SOTM 2020 notes #2

Posted by Øukasz on 7 July 2020 in English.

There might have been a misunderstanding…

A very useful talk, interesting especially to see how OSM is seen from the perspective of people who perhaps have the most authority on what actually goes in the map - the DWG. However, I am wondering about the format and channel of delivering this message. I’m not sure how many people have seen the talk, but I expect the sample to be self-selected from those who are already relatively deeply engaged with the OSM, and, consequently, they already know a lot of what was said here. A lot of this content is available through the wiki, but I’m wondering if it shouldn’t be made more prominent on OSM landing page. Or perhaps the ID editor could be augmented so that the walkthrough that is offered to the new mappers does not only focus on how to edit, but also what to edit?

An Incomplete History of Companies and Professionals in OpenStreetMap

A solid opening, and a lot of informative background. This talk makes me think it would indeed be useful to write up - perhaps as someone in the Q&A suggested, in a collaborative way (or at least with more than one perspective represented). Worth to watch together with the “Curious Cases of Corporations in OpenStreetMap 🎓” talk from Day 2.

An interesting talk, with humanitarian principles such as do no harm and informed consent at the centre. For me personally, it highlights the tension that is created when a company or an organisation is contracted to do some OSM work, and then, for reasons beyond their control, this proves to be challenging.

For example, a humanitarian mapping project may wish to create a tag, or an interrelated tagging system, for a specific aspect of humanitarian operations. They may have things well thought out based on the examples in front of them, but the OSM community could be resistant to the new tag because while it would work well in Africa it would not be applicable to Asia, or because the tag is too specific and only really suitable for this one particular project’s requirements. Now the organisation is in an uncomfortable position where they have to deliver on a contract, but the OSM community is rejecting their contribution.

Aa another example, another humanitarian mapping project may wish to upload a sizeable chunk of refugee camp facilities data into OSM. Even assuming there are no licensing issues, the organisation may or may not be the author of the data, the person doing the upload may or may not be the one who collected the data in the field, finally, the data may or may not fall under the Imports or Organised Editing categories. Again, the project is contractually obligated to upload the data, but the community may reject it.

Both of these examples point to the need for careful consideration of how to engage with the OSM community and understand procedures before undertaking any work contract.


Comment from anum shireen on 7 July 2020 at 14:53

this is very exciting website for mapping and all….. this is really helpfyl.. anyhow if you guys want to read exciting novvels then must visit………………….

Comment from ABZ_OSM on 8 July 2020 at 22:52

I’d be interested in your thoughts on mapping sensitive sites.

As you outline, the OSM community can reject contracted work, becasue for example, the sensitivity of the meaning of a tag / key in Africa vs it’s meaning in Asia.

Sensitivity can also be about strategic sites.

What do we do when we visit a location and unbeknown to us, a large site we are welcome at houses within it a large sensitive site the locals don’t want to talk about.

On such an occasion a fairly innocent question is “Hi, what is that right in front of us?”

It turned out to be, as far as I could gather, a chemicals plant, regarded as strategic, in a fairly unstable region.

Innocent questions as such seem to cause suspicion, though it is not aparent why, when your supposed to be welcome standing next to it !!

I’m fairly sure some of my edits, on that occasion, outlining that plant, were deleted, though how locals would connect “Hi, what is that over there” to OSM I do not know. Perhaps I innocently mentioned OSM at the time.

Comment from Øukasz on 9 July 2020 at 15:14


I think there are a lot of issues to unpack in here - the OSM community who want to see the data fit into their (possibly subjective) interpretation of what the rules and standards are, the local community living next to a strategically or nationally sensitive facility, the international aid community using OSM as their data platform… all of these are complex in itself.

As for the example you mention - it could be the locals who deleted the plant, or it could be anyone else in the country - or outside of it - that knew of its sensitive status and just checks in every now and then if it didn’t get mapped by someone. In some countries mapping military installations at all is completely prohibited, in others even using a map that has military installations on it is illegal.

I’m not sure if there is a specific question in your message that you’d like me to answer?

Comment from ABZ_OSM on 9 July 2020 at 16:05

Hi Øukasz,

thank you for your response.

Quote: “I think there are a lot of issues to unpack in here”. Yes.

