Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Peter J Stoner
In message on 22 Jul 2009,  Andy Allan  wrote:

>> Brian Prangle wrote:
>>
>>> Most public transport route maps do show the whole roundabout as part of
>>> the route - perhaps we should follow their example?

> No. Don't put in garbage into openstreetmap just to mimic other
> inferior maps!

The split roundabout is a better portrayal of the public transport.  
Excellent if you are prepared to do that.




-- 
Peter J Stoner
UK Regional Coordinator www.travelinedata.org.uk
Traveline

a trading name of
Intelligent Travel Solutions Ltd  company number 3826797
Drury House, 34-43 Russell Street, LONDON WC2B 5HA
- sign up to "watch this page" at www.travelinedata.org.uk/new.htm


___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Andy Allan
> Brian Prangle wrote:
>
>> Most public transport route maps do show the whole roundabout as part of
>> the route - perhaps we should follow their example?

No. Don't put in garbage into openstreetmap just to mimic other inferior maps!

On Wed, Jul 22, 2009 at 9:32 PM, Lennard wrote:
> If the return trip takes the same roads, eventually the bus will have
> navigated the whole roundabout, and then the whole roundabout *is* part
> of the route.

Blatantly not true for any roundabout that has been modeled with
flared approach roads.

Cheers,
Andy

___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Lennard
Richard Mann wrote:
> I've been merrily splitting up roundabouts. When we have separate 
> relations for each direction of a bus route, we'll even need to split 
> the ones with point junctions.

Perhaps it would be nice to know that in the AND import in The 
Netherlands, almost all roads have been segmented, so they run from 
junction to junction. Also, all roundabouts were modeled as consisting 
of separate segments between each connecting road.

So, even the pros are modeling it in this split way.

-- 
Lennard

___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread David Earl
Lennard wrote:
> Brian Prangle wrote:
> 
>> Most public transport route maps do show the whole roundabout as part of 
>> the route - perhaps we should follow their example?
> 
> If the return trip takes the same roads, eventually the bus will have 
> navigated the whole roundabout, and then the whole roundabout *is* part 
> of the route.

As usual, I have a counter example - the bus station leads off the 
roundabout and comes on to it further on, so there is one section of 
roundabout that none of the 40 buses an hour that use it travel over!

OTOH, some routes go round one-and-a-half times, so should I include 
that section twice in the relation? (No, I'm not being serious).

David


___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Lennard
Jennifer Campbell wrote:

>> If the return trip takes the same roads, eventually the bus will have 
>> navigated the whole roundabout, and then the whole roundabout *is* 
>> part of the route.
>>   
> Not the case, 99% of the time, there will be 2 segments of a roundabout 
> which is unused by a route.

Only if the connecting ways have lane divisions, so when they have been 
mapped as 2 separate ways entering/leaving the roundabout from a certain 
direction.

-- 
Lennard

___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Andy Allan
On Wed, Jul 22, 2009 at 10:33 PM, David Earl wrote:
> Lennard wrote:
>> Brian Prangle wrote:
>>
>>> Most public transport route maps do show the whole roundabout as part of
>>> the route - perhaps we should follow their example?
>>
>> If the return trip takes the same roads, eventually the bus will have
>> navigated the whole roundabout, and then the whole roundabout *is* part
>> of the route.
>
> As usual, I have a counter example - the bus station leads off the
> roundabout and comes on to it further on, so there is one section of
> roundabout that none of the 40 buses an hour that use it travel over!
>
> OTOH, some routes go round one-and-a-half times, so should I include
> that section twice in the relation? (No, I'm not being serious).

Yes. Relations are ordered, so you can (and should) put in the exact
sequence of ways that the route passes over.

Cheers,
Andy

___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Richard Mann
I've not exactly rushing to get to that stage, but I couldn't see any
obvious way to edit the ordering of a relation. Could anyone give me any
clues?

Richard

On Wed, Jul 22, 2009 at 11:06 PM, Andy Allan  wrote:

> Yes. Relations are ordered, so you can (and should) put in the exact
> sequence of ways that the route passes over.
>
> Cheers,
> Andy
>
> ___
> Talk-GB mailing list
> talk...@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-gb
>
___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Jennifer Campbell
Lennard wrote:

> Only if the connecting ways have lane divisions, so when they have 
> been mapped as 2 separate ways entering/leaving the roundabout from a 
> certain direction.

Which is the case for a large majority of roundabouts (not mini 
roundabouts) in the UK. I've been happily splitting roundabouts when I 
have been mapping bus routes. Accuracy is more important IMO.

Jeni



___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Richard Mann
I've been merrily splitting up roundabouts. When we have separate relations
for each direction of a bus route, we'll even need to split the ones with
point junctions.

I leave the junction=roundabout tags (if someone's put them in), so
presumably some zoom levels in some renderings end up with multiple
roundabout symbols overlapping one another. That doesn't misrepresent
anything, it just might look slightly distorted. That's for the renderer to
sort out (if they feel it's important).

Richard
___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Thomas Wood
2009/7/22 Jennifer Campbell :
> Lennard wrote:
>> If the return trip takes the same roads, eventually the bus will have
>> navigated the whole roundabout, and then the whole roundabout *is* part
>> of the route.
>>
> Not the case, 99% of the time, there will be 2 segments of a roundabout
> which is unused by a route.
>
> Jeni

Assuming flared junctions?

-- 
Regards,
Thomas Wood
(Edgemaster)

___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] Naptan import

2009-07-22 Thread Thomas Wood
2009/7/22 Thomas Wood :
> 2009/7/22 Peter Miller :
>>
>> On 20 Jul 2009, at 14:35, Thomas Wood wrote:
>>
>>> 2009/7/20 Peter J Stoner :

 In message on 20 Jul 2009,  Ed Loach wrote:

> I'm assuming that the naptan import when it happens will be as at a
> certain point in time, and won't include any new bus stops since that
> time?

> I'm asking because a bus route has changed in the last week or so that
> now passes my house both ways instead of just one way and rather than
> add bus stops on the other side of the road they've added a taped
> message "Buses stop here and opposite" to each of the existing bus
> stops on the road.

>

 Ed


 If the Transport authority has done its job properly then we will
 expect to see Custom and Practice stops appear in NaPTAN opposite the
>>
>> snip
>>


>>>
>>> The refreshed data is yet to be downloaded, so depending on the
>>> responsiveness of the LA, the stops may be in there by the time I get
>>> around to finalising the import.
>>
>>
>> I am conscious that it is now over 6 months since the data was offered. I do
>> realise that a lot of technical work and familiarisation has been taking
>> place but it would be great to be able to complete the import and move on.
>>
>> I am also aware that there is a 50K place gazetteer sitting there untouched
>> - last week I was adding villages in Norfolk by hand and the data is sitting
>> available in NPTG.
>
> It is, we need to start thinking about what we can do with it.
>
>> Do you need help with the NaPTAN import or are you just about ready to do
>> the work? Do we need to set up a wiki page where people can request imports
>> for their authority or are we going to do it without that?
>>
>
> I've been putting off working on it for a while as slightly more
> interesting projects seem to keep coming my way.
> Anyway, I'm now checking that the new tools that will be used to
> upload the data that have been written for 0.6 will meet our needs.
> For this I'm doing a few uploads to a dev server to see what the
> imported data looks like with regards the created changesets etc.
> I'm probably going to have to modify the uploader to record object ids
> that are being stored for missing references to stop areas.

I have just done a fairly thorough review of both the 0.6 API bulk
upload scripts. Neither works fully as expected.
I have three options, fix the python one, finish the php one, or port
the 0.5 perl one...

The first option is currently looking most tempting.

-- 
Regards,
Thomas Wood
(Edgemaster)

___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


[Talk-transit] Change of settings for talk-transit

2009-07-22 Thread Peter Miller

Just to let you know that I have added Frankie Roberto as an admin for  
the list as per an earlier discussion. Thanks for helping out Frankie.

I have also changed the setting for talk-transit so that replies go to  
the whole list by default which seemed to be what the majority who  
expressed a preference wanted.


Regards,



Peter


___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Jennifer Campbell
Lennard wrote:
> If the return trip takes the same roads, eventually the bus will have 
> navigated the whole roundabout, and then the whole roundabout *is* part 
> of the route.
>   
Not the case, 99% of the time, there will be 2 segments of a roundabout 
which is unused by a route.

Jeni

___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] [Talk-GB] Roundabout, ways and relationship policies

2009-07-22 Thread Lennard
Brian Prangle wrote:

> Most public transport route maps do show the whole roundabout as part of 
> the route - perhaps we should follow their example?

If the return trip takes the same roads, eventually the bus will have 
navigated the whole roundabout, and then the whole roundabout *is* part 
of the route.


-- 
Lennard

___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


[Talk-transit] Roundabout, ways and relationship policies

2009-07-22 Thread Brian Prangle
Hi all

I'm the guilty party who split up the roundabouts in question to put the bus
route relation through it. What do we want correct roundabout junction tags
or correct  bus routes? I did it this way  by splitting the roundabout
because it's obviously a roundabout from it shape and relationships to
adjoining roads  but the bus route doesn't physically use the whole
roundabout and so I felt it incorrect to show the route using bits of road
it doesn't physically pass over - especially on such  large roundabouts
Having said that it's a pain in the butt to keep splitting roundabouts for
bus routes and I plead guilty (again!) to using whole roundabouts elsewhere
on bus routes ( usually much smaller ones than the ones in question.)

Most public transport route maps do show the whole roundabout as part of the
route - perhaps we should follow their example?

The whole thing is inconsistent, including the mappers!  Some guidance on a
wiki somewhere needed perhaps, or a whole better way of adding route
relations. Another issue - some roads can get quite chopped up to cope with
route relations  where several different bus routes turn off and turn onto
the road/down adjoing roadsand then you can lose the sense of the continuity
of the road

Regards

Brian
___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] Naptan import

2009-07-22 Thread Thomas Wood
2009/7/22 Peter Miller :
>
> On 20 Jul 2009, at 14:35, Thomas Wood wrote:
>
>> 2009/7/20 Peter J Stoner :
>>>
>>> In message on 20 Jul 2009,  Ed Loach wrote:
>>>
 I'm assuming that the naptan import when it happens will be as at a
 certain point in time, and won't include any new bus stops since that
 time?
>>>
 I'm asking because a bus route has changed in the last week or so that
 now passes my house both ways instead of just one way and rather than
 add bus stops on the other side of the road they've added a taped
 message "Buses stop here and opposite" to each of the existing bus
 stops on the road.
>>>

>>>
>>> Ed
>>>
>>>
>>> If the Transport authority has done its job properly then we will
>>> expect to see Custom and Practice stops appear in NaPTAN opposite the
>
> snip
>
>>>
>>>
>>
>> The refreshed data is yet to be downloaded, so depending on the
>> responsiveness of the LA, the stops may be in there by the time I get
>> around to finalising the import.
>
>
> I am conscious that it is now over 6 months since the data was offered. I do
> realise that a lot of technical work and familiarisation has been taking
> place but it would be great to be able to complete the import and move on.
>
> I am also aware that there is a 50K place gazetteer sitting there untouched
> - last week I was adding villages in Norfolk by hand and the data is sitting
> available in NPTG.

It is, we need to start thinking about what we can do with it.

> Do you need help with the NaPTAN import or are you just about ready to do
> the work? Do we need to set up a wiki page where people can request imports
> for their authority or are we going to do it without that?
>

I've been putting off working on it for a while as slightly more
interesting projects seem to keep coming my way.
Anyway, I'm now checking that the new tools that will be used to
upload the data that have been written for 0.6 will meet our needs.
For this I'm doing a few uploads to a dev server to see what the
imported data looks like with regards the created changesets etc.
I'm probably going to have to modify the uploader to record object ids
that are being stored for missing references to stop areas.

-- 
Regards,
Thomas Wood
(Edgemaster)

___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit


Re: [Talk-transit] Naptan import

2009-07-22 Thread Peter Miller

On 20 Jul 2009, at 14:35, Thomas Wood wrote:

> 2009/7/20 Peter J Stoner :
>> In message on 20 Jul 2009,  Ed Loach wrote:
>>
>>> I'm assuming that the naptan import when it happens will be as at a
>>> certain point in time, and won't include any new bus stops since  
>>> that
>>> time?
>>
>>> I'm asking because a bus route has changed in the last week or so  
>>> that
>>> now passes my house both ways instead of just one way and rather  
>>> than
>>> add bus stops on the other side of the road they've added a taped
>>> message "Buses stop here and opposite" to each of the existing bus
>>> stops on the road.
>>
>>>
>>
>> Ed
>>
>>
>> If the Transport authority has done its job properly then we will
>> expect to see Custom and Practice stops appear in NaPTAN opposite the

snip

>>
>>
>
> The refreshed data is yet to be downloaded, so depending on the
> responsiveness of the LA, the stops may be in there by the time I get
> around to finalising the import.


I am conscious that it is now over 6 months since the data was  
offered. I do realise that a lot of technical work and familiarisation  
has been taking place but it would be great to be able to complete the  
import and move on.

I am also aware that there is a 50K place gazetteer sitting there  
untouched - last week I was adding villages in Norfolk by hand and the  
data is sitting available in NPTG.

Do you need help with the NaPTAN import or are you just about ready to  
do the work? Do we need to set up a wiki page where people can request  
imports for their authority or are we going to do it without that?




Regards,



Peter



>
> -- 
> Regards,
> Thomas Wood
> (Edgemaster)
>
> ___
> Talk-transit mailing list
> Talk-transit@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-transit


___
Talk-transit mailing list
Talk-transit@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-transit