Can I cycle vs. would I cycle?

18 posts / 0 new
Log in to post comments
Last post
13 February, 2011 - 22:37
#1
christhebull
Can I cycle vs. would I cycle?

This is an idea for a quiz, but it could also be good PR. Essentially, we take pictures of various locations, show them to people, and ask them if you can (legally) cycle there, and then ask them if they would cycle there. The idea is to get people against pedestrianised areas, pedestrian only road blockages, and contraflow bus lanes having cycling banned, as well as other obstacles to urban cycling like one way streets, but in favour of cycle paths on 70mph dual carriageways which are notionally open for cycling, but in practice treated as cars only.

Submitted by christhebull on 13 February, 2011 - 22:37
Top
2 March, 2011 - 09:32
yojimbo

I like this idea! Perhaps we copy the format of ‘ScenicOrNot’ (http://scenic.mysociety.org/). Personally I think people’s answers to the ‘Would I cycle there?’ would be more interesting, and potentially very useful for PR/lobbying. Might also be useful to get people to put down their age and gender, as other evidence suggests youngish men are more risk-loving and thus more amenable to current cycling conditions.

Submitted by yojimbo on 2 March, 2011 - 09:32

Jim Gleeson

Top
2 March, 2011 - 10:51
(Reply to #2)
christhebull
A good place to use for this would be Guildford (or at least, a place I'm familiar with). The A31 dual carriageway to the west has a 60 mph speed limit but would appear to be a good place for a cycle path. There is plenty of room, albeit much of it is between the two carriageways because they are so far apart; and quite possibly the demand, because a bus stop on the road has been closed and residents of a nearby village have complained that the road is unsafe to walk on, which is not surprising when there is no pavement. In contrast, the lower, cobbled part of the historic High Street is a one way street, mostly closed to all vehicles during certain hours. This is despite the fact that the pedestrianised section forms both directions of National Cycle Route 223, with stickers at the top pointing cyclists the "wrong way". This is hardly surprising when you look at the route traffic is supposed to take. The street is quite wide, so contraflow cycling would be safe even with parked cars.
Submitted by christhebull on 2 March, 2011 - 10:51
Top
10 May, 2011 - 08:02
pj.mcnally

this is an awesome idea – once back in the UK, I’ll start taking photos to suit!

(currently in Sri Lanka. don’t get me started on road conditions here).

Submitted by pj.mcnally on 10 May, 2011 - 08:02
Top
18 May, 2011 - 12:20
yojimbo

Just to follow up on my earlier comment, I’ve been thinking more about how to develop this idea, in a slightly different direction from Chris’s original thought.

One of the problems in cycling campaigning at the moment is the complete lack of a comprehensive, comparable indicator of the ‘cycle-friendliness’ or ‘subjective safety’ of different areas. It would be great to be able to gather people’s subjective impressions of different streets/junctions and aggregate them all up so that we know which areas (from individual streets to whole local authority areas) people think are more or less inviting to cycle around. This would be interesting in its own right but also potentially a useful campaigning tool. To date, the logistical obstacles to gathering such data have been fairly prohibitive. But it may now be possible, through the use of technologies like Google Streetview.

My proposal is for a website that presents people with a succession of Google Streetview images (displayed using the Google API), taken randomly from a set which includes a representative sample of streets and junctions, say 50 per local authority area. You ask people to tick a box indicating whether they would ‘Definitely’ cycle there, ‘Definitely not’, and points in between. If enough people do this, you start to build up a representative picture of how cycle-friendly each street and local authority is. And if you get people to identify at the start of the process whether they are male or female and young or old, then you can also build up a picture of how different demographic groups respond to different areas.

There are various design issues that crop up. It would be important to attract all kinds of people to the site, non-cyclists as well as cyclists, and to get people to keep on rating different areas. I’m not sure whether it would be better to manually pick the Google Streetview images (this would probably have to be done by people familiar with each area), or to have them come up randomly. I think the latter might over-sample areas where not many people live.

