On 26 May 2009, at 18:09, Roger Slevin wrote:
Brian
I would be very interested to have the evidence about the transposed
stops if that is possible. Thanks.
Do you not have access to “bearing” – it’s in the source data, but I
am not sure what you have access to. I can certainly let you have a
file with the bearing data in it, if you do not have access to this
from anywhere else.
I think the bearing must be one of the fields that has been 'culled'
during the import; personally I would consider it essential
information for the next import/update pass. Without that information
will be hard to know if the name/identifier or bearing should be
changed. It is actually necessary to have access to the schedules
themselves to know how the stops are being used.
CUS is a fact of life – nothing I can offer in response to that.
The Stop-Type field would indicate what sort of stop it is (CUS for
customary) if it had been imported. Without this information it is
hard to review the NaPTAN data well and provide hard-hitting feedback
on the data.
I know that as the number of tags for a node increasing that it
becomes difficult for Potlatch to manage, but I think that we are
going to solve that one to get the most out of this data-set. Possibly
someone needs to talk to Richard.
You can of course also check data you are not sure about using the ITO
NaPTAN service that we gave a few people in the area access to for
evaluation purposes during the initial import. That will reflect the
current data rather than the data at the time of the import and will
include all the tags including the ones that have not been imported
into OSM this time.
Regards,
Peter
If stops have been removed – but not removed or replaced in NaPTAN –
then I would again like evidence to take back to colleagues who
maintain NaPTAN. It is possible that they have changed the stop
sub-type from MKD to HAR – that is the preferred method of handling
this situation in NaPTAN. A formal HAR record has an entry point,
what I call an “anchor” point in the middle, and an exit point. The
guidance tells editors that these should all fall on the same named
road (but in practice we know that a lot of them do not adhere to
this rule). What is almost certainly the case in almost every case
is that the length of the HAR section is not the same as the full
length of the road ... most well-created HAR records would have
entry and exit points which are at least a few metres short of the
road junctions at each end.
Best wishes
Roger
From: [email protected] [mailto:[email protected]
] On Behalf Of Brian Prangle
Sent: 26 May 2009 17:19
To: [email protected]
Subject: [Talk-transit] NaPTAN data import in Birmingham update
Having surveyed (and resurveyed) about three hundred bus stops I
think I could draft (soon) a definitive report on how to proceed.
There's also a few points to emerge about the Birmingham data
1. I'm coming across regular transpositon of stops from opposite
sides of the road ( would be useful to have bearing data)
2. Not having the CUS marker on customary stops makes for hard work
surveying and editing on estates where practically every other stop
is CUS
3. Came across two routes (36 and 36C) where whole roads have had
their bus stops physically removed (still showing as NaPTAN data)
and the timetables show the roads as Hail and Ride zones. Couple of
things here - I presume that actual practice has gone beyond the
capacity of NapTAN data to keep pace ( I'll add the details to the
NaPTAN error log on the mappamercia wiki shortly) and how do we
import the HAR data - or represent it ( I can't remember where our
previous discussions got to). In the meantime I'll probably just tag
the roads as HAR=yes and HAR_route_ref= xx. (Without the timetable
there's nothing visible on the ground to indicate to surveyors that
they are in a HAR zone)
Note for Thomas - I know you're into exam season - none of this is
desperate - there's a few thousand more bus stops to survey and
verify yet!
_______________________________________________
Talk-transit mailing list
[email protected]
http://lists.openstreetmap.org/listinfo/talk-transit
_______________________________________________
Talk-transit mailing list
[email protected]
http://lists.openstreetmap.org/listinfo/talk-transit
