In Perths western suburbs there are quite a few 13~14 year old cycling routes created by the user “Richard Stallard” that aren’t really routes at all. The “X to Y” routes (e.g. NDJM Nedlands to Jolimont, DKCL Dalkeith to Claremont etc) are not signposted or appear on any “official” cycling route maps, nor do they use any dedicated cycling infrastructure for the majority of their route; with some of them mentioning being “user generated” routes—whatever that means—when they were renamed to distinguish them from “official” routes.
I’m thinking of deleting them, but I want to know if that’s in-line with how the Australian mapping community (or even more specifically, the Perth mapping community) handling cycling route relations.
I wonder if these are the remnants of the cycle routes that were deleted en mass a few years ago? There was some discussion about restoring them.
The NDJM Nedlands to Jolimont relation started life as NW203 Nedlands to Subiaco, which was an officially mapped route. I don’t know if the old blue signs still exist anywhere along that route; if they do, it should stay in OSM.
The government’s long-term cycle network plans don’t appear to use the same route (e.g. Smyth Road rather than the cycle path on the east side of the primary school).
NW203 doesn’t seem to exist anymore, there is no signage or acknowledgement of its existence at least on any official map since 2016. I also couldn’t actually find any official maps containing it, but finding search results from 13 years ago is hard .
Not that I really trust the official maps either, for example the official bike map for Perth, Fremantle and Stirling claims that the continuously signed route “NW22” runs down Waratah Avenue, but there is nothing there at all, not even sharrows, and the Long term cycle network flags the avenue’s biking experience as “Non-existing (inadequate)” and its bike infrastructure as “No infrastructure”, unless you looked at perths cycle network pdf you’d have no idea that NW22 even exists.
I’ll also note that the Long term cycle network isn’t a map of cycle routes but just cycling infrastructure in general, maybe this is not the consensus but from what I’ve read a cycle route needs to be something that at least has some form of signage (even if its terrible sharrows) telling you where to go, if these routes were mostly just straight lines with some sharrows down their length I’d probably leave them be, but often they make turns onto new streets or across intersections with no signage whatsoever.
Personally I am a fan of mapping whats on the ground, not what the government claims there to be even if these routes at some point contained signage as Richard claims in that thread you posted, they don’t anymore or might’ve never had in the first place, and offer no indication that they are intended for cycling whatsoever.
Yep, I totally agree that they should be deleted where there’s no signage or indications on the ground of how to follow these routes.
Good point about the LTCN not being about actual end-to-end routes. That’s probably a good idea really, on their part, given the general dismal state of the last couple of decades’ efforts in maintaining signposted routes.
The routes on OCM with 4-letter names are ones that I mapped a few years ago based on my local knowledge.
I believe the NDJM still represents the best bike route on low-traffic streets.
I strongly believe that routes on the DoT long term cycle network should not be mapped until necessary improvements are implemented, which may take many years. Smyth Rd is, in my view, certainly not bicycle-friendly in its current state.
In general, I believe the OCM route should represent viable cycling routes that are useable now, rather than future aspirations of DoT.
I have it on good authority that at one stage in recent years, none of the staff of the DoT cycling team actually rode bikes, and new people were brought into the group on the basis of playing in a basketball team rather than any relevant experience in planning or even using bicycle routes. That may well be part of the reason for some of their recent failures such as the Nedlands so-called Safe Active Street which leads cyclists up the steepest hill in the area, and totally ignored pre-existing well used routes connecting to UWA and Perth CBD, as well as having severe speed bumps twice as steep as DoT’s own guidelines.
Richard Stallard
Perth and Denmark, Western Australia
Hello Richard, thank you for taking the time to reply.
My issue is that these routes aren’t meeting the criteria for being cycle routes per both the OSM wiki’s cycle routes page and the Australian tagging guidelines for cycling and foot paths which states the route needs physical signage of some form, they may very well be nicer roads and paths to cycle down compared to other roads but that in itself doesn’t make them cycle routes. I don’t think it would be right of me for example to—say, add a walking route relation down a couple sidewalks because I think its the best way to get from Cottesloe Train Station to Cottesloe Beach, from all I’ve seen mapping what’s on the ground is generally best practice instead of mapping what a government or institution claims there to be, and these routes have neither government recognition or infrastructure the ground at all.
