation).

• CID says: ‘Cycle lanes should be 2 metres wide on busy roads, or where traffic is travelling in excess of 40 mph’ (7.4.2, p37). We feel that 2m should be given to cyclists wherever motor speeds of up to 40 mph are permitted (not merely 40 mph+).

3) CID and off-road provision

a) What’s welcome

• On or off-road? Along with the Hierarchy of Provision, which asks practitioners to consider converted pavements as a last resort, CID
  

• Consultation: It’s good to see CID say that ‘it is important to consult with cyclists and pedestrian groups’ when designing off-road facilities (8.1.2, p41).

• Design speed: CID recognises that ‘Cyclists tend not to favour cycle routes that frequently require them to adjust their speed or stop’ and that ‘routes with design speeds significantly below 20 mph are unlikely to be attractive to regular commuter cyclists and it may be necessary to ensure that there is an alternative on-carriageway route for this user category’. (8.2, p41).

• Width for off-road facilities: CID says ‘Wherever it is possible, widths larger than the minimum should be used. Practitioners should not regard minimum widths as design targets’ (8.5.1, p42).

• Obstacles: CID explains that obstacles such as sign poles, lighting columns, pillar boxes, kiosks etc may need moving when footpaths/ways are converted – this is often overlooked! (8.11.1, p47).

• Surfaces: CID says: ‘...minor surface defects and debris that would be of little consequence for motorised traffic can be uncomfortable to cyclists and may present a hazard.' (8.8.2, p44).

• Cycle & car lamps: CID asks designers to be aware of the dazzling that cyclists may suffer at night when using two-way cycle tracks alongside unlit carriageways; and drivers’ confusion when they see cycle lights approaching on their nearside – which can have serious consequences. (8.12.2, p47).

• High quality: CID says: ‘Pedestrians and cyclists will use high-quality, well-maintained, traffic free routes away from the carriageway if they are more direct than the equivalent on-road alternative and there are no personal security issues’. (1.3.9, p12). Opening up paths through parks or along towpaths etc. to cyclists, often provide such opportunities.

b) What’s missing

Road geometry as a factor in the on- or off-road decision-making process: the table (1.3, p13), that links traffic flow and motor vehicle speed to the most appropriate option (i.e. no facility, cycle lane, or cycle track) should also refer to road geometry/available width as crucial factors too.

c) What’s weak

• Cycle track priority over crossings:
• CID is less firm than CFI2 on the need to give primary cycle tracks priority over minor side roads.
• However, CID does stress the problems and risks, particularly to young children, of frequently having to cross side roads whilst using an off-road (road-side) cycle track (10.3.2, p65).
• CID also advises variously (10.3-4. p62-66) on a range of treatments including: flat-topped road humps; ‘bent-out’ crossings; or taking cyclists back onto the carriageway for the duration (CTC’s preferred option where traffic volume and speed allow because it avoids loss of priority altogether).
• CID also quotes research that suggests that if satisfactory crossings of minor roads can’t be provided, a cycle track may not be a ‘sensible’ creation after all (10.3.4, p65).

• Surfacing: CID cautions against unbound surfaces, but should say that bound surfaces are best for all cycle tracks, urban and rural.

4) Other features and measures

a) What’s welcome

• Risk assessments: CID emphasises the importance of understanding ‘the relative risks of various options’, guarding practitioners against assuming that such decisions are ‘straightforward’.

It’s common, for example, to install cycle tracks alongside a carriageway on the assumption that they simply must be safer than the on-road option. To illustrate the risks of such an approach, CID says: ‘A cycle track frequently interrupted by side roads can have a significantly worse potential for accidents than the equivalent on-carriageway facility’.

This is something that many campaigners have been trying to put across for years – and it’s good to note the advice that ‘it can be beneficial to involve user groups’ in the risk assessment process. (1.6.2, p14).

• Cycle audits: CID reminds readers that the cycle audit process isn’t just about checking that schemes don’t affect cyclists unduly, but that they should also be used to identify opportunities for improving conditions for them (1.7.1, p15).

•  Innovation & fear: CID repeats the Manual for Streets’ recognition of the ‘reluctance of some authorities to implement innovative schemes or schemes that do not meet all safety criteria, for fear of litigation’.

CID reassures readers that ‘…the vast majority of claims against highways authorities relate to maintenance defects rather than deficiency in design’ and explains how to avoid problems in the first place: ‘An authority should not be exposed to claims if there are robust design procedures in place where the resulting decisions are recorded in an audit trail’.

This should help encourage practitioners to be more confident about exploring untried ideas (some of which might be suggested by cyclists themselves!) (1.6.1, p14).