More fundamentally, I am no coder and don’t really how to go about making something like this (though I would be able to have a good stab at analysing the data). Tom Scott has done a funny game about guessing crime levels from Street View images (http://www.tomscott.com/stabbed/), and when my college exams finish in June I intend to approach him to see if he has any helpful tips for how to go about making something along these lines. But in the meantime I’d be very interested to hear views from anyone on the forum!

Submitted by yojimbo on 18 May, 2011 - 12:20

Jim Gleeson

Top
18 May, 2011 - 13:36
(Reply to #5)
sallyhinch

although as you say I suspect doing it purely randomly might be a problem. Would it make more sense to pick streets from within a radius of their postcode, or within their postcode if they only give the first half? That would mean they’re more likely to be familiar with the street in question – not sure if that makes it more or less useful. I suppose how to do it depends on the hypothesis being tested. If you just want to find out what kinds of streets are more inviting, or which local authority is most bike friendly overall then random would be the way to go, but if you want to test whether streets with actual bike infrastructure are more inviting than without then you’d need to provide a balanced set of paired examples (similar looking streets with and without infrastructure). I think you’d also want to identify beforehand how much people actually cycle as well because that will affect the results.

What might also be interesting is to find out whether google street view images which have (randomly) captured a bicycle look more inviting than those that don’t (e.g. by removing the cyclist from the scene)… In fact, there’s loads you could do with this.

Submitted by sallyhinch on 18 May, 2011 - 13:36

AKA TownMouse

Top
18 May, 2011 - 14:00
(Reply to #6)
Jim

There’s certainly loads you could do with this. This is a brilliant idea and discussion that I think needs to go into the ‘website development’ part of the meeting in Manchester this Saturday. Mapping is right up Anthony’s street (no pun intended).

Submitted by Jim on 18 May, 2011 - 14:00
Top
18 May, 2011 - 19:52
(Reply to #7)
yojimbo

Thanks for the feedback Sally. The idea of getting people to compare similar streets with different levels of infrastructure is an interesting one, though it might be hard to find good examples! It may be possible to get at the question by separately coding each image in the database in terms of infrastructure then comparing those codes with the subjective scores given by the public. And as you say it would be interesting to see how people respond to images that contain bikes (or HGVs …)

Other Jim, unfortunately I can’t make the Manchester meeting but obviously very happy for this to go on the agenda.

Submitted by yojimbo on 18 May, 2011 - 19:52

Jim Gleeson

Top
18 May, 2011 - 22:01
(Reply to #8)
sallyhinch

whether some sort of a mashup could be done between Open Streetmap (which has coding of cycle facilities – at least in places) with Google streetview… Unfortunately OSM isn’t that complete and certainly some local authorities are more complete than others

Submitted by sallyhinch on 18 May, 2011 - 22:01

AKA TownMouse

Top
7 September, 2011 - 09:38
(Reply to #9)
PaulJames
PaulJames's picture

Hey Jim, just come across this post via your blog. This is an awesome idea. I might be able to help with implementation, this sort of dev stuff is what I do as the day job.

Submitted by PaulJames on 7 September, 2011 - 09:38
Top
7 September, 2011 - 13:44
(Reply to #10)
Jim

Thanks Paul

One issue that was raised at the launch picnic was that a photo only captures part of the story and is no real substititute for riding it. More specifically, just looking at a picture gives no indication of the purpose of a particular road, such as whether it feeds into a Trading Estate, or if it just skirts a new housing development, or an out of town shopping development. All will be factors on type of traffic, volume and speed.

This is going to be more complex but it would be worth seeing, when the photo is brought up, if anyone is already campaigning locally in that area as they will have a greater knowledge of any issues (such as ‘has this road been raised with local Councillors and the Highways Authority as an issue?’ or ‘are there already plans afoot to add this to a local ‘network’ by converting the pavements to shared use etc?’). It would be worth getting people to leave comments for each photo and even organising an ‘Infrastructure Safari’ where a group can ride out to the area covered in the photo to give a little report on impressions.

That said, as a simple exercise in ‘Subjective Safety’, it would be interesting however it is implemented.

Following on from what Chris said elsewhere in this thread, the A31 and Guildford town centre are very well known to me. I’d happily write a thesis for you on how lethal the A31 has become between Guildford and Farnham and, more crucially how the space is there to make it better.