I’d also note that this is why I’m also against adding “official” “routes” published by the government like NW22 down Waratah Avenue, as that also doesn’t have any physical indication of its existence like signage; heck, I’d even leave these routes alone if the opposite were true and some Good Samaritan put up unofficial cycling signs for these routes (provided the councils don’t take them down that is).
(also I want to reiterate that the DoT’s Long Term Cycle Network is a map of proposed infrastructure, not routes, and is completely irrelevant to this discussion.)
Basically, if the map only shows signed routes, there would be hardly anything to show.
Even the Nedlands Safe Active Street project, for instance, which cost something like $3 million for a few km, has no signs to say where it leads, therefore, it should not be on the map.
My personal view is that, due to the almost total lack of signed bicycle routes around Perth suburbs, other than some PSP’s, a map of useful routes is actually very important. If routes are comprehensively signed, you hardly need a map.
Anyway, I am pretty much done with contributing to OCM in the Perth area since the bulk deletion of previously mapped routes.
I don’t hold out much hope that DoT and local governments will get their act together any time soon to develop a comprehensive network of signed bike routes through the suburbs. Such a network was planned in the late 1990’s, but it was never fully implemented and signage on the routes which were completed has not been maintained, despite commitments from local governments to do so. I know because I was there.
Richard Stallard
Perth and Denmark, Western Australia
Well yes, but that’s the rub, the map currently makes it out to seem like there are more cycling routes then there actually are. There would be hardly anything to show because unfortunately there is hardly anything to show, at least until perth expands its long term cycle network.
Absolutely not, we still map road routes and the like because its still useful for getting a birds eye view as well as helping you get to where you actually need to go once you’ve gotten off the route, not to mention it helps with navigational software; I’m sure if you’ve ever used a satnav you’ve heard something like “Continue on State Route 51” or “Turn onto State Route 13” and the like, this generally assumes clear signage around how to continue or get on/off the signed route which these cycle routes do not have, if someone was using routing software and was told to “Cycle onto NDJM Nedlands to Jolimont” there would be no signage to actually tell them what the routing software is talking about (which would also be another reason to not add “official” routes that don’t have signage either).
I’m going to go ahead with removing a some of the routes in the western suburbs, currently:
NDJM Nedlands to Jolimont
DKCL Dalkeith to Claremont
NDDK Nedlands to Dalkeith
NDCL Nedlands to Dalkeith min. grade
UWSU UWA to Subiaco
UWCL UWA to Claremont
Bicycle routes in OpenStreetMap are meant to represent actual routes which exist on the ground, either through direct route signage with a route name or code, or implied through arrows or other markings which connect roads into a route.
It’s not for our opinion on what we think would make a good cycling route, everyone will have a different opinion and would go against the concept of Verifiability - OpenStreetMap Wiki.
There is a place for people to publish their own suggested cycling routes, but OpenStreetMap isn’t that place.
The typical government maps show all bicycle infrastructure as a “network” but type=route + route=bicycle isn’t for networks, it’s for routes.
Gosh. Exactly. If we really want to map the “Bikedirect Network” (I don’t have the local knowledge to have a view on if we should or shouldn’t), then use something like network=AU:WA:Bikedirect per Key:network - OpenStreetMap Wiki on each of the ways.
I believe they’re just roads considered “better” for cyclists. If they were to remain I’d definitely want to break them up either by those maps, some kind of already given grouping, or by LGA, and then make a super-relation to join them all together.
I’ll probably do some more research and make a separate thread about this specific case.
At least some of the BikeDirect “secondary road” routes are physically signed, small rectangles with a bike symbol on a blue background, or a blue arrowhead on a white background.
Main roads are never signed as part of the network in the same way. I don’t think they are (or at least weren’t originally intended to be) included in the “network”, and only come with the dataset for the local road routes because they indicate which roads are main roads on the “Cycle Instead” maps and whether or not they have bike lanes.
.
.