[OSM-talk] Fw: conclusions from my postgres / postgis experiments

[Nick Whitelegg]

Sorry misread your post. and realised you weren't looking for examples. 
I'd have thought that applications which need to treat a single road as an 
entity for any reason would require less processing if a road was treated 
as an entity. There are also labelling issues - e.g. if a superway road 
has one name, you might not want two individual labels close together on 
constituent sub-ways. 

Nick
---------------------- Forwarded by Nick Whitelegg/FT/Southampton 
Institute on 23/11/2006 12:09 ---------------------------




Nick Whitelegg
23/11/2006 12:08


To:     SteveC <steve at asklater.com>
cc:     talk at openstreetmap.org 
Subject:        Re: [OSM-talk] conclusions from my postgres / postgis 
experiments 





>I'm not entirely sure why... I'm imagining a routing algorithim that
>will notice that its on a way with the same name as the one it was just
>driving on and that's it. Are there any use cases for super ways?

I would have thought so, e.g.:

1. If the A1234, as it passes through the town, takes on various names. 
Each individual named bit is the way and the A1234 is the super-way.

2. If the first half mile of Bloggs Street is an unclassified road used 
for through traffic and the remaining quarter mile is a residential 
street. The unclassified and residential bits are the ways, Bloggs Street 
is the superway.

3. Those branched estate roads. Each branched bit could be the way, the 
whole thing the superway.

Nick