Re: [OSM-talk] Automated Populate/Update Problem

[Warin]

G'day,

Yes to contacting the Ozies on au-talk.

In general I'd be careful on changing the 'old' data as some of it maybe 
better than the departments data, needs a judgement call. I'd think 
adding to it is fine - reference numbers that are missing, that kind of 
thing.


Adding new stops should be fine - assuming that the data is reasonable 
location and quality wise.



Best .. talk to the locals, enter a few examples, say 20, of the new 
data and state some of the conflicting stops differences between OSM and 
the departments data - say 10? This would give a small sample 
demonstration of what is intended and is easier to correct than 27,000 
of them.


On 28/9/22 20:38, Andrew Harvey wrote:
Yep you'll reach Victorian and Australian mappers better on talk-au as 
some might not join the global talk list -> 
https://lists.openstreetmap.org/listinfo/talk-au.


I'll echo other's comments here, if you are planning or have done the 
conflation I'd suggest sharing those results so the community can 
review and share feedback.


On Wed, 28 Sept 2022 at 19:16, Phil Wyatt  wrote:

Hi Rob,

Given you are in Australia I would try the talk AU list as well.
Maybe also the discord channels as there are a few Ozzie folks
there in the Oceania channel with lots of transport experience.

Cheers - Phil
(On the phone so apologies for any typos)


On 28 Sep 2022, at 6:36 pm, rob potter  wrote:


Thanks for your reply.

I have read the guidelines.

I'm in Victoria, Australia

Rob

On Wed, 28 Sept 2022, 18:07 Eugene Alvin Villar,
 wrote:

Hi,

I work for the state transport department


Sorry if I missed this somewhere, but which state and which
country? Depending on the answer, there might be a local
community that can help and provide guidance as well with the
conflation/import process.

Thanks,
Eugene


On Wed, Sep 28, 2022 at 3:24 PM rob potter
 wrote:

Hi,

I work for the state transport department and we are
looking to become an active member of the community and
as a first dataset we have focused on is our public
transport stops, bus and tram initially and then stations.

I would like your advice on how to achieve the outcome.

There are a number of considerations:

  o Currently in the state there are ~9,100
highway:bus_stop
  o our GTFS - stops.txt has ~27,000 stops
  o the current accuracy of highway:bus_stop needs
review.
  o stops.txt location appears to be of a much better
quality

My initial thought was extract current, match data
location, enrich what stops.txt has then create all new
and remove existing as final step.

I would guess there are people screaming NO!! if so,
please advise of a viable way of making such a significant

Regards,

Rob
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk


___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Automated Populate/Update Problem

[Andrew Harvey]

Yep you'll reach Victorian and Australian mappers better on talk-au as some
might not join the global talk list ->
https://lists.openstreetmap.org/listinfo/talk-au.

I'll echo other's comments here, if you are planning or have done the
conflation I'd suggest sharing those results so the community can review
and share feedback.

On Wed, 28 Sept 2022 at 19:16, Phil Wyatt  wrote:

> Hi Rob,
>
> Given you are in Australia I would try the talk AU list as well. Maybe
> also the discord channels as there are a few Ozzie folks there in the
> Oceania channel with lots of transport experience.
>
> Cheers - Phil
> (On the phone so apologies for any typos)
>
> On 28 Sep 2022, at 6:36 pm, rob potter  wrote:
>
> 
> Thanks for your reply.
>
> I have read the guidelines.
>
> I'm in Victoria, Australia
>
> Rob
>
> On Wed, 28 Sept 2022, 18:07 Eugene Alvin Villar,  wrote:
>
>> Hi,
>>
>> I work for the state transport department
>>>
>>
>> Sorry if I missed this somewhere, but which state and which country?
>> Depending on the answer, there might be a local community that can help and
>> provide guidance as well with the conflation/import process.
>>
>> Thanks,
>> Eugene
>>
>>
>> On Wed, Sep 28, 2022 at 3:24 PM rob potter  wrote:
>>
>>> Hi,
>>>
>>> I work for the state transport department and we are looking to become
>>> an active member of the community and as a first dataset we have focused on
>>> is our public transport stops, bus and tram initially and then stations.
>>>
>>> I would like your advice on how to achieve the outcome.
>>>
>>> There are a number of considerations:
>>>
>>>
>>>- Currently in the state there are ~9,100 highway:bus_stop
>>>   - our GTFS - stops.txt has ~27,000 stops
>>>   - the current accuracy of highway:bus_stop needs review.
>>>   - stops.txt location appears to be of a much better quality
>>>
>>> My initial thought was extract current, match data location, enrich what
>>> stops.txt has then create all new and remove existing as final step.
>>>
>>> I would guess there are people screaming NO!! if so, please advise
>>> of a viable way of making such a significant
>>>
>>> Regards,
>>>
>>> Rob
>>> ___
>>> talk mailing list
>>> talk@openstreetmap.org
>>> https://lists.openstreetmap.org/listinfo/talk
>>>
>> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Automated Populate/Update Problem

[Phil Wyatt]