• Perception of safety: CID recognises that: ‘Not only must infrastructure be safe, but is should be perceived to be safe’. As we know, cycling is a lot safer than some people think – a message that’s reinforced where genuinely ‘safe’ infrastructure looks and feels ‘safe’ as well. (‘Safety’, p11).

• Consideration for cycling at the outset: CID says that cycle and pedestrian links should not be left as an afterthought when planning a new road scheme or other major works, but ‘considered from the outset, rather than being left until later.’ (1.7.7, p15).

• Length of cycle trips: CID acknowledges that cycle journeys are not necessarily short. It says that over five miles is not uncommon for commuters and that novices and leisure cyclists will cycle longer distances ‘where the cycle ride is the primary purpose of their journey’. (1.5, p14).

• Non-standard cycles: CID recognises that designers should anticipate the use of ‘nonstandard cycles’, e.g. tandems, tricycles etc. (2.6.2, p18).

• Cycle networks & accessibility: CID recommends that ‘cycling networks should link trip origins and key destinations, including public transport access points.’ (‘Accessibility’, p11).

• Cycles in bus lanes: CID supports the use of bus lanes by cyclists because they afford ‘direct routes to town centres and other important destinations.’ CID also recognises that they ‘are often preferred over off-road facilities as a result of the advantage of remaining in the carriageway and therefore having priority at side roads.’ (6.1.1. p32).

• Vehicle restricted areas (VRAs):
  • CID recognises that ‘in vehicle restricted areas where the whole street width is available, cyclists can usually mix safely with pedestrians, especially outside the main retail trading hours’.
• Campaigners are often keen to see cycling permitted in VRAs (e.g. pedestrian priority high streets) and will find this advice supportive, particularly when challenging restrictions or bans. (1.3.13, p12).
• The VRA section itself stresses the desirability of allowing good cycle access. (4.3.4, p23).
• CID explains that the potential and feasibility of cycle-shopping trips should not be underestimated. (4.3.2, p23).
• In situations where a council is proposing to allow cycling in VRAs but meets with opposition, CID suggests that experimental traffic regulation orders (TROs) should be introduced to see what happens. As trials rarely (if ever) confirm the worst fears of objectors, this is an advantageous tactical approach. (4.3.7, p24).
• CID warns that marked cycle routes within pedestrianised areas may lead to higher cycle speeds and a greater potential for conflict (4.3.8. p24); and explains that ‘Careful urban design can help to create an attractive and functional environment in which cycle speeds are low and pedestrians have priority’ and advises that trees and benches etc can suggest a ‘preferred route for cyclists’ without the use of road signs. (4.3.10, p24).

• Road closures: CID advises that, when closing off the end of a street, ‘consideration should always be given to allowing cyclists to continue using the route by installing a cycle-gap in the closure’ - excellent advice in the interests of accessibility, directness and convenience (although we’d recommend that 1.5m should be the minimum gap, not 1.2m). (4.2.1, p22).

• Cycle exemptions: CID states that, unless safety concerns dictate otherwise, ‘Cyclists should usually be exempt from prohibited turning movements or manoeuvres…An Order giving effect to the prohibition will need to exempt cyclists.’ (4.2.4, p23).

• Removal of centre lines: CID explains why and in what circumstances the removal of centre white lines may help reduce speeds and how this can be done to benefit cyclists – (e.g. by providing adequate advisory cycle lanes on each side) (5.3, p28).

• Coloured surfacing: CID sets out a comprehensive list of situations in which coloured surfacing may be appropriate (3.2.3. p19). It also states that such surfacing is ‘especially useful for cycle lanes away from the kerb, such as non-nearside cycle feeder lanes for an advanced stop line [ASL] layout, or where a cycle lane runs along the offside of a dedicated left-turn lane’ (3.2.2, p19) (but it would be good to refer to this advice later in the discussion on non-nearside ASL feeder lanes (9.4.9, p56)).

CID doesn’t mention the option of continuing the cycle lane colour, but not the lane markings, when it crosses a side-road entrance. (3.2.2, p19). See also 2.b, 2nd bullet point above.

• Access control: CID spells out the problems that some types of barriers (e.g. those barring motorcycles) can cause cyclists, particularly riders with non-standard machines, panniers, etc and offers advice on best practice (8.14, p48).

CID also stresses that ‘Motorcycle barriers should only be introduced after a definite need has been established…’ – not always the case, unfortunately. (8.14.1, p48). We would, however, like to see CID stating a preference for cattle grids over self-closing gates to prevent livestock escaping, because the latter may force some cyclists to dismount.

• Roadworks: CID recommends a minimum gap of 4 metres so that cyclists can get past roadworks safely (unless there’s a cycle bypass or additional features significantly to reduce motor speed). (5.7.2. p30).

• Gully Gratings: CID says that gully slots should be installed at right angles to the direction of cyclists’ flow to avoid catching wheels (although this should apply wherever they are located, not just at transition points from cycle track to carriageway). (8.9.3, p46).