The particular question I put was about a particular installation, however I see there are a range of possible answers and interpretations about what constitutes a sensitive location, and there are many possibilities for sensitive site types.

eg a refuge for women from abuse, as was already pointed out in this post. Clearly such a location is sensitive and only a person intended to be there and relevant Social services should know the location. Not somewhere to be mapped.

There is always more to learn.

This chemical plant in question was a hush hush topic locally, even though I had to walk all the way around it within the site, where I was welcome.

Quote: “As for the example you mention - it could be the locals who deleted the plant, or it could be anyone else in the country - or outside of it - that knew of its sensitive status and just checks in every now and then if it didn’t get mapped by someone. “

I think it was something like that. Though I can’t remember how to use to find out if I ever mapped it in the first place.

I’m happy to leave the deletion, if it was indeed deleted, alone; I understand that what is a good mapping opportunity to me may constitute an threat to others.

This should be respected.

In any case antagonising anyone over this seems poinless, but exploring the ideas around what makes something a sensitive issue is always worth it (If you have enough conversational energy left and it’s not to far towards the end of the day :-) ).

All insights on sensitive sights are welcome :)

Comment from Øukasz on 13 July 2020 at 15:09

ABZ_OSM, if you want to trace a deleted issue, the easiest is probably to use to find the feature matching the tag you’re after (so I guess the tag of the chemical plant), and run the query to show the state of the planet at a specific point in time (so ideally right after you mapped it). It should then return a node/way ID that you can use to audit its history, even though it does not appear in the live OSM database anymore.

For example, consider this query showing a refugee camp and compare to this one, which shows what the OSM looked like 2 minutes before. I accidentally created a duplicate that I removed later, but the second query shows both points. You can click the node, and then in the popup click the node ID (so in this case 7676579280) and then select “view history” to see what has happened to this feature in the past even though it is now deleted.

Hope this helps

Comment from ABZ_OSM on 15 July 2020 at 10:51

Hi Øukasz ,

It does help thank you.

Thank you also for your response, which includes nice overpass queries, which is very informative.

I am quite familiar with overpass and have used it extensively, but am certianly no expert, and often struggle with putting queries together, particularly the complex queries.

I can indeed see the remaining and the deleted node, at the camp, by time.

I had, during my searches, some days earliear, found and used a very similar, if not exactly the same query, for isolating changes to the chemical plant. Your query shown below :

out geom;

Specifically I used the

[date:”2020-07-02T09:59:00Z”] \


I feel that something looking like this

[date_between::”2020-07-02T09:50:00Z”,”2020-07-02T09:59:00Z”] \

would be much more useful, for such a query, simply to know the query is isolated in a time frame. Certianly SQL does allow a query between two time stamps.

I searched for such a possibility but found nothing in my searches.

Also, I do not understand the difference between



out geom;

why would you choose one and not the other?

Comment from Øukasz on 16 July 2020 at 08:24

Thanks for the challenge!

For your first point, you are probably looking for the diff setting. Using the same example.

As for the difference between out; and out geom; - one will return more concise data - so for example just the OSM way element with reference of the nodes it is built from - while the other will return complete geometry of all the elements. To see this, you have to switch to the Data tab on the right-hand side of the OT interface.

Comment from ABZ_OSM on 16 July 2020 at 09:53

Hi Øukasz,

Thanks for the useful query. I used it as such o the industrial site:

out geom;

and no out line nodes for the ways I might have traced appreared.

So I assume I never actually put it in.

This is a very useful query as I have not re-familiarised myself with osmcha, which does not seem straightforward, at all to use. It probarly is easy to use, and it’s a simple case of the initiated never see the reason it is not easy to use. Could’nt quickly find a simple short usage video on youtube so had to apss on that.

This diff command is going to be a really useful way of checking for local map vandalism, though usually threse are just unintended errors.

Since I mapped 99.99% of the local town, and surrounding vilages, do you know if I can look for all changes not by me. I tried this query, which works, but when I tried to put the ! , to not it, in all the possible places, it did not work

out geom;

Thanks for the expanation on out; vs out geom;

I have used the data tab quite extensivley in the past for collecting counts and other street level information. eg outputting addresses of local schools as a list

Log in to leave a comment