Hi Rob,Given you are in Australia I would try the talk AU list as well. Maybe also the discord channels as there are a few Ozzie folks there in the Oceania channel with lots of transport experience. Cheers - Phil(On the phone so apologies for any typos)On 28 Sep 2022, at 6:36 pm, rob potter  wrote:Thanks for your reply.I have read the guidelines.I'm in Victoria, AustraliaRobOn Wed, 28 Sept 2022, 18:07 Eugene Alvin Villar,  wrote:Hi,I work for the state transport departmentSorry if I missed this somewhere, but which state and which country? Depending on the answer, there might be a local community that can help and provide guidance as well with the conflation/import process.Thanks,Eugene On Wed, Sep 28, 2022 at 3:24 PM rob potter  wrote:Hi,I work for the state transport department and we are looking to become an active member of the community and as a first dataset we have focused on is our public transport stops, bus and tram initially and then stations.I would like your advice on how to achieve the outcome.There are a number of considerations:Currently in the state there are ~9,100 highway:bus_stopour GTFS - stops.txt has ~27,000 stopsthe current accuracy of highway:bus_stop needs review.stops.txt location appears to be of a much better qualityMy initial thought was extract current, match data location, enrich what stops.txt has then create all new and remove existing as final step.I would guess there are people screaming NO!! if so, please advise of a viable way of making such a significant Regards,Rob
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk


___talk mailing listtalk@openstreetmap.orghttps://lists.openstreetmap.org/listinfo/talk___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Automated Populate/Update Problem

[rob potter]

Thanks for your reply.

I have read the guidelines.

I'm in Victoria, Australia

Rob

On Wed, 28 Sept 2022, 18:07 Eugene Alvin Villar,  wrote:

> Hi,
>
> I work for the state transport department
>>
>
> Sorry if I missed this somewhere, but which state and which country?
> Depending on the answer, there might be a local community that can help and
> provide guidance as well with the conflation/import process.
>
> Thanks,
> Eugene
>
>
> On Wed, Sep 28, 2022 at 3:24 PM rob potter  wrote:
>
>> Hi,
>>
>> I work for the state transport department and we are looking to become an
>> active member of the community and as a first dataset we have focused on is
>> our public transport stops, bus and tram initially and then stations.
>>
>> I would like your advice on how to achieve the outcome.
>>
>> There are a number of considerations:
>>
>>
>>- Currently in the state there are ~9,100 highway:bus_stop
>>   - our GTFS - stops.txt has ~27,000 stops
>>   - the current accuracy of highway:bus_stop needs review.
>>   - stops.txt location appears to be of a much better quality
>>
>> My initial thought was extract current, match data location, enrich what
>> stops.txt has then create all new and remove existing as final step.
>>
>> I would guess there are people screaming NO!! if so, please advise of
>> a viable way of making such a significant
>>
>> Regards,
>>
>> Rob
>> ___
>> talk mailing list
>> talk@openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk
>>
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Automated Populate/Update Problem

[Eugene Alvin Villar]

Hi,

I work for the state transport department
>

Sorry if I missed this somewhere, but which state and which country?
Depending on the answer, there might be a local community that can help and
provide guidance as well with the conflation/import process.

Thanks,
Eugene


On Wed, Sep 28, 2022 at 3:24 PM rob potter  wrote:

> Hi,
>
> I work for the state transport department and we are looking to become an
> active member of the community and as a first dataset we have focused on is
> our public transport stops, bus and tram initially and then stations.
>
> I would like your advice on how to achieve the outcome.
>
> There are a number of considerations:
>
>
>- Currently in the state there are ~9,100 highway:bus_stop
>   - our GTFS - stops.txt has ~27,000 stops
>   - the current accuracy of highway:bus_stop needs review.
>   - stops.txt location appears to be of a much better quality
>
> My initial thought was extract current, match data location, enrich what
> stops.txt has then create all new and remove existing as final step.
>
> I would guess there are people screaming NO!! if so, please advise of
> a viable way of making such a significant
>
> Regards,
>
> Rob
> ___
> talk mailing list
> talk@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Automated Populate/Update Problem

[Marc_marc]

Hello,

Le 28.09.22 à 09:18, rob potter a écrit :
My initial thought was extract current, match data location, enrich what 
stops.txt has then create all new and remove existing as final step.


no :) keep the historic, no need to delete-create again.

a better workflow :
extract current
match data location
enrich what stops.txt has
read import guidelinee and follow it :)
modify existing stop and create missing one
delete a stop only if it doesn't exist anymore, not to change
the location nor a tag
also take care that some operator (may be you, I don't known)
have 1 stop where osm have 2 stop, one at eatch side of the road

Regards,
Marc



___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Automated Populate/Update Problem

[Martin Koppenhoefer]

Welcome to osm! Regarding the "remove existing as final step": we generally
expect people to conflate new data with existing data, so you should not
remove the existing ones, but enrich them (including position acuracy if
possible) with your data, and only create new objects for things that
weren't mapped before.

If you haven't yet, read the guidance for automated edits
https://wiki.openstreetmap.org/wiki/Automated_edits and follow the import
guidelines https://wiki.openstreetmap.org/wiki/Import/Guidelines

Cheers,
Martin
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

[OSM-talk] Automated Populate/Update Problem

[rob potter]

Hi,

I work for the state transport department and we are looking to become an
active member of the community and as a first dataset we have focused on is
our public transport stops, bus and tram initially and then stations.

I would like your advice on how to achieve the outcome.

There are a number of considerations:


   - Currently in the state there are ~9,100 highway:bus_stop
  - our GTFS - stops.txt has ~27,000 stops
  - the current accuracy of highway:bus_stop needs review.
  - stops.txt location appears to be of a much better quality

My initial thought was extract current, match data location, enrich what
stops.txt has then create all new and remove existing as final step.

I would guess there are people screaming NO!! if so, please advise of a
viable way of making such a significant

Regards,

Rob
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk