[OSM-talk] prohibited/restricted/priority and mandatory manoeuvres

[Peter Miller]

Sure, I will copy it to the wiki. But I also think we should aim to get this
settled in the next week or two, or we are going to find people are getting
on with coding using the 'draft' proposal which will make it harder to then
change.

I think it would be good to describe all the tags we can predict we are
going to need at the start, and I also think we should make it as easy as
possible for professional users to migrate to OSM by using the same phrases
for the same purposes (ie not using 'restricted' for a concept where the
convention is to use 'prohibited', especially as the term 'restricted' means
something else). Also there is the practical question that if we use the
term restricted to cover 'prohibited' then what do we use for what the
industry call restricted? I guess we could use 'difficult' or 'inadvisable'
or 'impossible', but none of them quite work.

I would be happy to use Ipswich as the basis for some real comprehensive
coding. I would aim to catch all the car, hgv and cycling restrictions for
at least the core of the town and make it available to anyone doing routers.

As a matter of fact Simon Nuttall has already been testing the Cambridge
Cycle Planner out using Ipswich OSM data over the past month, and he might
be happy to add this functionality to his planner.




Regards,



Peter


> -----Original Message-----
> From: David Earl [mailto:david at frankieandshadow.com]
> Sent: 20 October 2007 16:26
> To: Peter Miller; OSM
> Subject: Re: [OSM-talk] prohibited/restricted/priority and mandatory
> manoeuvres
> 
> It's only a tag name: I really don't think the name matters, but if it
> increases the sum of happiness, feel free.
> 
> May I suggest you add this to the discussion on that page of the wiki,
> so we capture the discussion.
> 
> Dave
> 
> 
> On 20/10/2007 16:10, Peter Miller wrote:
> >
> >
> > So we now have relationships to play with, which is just great. Now we
> > need to decide how to use them before we encode lots of data, and before
> > people start coding journey planners.
> >
> >
> >
> > Currently we have a proposal of 'turn_restriction'.
> >
> > http://wiki.openstreetmap.org/index.php/Relationships
> >
> >
> >
> > IMHO, I don't think the word 'turn' is correct, because we might be
> > talking about something that is straight-on, so the phase 'manoeuvre' is
> > actually the more appropriate Engish term (or actually probably the
> > better French term that has now been appropriated by the English from
> > the spelling). Also, the term restricted is not quite right. See below
> > for how the other standards use the phrase.
> >
> >
> >
> > If one looks at other standards, such as GDF
> > (http://www.ertico.com/en/links/links/gdf_-_geographic_data_files.htm)
> > or the equivalent from the Ordnance Survey
> >
> (http://www.ordnancesurvey.co.uk/oswebsite/products/osmastermap/userguides
> /docs/OSMM_ITN_userguide_v1.0.pdf)
> > one finds that they tease out other important different sorts of
> > manoeuvre. There is also consistent terminology between these standards
> > (and GDF is now an ISO standard).
> >
> >
> >
> > In these documents they identify the following routing limitations.
> >
> >
> >
> > 1) Prohibited_manoeuvre ; one that there is a not legal, but might be
> > physically possible (GDF and ITN).
> >
> > 2) Restricted_manoeuvre ; that that it is not impossible to make (often
> > for identified classes of vehicle) (GDF and ITN)
> >
> > 3) Priority_manoeuvre ; one that does not need to be announced as you
> > are following the same road. (GDF and ITN)
> >
> > 4) mandatory_manoeuvre ; one that you must make. (ITN only)
> >
> > 5) Finally, there is also a special type of prohibited manoeuvre with
> > 'no U turn' for dual carriageways. The actual encoding of the
> > prohibition includes three ways, the one on the initial dual
> > carriageway, the short stub across the carriageway and then the link on
> > the other carriageway coming back again. For this we would need to have
> > three associated ways 'from' 'via' and 'to'. (ITN only)
> >
> >
> >
> > For all of these above types, they can be restricted by time and by
> > vehicle class.
> >
> >
> >
> > I propose that we use the above phases as appropriate in place of
> > 'turn_restriction' and keep the 'roles' as described in the proposal.
> >
> >
> >
> > If anyone is interested, there is a comparison between GDF and OSM here:
> >
> > http://wiki.openstreetmap.org/index.php/GDF
> >
> >
> >
> >
> >
> >
> >
> > Regards,
> >
> >
> >
> >
> >
> >
> >
> > PeterIto
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > ------------------------------------------------------------------------
> >
> > _______________________________________________
> > talk mailing list
> > talk at openstreetmap.org
> > http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk