[OSM-talk] Potlatch & duplicated node

[Charlie Echo]

Just a clue:
When the server (or network) is slow, which happens sometimes, or when the "server cannot be reached" (in which case I "retry"), I notice it creates a lot of duplicate nodes. It happens probably when deselecting (i.e. clicking where there is nothing) or moving the map so as to see if the network reacts.

When the "buffer of actions" is cleaned, I suspect the same clicks / nodes are sent many times to the server.


----- Mail Original -----
De: "David Groom" <reviews at pacific-rim.net>
À: talk at openstreetmap.org
Envoyé: Mercredi 27 Août 2008 18:59:37 GMT +01:00 Amsterdam / Berlin / Berne / Rome / Stockholm / Vienne
Objet: [OSM-talk] Potlatch & duplicated node

Whilst reviewing data I've noted recently a large number of duplicated 
nodes.  These occur where one way joins another.

I've managed to create a few myself, but have noted that many other users 
are doing the same.  In each instance the relevant ways have "created by 
Potlatch" as tags.

I presume that :

(1) Either there must be a bug in Potlatch
(2) Its far too easy for a user to inadvertently create duplicated nodes.

Take for example the area 
http://www.openstreetmap.org/index.html?mlat=13.166319744096374&mlon=-61.239985647460315&zoom=12 
.  The validator in JOSM reports 176 duplicated nodes.

Same area  as shown on informationfreeway 
http://www.informationfreeway.org/?lat=13.153544792952308&lon=-61.222993541715226&zoom=15&layers=0000T0B0F

I would stress I'm not blaming the users who created the data in that area. 
I count myself as a fairly experienced OSM mapper, and as I stated earlier I 
have been managing to create some duplicated nodes, and that's even after 
reading and re-reading the Potlatch editing FAQ.

If its not a bug in Potlatch, then could the program be changed so that it 
is much harder to duplicate nodes.

David 



_______________________________________________
talk mailing list
talk at openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk