[Talk-GB] NaPTAN Data

Tue Mar 9 12:56:47 UTC 2021

I have no tutorial, but here's my memory of it.   If there are better ways to query changesets I haven't found them.   The mission would be 'return a list of bus stops that have ever been edited by StreetComplete and load to JOSM' - but the 'StreetComplete' tag applies to the changeset, not the bus stop node, so I *think*a multi-step process is needed.
 1. Visit https://osmcha.org/
 2. Hit 'Filters' on the top left.   Now you're building a filter to catch changesets of interest.
   1. Under comment, I entered 'bus stop', since all the relevant Street Complete challenges contain that text in their comment field.   (I've since noticed that there is a changeset tag "StreetComplete:quest_type" to play with too).
   2. Under 'Location' I keyed in 'London', but you can also set a bounding box.  Either way, large changests that cross the area will get picked up, unless you set a 'BBox size bound' (setting "2" will allow changesets up to twice the size of the BBox.)
   3. Under 'Users & Teams' I set my own user name.
   4. Under 'Editor', I keyed 'StreetComplete'
 3. You can save the filter, or just hit Apply (top right).
 4. On the left-hand side you'll now get a scrolling list of matched changesets.
 5. I copied and pasted into a spreadsheet to create and did a bit of jiggery-pokery to get a list of osm URLs in the form - https://www.openstreetmap.org/changeset/94219178   (There are almost certainly better ways to do this part)
 6. You can load these into JOSM one-at-a-time by using the 'Open Location' - ctrl-L option and pasting in the URL. (There are almost certainly better ways to do this)
 7. Now you have a collection of bust stop nodes in JOSM.   You may have to fight with the Conflicts panel if there have been lots of changes to the node since the StreetComplete edit(s) were made.   I always kept the latest node versions in these cases.
 8. If you hit ctrl-H with one or more of the nodes highlighted, you can see the entire edit history, and pick out the date of the latest StreetComplete edit to use as a last_checked date.   But possibly, there are other, later edits which can be attributed to an in-person survey too.
 9. I was able to process about 60 in an hour; I'd be quicker in future since I wouldn't be feeling my way.  The time-consuming parts where - compiling the list of changeset urls and feeding them into JOSM; fighting with the Conflicts panel; getting spooked and checking everything 4 times.   

---
https://hdyc.neis-one.org/?spiregrain
spiregrain_osm at ksglp.org.uk

On Tue, 9 Mar 2021, at 10:16 AM, Jay Turner wrote:
> Hi Ken,
> 
> I like this approach. 
> To you have a tutorial on how to do this? I'm unfamiliar with OSMCha
> 
> Jay Turner
> 
> On Tue, 9 Mar 2021, 09:51 Ken Kilfedder, <spiregrain_osm at ksglp.org.uk> wrote:
>> __
>> Hi Jay and Spike,
>> 
>> After reading this thread it occurred to me that StreetComplete has several questions about bus stops - whether they have a shelter; whether they have a bench, etc.    If there are answers to these questions in the database, that strongly implies that someone has been there, and confirmed the location.
>> 
>> So, as an experiment, I used OSMcha to find all the changesets I ever contributed to StreetComplete bus stop challenges, and set them to naptan:verified=yes with a check_date set to the date of the last StreetComplete-sourced change (which might not have been by me).   I also set a source:naptan:verified for posterity.
>> 
>> Happily, these were all bus stops I have a passing familiarity with.  https://www.openstreetmap.org/changeset/100491768
>> 
>> If you care about naptan:verified tags, this might be an approach to safely armchair-mapping a few.
>> 
>> ---
>> https://hdyc.neis-one.org/?spiregrain
>> spiregrain_osm at ksglp.org.uk
>> 
>> 
>> On Sat, 27 Feb 2021, at 10:52 AM, Jay Turner wrote:
>>> Hi all,
>>> 
>>> There's talk about depreciating naptan: verified in favour for check_date:*=*
>>> 
>>> I'd suggest check_date:naptan:AtcoCode=* or check_date:naptan:NaptanCode=* dependent on what's displayed. (As that varies regionally)
>>> 
>>> The message is here: https://github.com/streetcomplete/StreetComplete/issues/2566#issuecomment-787049474
>>> 
>>> Feedback requested
>>> 
>>> J
>>> 
>>> On Wed, 24 Feb 2021, 22:32 Silent Spike, <silentspike100 at gmail.com> wrote:
>>>> 
>>>> 
>>>> On Mon, Feb 15, 2021 at 8:38 PM Jay Turner <jaynicholasturner at gmail.com> wrote:
>>>>> 
>>>>> I don't really know how to analyse the data of the other 65% but I'll place a bet it's not in a great state. Additionally, 95% of all `naptan:verified` tags are `no`.
>>>>> 
>>>>> FYI, I think the correct tagging for verified is `naptan:verified=yes` - I know some people delete the tag.
>>>> 
>>>> The original import surveying guidelines actually say to delete the tag which is likely why it's 95% of the values:  https://wiki.openstreetmap.org/wiki/NaPTAN/Surveying_and_Merging_NaPTAN_and_OSM_data
>>>> 
>>>> I only know this because I did a deep dive into those wiki pages back when I was planning my more recent Aberdeen import.
>>>> 
>>>> 
>>>> On Mon, Feb 15, 2021 at 8:38 PM Jay Turner <jaynicholasturner at gmail.com> wrote:
>>>>> 
>>>>> *I want to know if it's possible to reimport/mass update NaPTAN data from the latest set of information*. Additionally, I propose a few extra tags;
>>>> 
>>>> It's certainly possible to join existing  naptan:AtcoCode  tagged features to the NaPTAN dataset and apply tags based on that. I originally intended to update the script I wrote for the import to do exactly that, but never got round to it. As for the extra tags, there's an old thread on this list from when I proposed my import where some insightful discussion of the tags to import took place. In the end I imported less than the original import because bloating OSM elements with tags from another data set is not desirable when the identifiers (primary keys) serve as a link. However, I think your proposed tags are reasonable and more generally applicable to OSM.
>>>> _______________________________________________
>>>> Talk-GB mailing list
>>>> Talk-GB at openstreetmap.org
>>>> https://lists.openstreetmap.org/listinfo/talk-gb
>>> _______________________________________________
>>> Talk-GB mailing list
>>> Talk-GB at openstreetmap.org
>>> https://lists.openstreetmap.org/listinfo/talk-gb
>>> 
>> 
>> _______________________________________________
>> Talk-GB mailing list
>> Talk-GB at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk-gb
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-gb/attachments/20210309/46bf62c4/attachment.htm>