[Tagging] Wild changes to wiki pages changing the cycleway tagging scheme

Andrew Davidson theswavu at gmail.com
Sun Mar 17 03:47:26 UTC 2019

On 17/3/19 10:42 am, Martin Koppenhoefer wrote:
> 49 out of 65000. 

Not sure what I am supposed to do with this factoid. Maybe if I try and 
explain the problem in a form that you can't just look up on taginfo:

Let's say we have two keys: key_a and key_b. key_a can have a number of 

value_1: which also implies the presence of an object of type O
value_2: which also implies the absence of an object of type O

and for both values of key_a there is also key_b which can have a number 
of values but let's assume that this is value_3 for both cases. So the 
logical problem is that both values of key_a map to the same value of 
key_b, making it is impossible to definitively state if an object of 
type O is present or not given only the fact that key_b=value_3.

There are some thousand of shared_lane though. I didn’t know this tag, 
historically the cycleway tags were used for bicycle infrastructure, 
seems people are working to change this.

I hadn't realised that shared was so new and radical. However if you 
want to make cycleway:[...]:oneway=-1 the recommended way to map these 
things then you are going to have to have a value for cycleway that will 
work to replace the existing tag cycleway=opposite. I suppose that an 
option would be to recommend that mappers don't use cycleway=opposite 
and just use oneway:bicycle=no instead.

More information about the Tagging mailing list