Submitted by Jim on 7 September, 2011 - 13:44
Top
7 September, 2011 - 22:19
(Reply to #11)
PaulJames
PaulJames's picture

I’ve been having a think about this this evening, the Place Pulse way of having two pictures to compare would work the best IMHO as it’s simple for people, since this will need to be aimed at the general public, and not people who cycle.

Rating on a scale is hard too, people need to compare with other pictures and we’ll end up with most people will just saying no, they wouldn’t feel safe to all of them. Which won’t tell anyone anything.

The trick is sourcing the imagery and the locations, we’d need a good selection of both good and bad infrastructure, from a variety of locations (urban and rural), and whether we want to restrict which images are shown with which, so as to only compare items of a similar type (high street with high street, etc.).

Submitted by PaulJames on 7 September, 2011 - 22:19
Top
11 September, 2011 - 00:30
(Reply to #12)
christhebull

This sounds good. My original idea was to point out that restrictions on cycling bear no relation to the suitability of the route or the risk posed to the cyclist and therefore the bans in pedestrianised areas get ignored while miles and miles can be driven on our wonderful trunk roads without seeing a single cyclist, despite being “all purpose roads” in name, and in name only due to their design and / or traffic volume and speeds. A pictorial comparison of infrastructure could also be done as well.

I certainly don’t think that there should be more restrictions on any type of road, even if it happens to be the A3 Kingston Bypass. But I think it is blatantly obvious that there are certain routes that are effectively off limits already (such as the aforementioned section of the A3), where the right to ride is NOT going to be protected by law if something goes wrong.

Submitted by christhebull on 11 September, 2011 - 00:30
Top
16 July, 2015 - 19:27
(Reply to #13)
Cliff Matthews
Cyclability survey

The idea of polling cyclists and non-cyclists about a range of junctions etc in their area is great. The difficulty is in capturing the dynamic feel of a junction. Some junctions I'd happily let my 7 year old navigate alone at 6:00am whereas at 5:00pm I would be wary.

Capturing the dynamic aspects by sampling streetview is difficult and the point of view is an elevated camera on a car, often (where the google car has only gone one direction) from a viewpoint on the wrong side of the road.

Streetview also fails where a cyclists can go off-road along a cycle path, pedestrian area, footbridge, canal tow-path etc. Some features are seen as good cycling infrastructure by town planners but are lamentable to ride due to poor surface, obstacles, narrowness etc.

My view is that the only way to capture the dynamic cylists eye view is to establish a databank of set format videos provided by cyclists with action-cams. I'm not thinking of the expletive laden near miss postings of some you tube afficionados but simple short sequences of junctions, roundabouts etc with a cyclist executing a safe manoevre in different traffic and weather conditions. These could be easily managed in a national database referenced by standard postcode proximity

 

Submitted by Cliff Matthews on 16 July, 2015 - 19:27
Top
16 July, 2015 - 22:29
(Reply to #14)
Clive Durdle

Isn't it not only physical structure but time and weather? M25 is possibly fine at 4 in the morning :-)

my local streets are good until a boy racer or merc driver turns up!

Submitted by Clive Durdle on 16 July, 2015 - 22:29
Top
16 July, 2015 - 22:30
(Reply to #15)
Clive Durdle

Isn't it not only physical structure but time and weather? M25 is possibly fine at 4 in the morning :-)

my local streets are good until a boy racer or merc driver turns up!

Submitted by Clive Durdle on 16 July, 2015 - 22:30
Top
16 July, 2015 - 22:32
(Reply to #16)
Clive Durdle

Isn't it not only physical structure but time and weather? M25 is possibly fine at 4 in the morning :-)

my local streets are good until a boy racer or merc driver turns up!

Submitted by Clive Durdle on 16 July, 2015 - 22:32
Top
17 July, 2015 - 21:25
pete owens

150 examples of "could cycle" (indeed supposedly designed for that purpose) here:

http://www.warringtoncyclecampaign.co.uk/facility-of-the-month

Submitted by pete owens on 17 July, 2015 - 21:25
Top
Log in to post comments