• Tactile paving: this explains correct installation. Our experience shows that authorities/contractors have been known to get this wrong. It would be helpful, though, if the advice emphasised the dangers of cyclists’ skidding on this sort of paving rather more strongly. (8.16, p50).

• Junctions: highlights in this section (Chapter 9, p53-59) are advice on:
  • why, when and how to extend the inter-green period for the benefit of cyclists at signalised junctions (9.2.2, p54); cycle lanes that bypass the main signals with their own phase or green signal (9.3.2, p54);
  • the strong advantages of continental-style roundabouts for cyclists (9.7.3, p58);
  • the advice that keeping well to the nearside on the circulatory carriageway puts cyclists ‘in the most hazardous position for being hit by vehicles entering or leaving the roundabout’ (9.8.1, p58); and how cyclists do (or ideally should) negotiate roundabouts and the ways in which design can serve their best interests (e.g. by inducing lower speeds and reducing capacity).

b) What’s weak

• Busy road crossings: CID says: ‘There should be provision for crossing busy roads and other barriers…’ (‘Accessibility’, p11), but rather weakens this shortly afterwards: ‘The needs of pedestrians, cyclists and equestrians should be considered where their routes cross busy roads, especially in rural areas’. (‘Safety’, p11).

We believe that the provision of good crossing points for non-motorised users (NMUs) over all busy roads is non-negotiable. Busy roads create one of the most serious and perennial barriers to NMU movement, not to mention community severance.

• Two-way cycling:
  • CID considers two-way cycling largely in the context of cycle lanes (7.9, p40). Rightly cautious about two-way cycle lanes (see 2.i, 4th bullet, above), it should however also mention that two-way is a good option for cycle tracks along high speed dual carriageways (if safely set back).
  • The advice on minimum widths for two-way cycle tracks and the need to make extra allowance where there are kerbs, walls etc, is somewhat unclear. In ‘Cyclists passing other cyclists’ (2.4, p17), CID says that 2.5m is the minimum desirable width for a two-way cycle track, but doesn’t note the extra allowance that needs to be made for kerbs, walls etc, even though this is detailed in the previous section (2.3, p16). Later, under off-road cycle routes (8.5.3, p43), CID does ask practitioners to make additional allowances for vertical features, but quotes 3m, rather than 2.5m for a two-way track: ‘The minimum recommended width for a two-way cycle track is 3 metres. If these widths cannot be realised, the facility may become difficult for some people to use.’ (8.5.2, p42-3). It is not therefore entirely clear if CID is saying that it thinks 2.5m (minimum, regardless of necessary clearance) caters for two-way cyclists only; while 3m (minimum, regardless of necessary clearance) caters for shared use (as it states later in 8.5.3, p43). In any case, pedestrians usually share two-way cycle tracks with cyclists, so 3m is usually the most appropriate minimum width (before taking necessary clearances into account).

• Contraflow:
  • We welcome CID’s support for cycle contraflow schemes because they help cyclists avoid a ‘longer, heavily trafficked alternative route’ that might not necessarily be safer just because it’s ‘with-flow.’ (1.6.3, p15); and that they provide ‘…permeability for cyclists when the movement of other traffic is restricted by one-way systems.’ (7.6.1, p37)
  • However, it is disappointing to see CID say that it is acceptable to ‘reduce or omit the buffer zone sometimes provided between parking bays and cyclists’ in contraflow schemes (7.7.1, p39). This conflicts with 7.1.4 (p35), which says that cycle lane arrangements should reflect cycle training advice. National Standard instructors always tell cyclists to keep out of striking distance of carelessly opened doors. Buffer zones should never be omitted. If width is a problem, then no cycle lane should be marked and cycle symbols considered instead.
  • The guidance should also ask planners and engineers to consider removing parking spaces altogether where they could obstruct or endanger contraflow cyclists.
  • Echelon parking on the nearside of contraflow should be ruled out unequivocally. (7.7.2, p39).

• Traffic calming, narrowings and cycle bypasses: CID explains how physical traffic calming, narrowings and other measures that have the effect of reducing speed (speed humps, cushions, build-outs, refuges, hatching etc) can create problems for cyclists.

It rightly recommends installing cycle bypasses at the pinch-points that some of these features create, where practicable. We would argue, however, that gaps at road closures, and clear spacing of bollards, should be a minimum 1.5m wide to allow for tandems, trailers etc (rather than 1.2m); that gaps from kerb to a speed cushion should be 0.75 – 1.2m, with 1.2m preferred (rather than between 0.75m and 1m)3; and that an alternative to chicanes is always preferable where safe and convenient cycle bypasses can’t be fitted. (5.2-5.8, p28-31).

