[OSRM-talk] Island hopping

Florian Lohoff f at zz.de
Tue Oct 14 08:28:14 UTC 2014


On Tue, Oct 14, 2014 at 02:44:11AM +0200, Zenon Panoussis wrote:
> I've been testing OSRM on Greece and I run into some oddities.
> I use start: ιεράπετρα, end: χαλκίδα and they are understood
> correctly and transformed into something perfectly reasonable.
> But the route goes to Piraeus, then to Kythnos, from there to
> Tziá (Kea) and then finally heads in the right direction through
> Lavrio towards Chalkida:
> http://s2.postimg.org/3t8nqy949/ierapetra_chalkida0.png
> The cause of this can be seen here:
> http://s29.postimg.org/h8ebb5g9j/ierapetra_chalkida1.png
> The routes from Heraklion and Kythnos do not fully connect to
> the harbour (this is an OSM mapping error), but they do connect
> to each-other at the entrance to Piraeus port. The rest is OSRM
> trying to do its job as best it can.
> However, this is also a bug in OSRM: it calculates routes that
> require a mid-sea change of ship. It shouldn't. There needs to
> be some logic that says "if at_sea: ignore_all_intersections"
> to get rid of this behaviour.

How can you detect whether there is an intersection at sea or in
the harbour? It sounds easy to build but i'd rather prefer exactly
these types of errors to see whether OSM data is broken. 

If OSM data is broken expect broken behaviour. 

Dont work around broken data too much - we want to see where
its broken and correct it. Otherwise you put these workaround
into ~1000 Applications working with OSM data instead to fix the
Problem itself.

Florian Lohoff                                                 f at zz.de
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 828 bytes
Desc: Digital signature
URL: <http://lists.openstreetmap.org/pipermail/osrm-talk/attachments/20141014/17af39a7/attachment.sig>

More information about the OSRM-talk mailing list