[OSM-talk] When is a bridge not a bridge?

Chris Hill chillly809 at yahoo.co.uk
Thu Jan 22 16:00:26 GMT 2009

David Earl wrote:
> In view of some changes that I've seen going through in my area 
> recently, I'd be interested to know people's opinion on what constitutes 
> a bridge. In particular to what extent are the approaches part of the 
> bridge? I know I'm likely to hear wildly contradictory answers and 
> there's probably no right answer.
>          ______________________
>         /                      \
>        /                        \
>       /                          \
> ____/                            \______
> Are the ramps part of the bridge? If they are on solid embankments? If 
> they are a lattice structure supported on pillars, perhaps like 
> intermediate supports of the bridge itself? If they have parapets or 
> not? Or do people feel the bit marked as bridge should strictly be only 
> the span (or spans) itself?
A bridge is usually there to cross something.  So I would say, 
generally, what ever was built or built-up or added to make the bridge 
function is part of the bridge.  So ramps or approaches on embankments 
even might well be judged to be part of the bridge.  Maybe a note 
attached to briefly describe your decision will help future OSMers.  As 
always there are exceptions. 

I can think of a bridge that no longer crosses anything (the drain was 
filled in) but it would cost too much to knock it down, I mapped it as a 
bridge because it is a landmark.

In the case of very large structures there are sometimes smaller bridges 
under the approach roads that are worth showing as separate bridges.

On the bridge front, it would be nice to group many ways that cross a 
single bridge onto a single, named object.  Does the proposed bridge 
relation get usefully used yet?  Would a bridge area with ways across it 
for complex bridges make sense?

Cheers, Chris

More information about the talk mailing list