• Roundabouts:
  • CID should state categorically that, as annular nearside cycle lanes put cyclists in danger, they should be ruled out. (9.10.2, p59).
  • CFI rightly advised that dedicated left-turn slip lanes should ‘preferably be avoided entirely’ unless controlled by signals (17.2.3, p63); CID is weaker: ‘they are not generally recommended on cycle routes’. (9.7.1, p58).

• Dropped kerbs: CID says: ‘Dropped kerbs should ideally be flush with the road surface’ (‘Comfort, p11; 8.9.1, p46). We believe that dropped kerbs should always be flush, given the hazard that any upstand poses to cyclists (particularly when hidden by leaves or puddles etc).

• Wheeling ramps: we welcome this section, but suggest the norm should be to provide usable ramps on both sides of a flight of steps. (10.9, p68).

• Trams: while CID sets out the hazards that tramlines and tram services present to cyclists and some solutions (12.5, p78), it should stress the importance of factoring in cycle safety considerations right at the start of designing a tram project.

• Conflict with walkers: a significant proportion of walkers undoubtedly find sharing space with cyclists intimidating, but CID should state that actual conflict is more a fear than a reality.

• Maintenance (8.17, p50 – 52):
• As this is covered under ‘off-road’, it tends, unfortunately, to ignore road defects such as potholes and the criteria for repairing them. Also, the advice explains the importance of maintaining ‘cycle routes’, whereas we would argue that the whole network – or at the very least, all roads and paths that are well-used by cyclists - need attention.
• The ‘Typical maintenance programme’ table (8.4, p51-2) should advise: that leaf litter and debris needs to be cleared 4 times a year, including at least twice during leaf-fall; that the actual need for each access barrier should be reviewed during inspections; and it should also mention repair to markings.
• However, CID does go beyond the standard reference on maintenance standards for cycle routes, and even suggests that ‘routine and safety inspections are best carried out from a bicycle.’

c) What’s missing

• Climate change: a direct reference, in the introduction, to the contribution that cycling can make toward helping a local authority address climate change.

• More on 20mph: CID should mention 20mph much more often, simply because it is perhaps the most useful cycle facility of all. 20mph does appear in ‘Speed-reducing measures’ (Table 5.1, p27), but there are a number of other contexts in which it should be highlighted (e.g. as the design speed for new-build residential streets (4.4.3, p25); or amongst the recommended requirements for child cyclists (1.3.8, p12) etc. There is no reference to enforcement (e.g. via cameras etc) at all.

• HGV management: this is another ‘invisible infrastructure’ measure that CID leaves out, although it was covered in CFI (e.g. 10.3.1, p43, CFI). HGVs have a disproportionate impact on cyclists’ safety.

• Removal of ‘Cyclists Dismount’ signs: CID explains that ‘Cyclists Dismount’ signs are overused and rarely appropriate, adding that practitioners should ‘be able to defend their decision [to use the sign] and explain why it cannot be avoided by design’ (3.6, p20-21). This is good, but CFI’s advice was stronger – it said that employing other techniques should lead to the signs’ being ‘progressively removed’ (12.1.7, p53, CFI).

• Relative safety for cyclists at different types of junctions: CID’s chapter on junctions (9, p53) should set the scene with this vital discussion.

• Grade separation at large roundabouts: CID should advise specifically that this is an option for cyclists/pedestrians over large, busy roundabouts, esp. unsignalised. (9.9. p59; 10.7, p67).

• On-carriageway cycle logos: in ‘Signing Issues’ (3.1.3, p19), CID asks practitioners to avoid unnecessary visual intrusion with the over-use of signs and coloured surfacing. It suggests alternatives (e.g. wall mounted signs), but omits the relatively unobtrusive, but useful oncarriageway cycle logo.

In fact, logos (maybe combined with coloured surfacing) should be flagged up as good way, in certain situations, of indicating where cyclists expect to share space with motor traffic, but need more space than a marked cycle lane might allow, e.g. past the mouths of junctions, parking bays etc. CID covers the ‘Cycle Symbol’ in 3.3, p20.

• Legal issues associated with footpath conversion: information on this is lacking, as is a recommendation that authorities dedicate a ‘restricted byway’ rather than a cycle track when converting a footpath because once it’s converted into a cycle track, the path is removed from the definitive map – something that tends to generate objections.

• Residential developments – levels of cycle parking: CID notes the significant effect on cycle use that having a ‘cycle ready and available at the front of a house, rather than locked away at the back’ can have (11.2.2, p71). However, it would help if it also set out the recommended levels of cycle parking at new developments.

• Road studs as a hazard: there’s no advice on the installation of road studs, which can cause skidding if they are thoughtlessly situated and/or designed.

• Loop detectors for cyclists at bus gates, rather than forcing them to stop to press a button.