[talk-au] RMS (NSW)'s Bicycle infrastructure map in beta ( http://www.bicycleinfo.nsw.gov.au/maps/cycleways.html )

Sam Russell g.samuelrussell at gmail.com
Mon Aug 20 03:53:13 BST 2012


Ben wrote:

> I had a quick look at it this morning, and it seems a bit inaccurate in
> areas where I know OSM is good. (Ashfield, Marrickville, Burwood)
> Essentially routes marked on the map that are not cycle routes and actual
> cycle routes missing.
>
> Not surprising for the RMS, their data or their support of cycling related
works inside RMS.  Also, as you note below, not surprising for cycling
infrastructure.


> That said, there are probably places where it lists real cycle routes that
> OSM does not have, but possibly it is no more useful than council maps in
> this regard. Either way you need to survey, but proving that something
> isn't a cycle route tends to be more time consuming than proving something
> is.
>
The two things I am interested in mapping are legal designations that don't
appear on other maps—riding out the bounds of "Shared Path" designations
where these are rideable.  I think these designations often provide
necessary bridging infrastructure such as the Shared Path on the south side
of Cleveland street linking Moore Park to the Bourke Street cycleway.  And
secondly, confirming claims by authorities that routes or infrastructures
exist.  I think it could help to clarify some of these things over time,
and then write them up for the wiki.

I do the first because I often feel ashamed to make a logical decision to
use what appears to me to be a footpath, only to discover months later when
wandering that it is in fact designated as a shared path, but that the
signage is located 1200m away behind six corners in the continuous path.

What is a OSM highway=cycleway?  A 2m minimum width path with any
designation excluding motorists and permitting cyclists?  (
http://www.mackay.qld.gov.au/__data/assets/pdf_file/0020/14780/15.08_-_Cycleway_and_Pathway_Design_V2.pdf)
 Cleveland street south would meet that, but shouldn't be marked as
such
because of its sight lines, etc.

What if the primary purpose of a path as narrow as 80cm in relation to
cycling infrastructure is linking dedicated cycleways with heavily
sharrowed "routes."?  (I'm still not sure about what to do with
centre/right of lane bicycle marked roads that make sense to cycle on).

In an ideal world the legal designation, customary use, way surface and
smoothness, way width would all be available.

Two further questions:

I know that in the distant past (before the blue bicycle symbols on
telegraph poles, even) councils tried to claim that bicycle advisory signs
marked on roads indicate a route.  This is fairly ridiculous, and the
modern signage guidelines are better (but not implemented).  It would be
good to get a citation for this in relation to discovering / refuting
declared routes.

The idea of fully compliant routes in Australia is ridiculous.  While this
is a point of contention, verifiability of routes is needed.  I'd put the
level of refutation at a discontinuity of infrastructure and markings, with
no signage, and with no custom or practice amongst cyclists of using the
route.  Wilson St Darlington is a great example of an existing route.  It
is a highly trafficked customary route, with continuous infrastructure, and
a presence on council maps indicating a route.  Without either the
infrastructure OR the high customary traffic, I wouldn't mark it as a route.

Ian wrote,

I'm sure we can do better.
>>
>
I'm sure we can.  RMS's data acts as a spotting guide at best.


>  I was following apmon's remapping idea on the weekend, and for each
>> suburb pair in Sydney produced a travel distance for car routing
>> (using all motor vehicle accessible roads) and a "quiet cycle" routing
>> (using only highway=residential|cycleway, cycleway=*, lcn=*, rcn=*),
>> then sorted the resulting grid in terms of % difference.
>>
>> I'm yet to come to terms with exactly what the numbers mean (if
>> anything :-), but I think there are certainly some pointers to further
>> urban exploration in there.
>>
>
We can use this to get average "directness" values, and look at outliers.
 We can then determine if outliers are due to missing infrastructure
(missing cut throughs, unmapped lcns).  While RMS/Councils have a bad score
for directness in many projects, even they will come towards an "average"
directness value.  (As shockingly indirect as such a routing may be).

thanks—the RMS maps gives plenty of rides to plan,
Sam.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-au/attachments/20120820/852a9640/attachment-0001.html>


More information about the Talk-au mailing list