OpenStreetMap

My OSM process (2021)

Posted by ortho_is_hot on 9 December 2021 in English. Last updated on 25 December 2021.

Well, this is my first diary entry. Here I will outline the tools and process I use when editing on OSM.

Editors

  • I prefer to use JOSM when doing detailed edits, particularly when doing any work with route relations. Enhanced road and lane attributes is a very useful map paint style when lane mapping, and the reverter plugin is useful for handling more complicated reverts (or partial revers)
  • iD is useful for quick or simple edits, and I also find the master branch of iD useful when adding things like powerlines or other objects where every node needs tags (yes this can be done with JOSM however I prefer iD for this task)
  • Vespucci is good when doing in-depth edits on a mobile phone
  • Other editors like Level0 or RevertUI can be used for more simple tasks, such as changing a single tag in the case of the former or reverting a simple changeset with the latter

QA

  • Geofabrik’s OSM Inspector is very useful for picking up problems with routing in particular, which I generally check and fix most obvious errors every couple of days or so; there are also other useful things it can check for
  • Osmose is also good for picking out other errors
  • I keep track of edits using osmcha, using the area boundary tool and a custom bounding box

Other tools

  • hdyc by neis-one is useful for keeping track of my (and other’s) editing statistics, as well as collating changeset comments in a single place
  • I do have a page or two on the OSM wiki, however work has been pretty stagnant for a while

Third-party data

  • Transport related tasks in NSW can use TfNSW’s GTFS data, some of which I have imported (along with others); route relations can be created manually using
  • Mapillary and KartaView are useful when the area you’re editing has imagery available (I prefer to use the web interfaces on another monitor when available, rather than the in-editor tools)

Hardware

  • I have 2x GPS receivers connected to a Raspberry Pi which continuously log the current location every second
  • I have a dashcam which helps with mapping out many features, however reading small writing (and in some cases street signs) can be difficult. I plan on getting a second, higher resolution camera which should help this issue
  • I have a 360 camera which has not seen much use, however is useful for capturing imagery that I’m not going to upload anywhere

(Edit: added hyperlinks)

Discussion

Comment from arukuni on 23 December 2021 at 19:10

Great post! It looks very useful, IMO it would be even cooler if you added links to the tools you are using.

I’m wondering how this 2xGPS + RPi works - is the accuracy comparable to the average modern smartphone? If so, you could maybe describe this setup in detail in the future :)

Comment from ortho_is_hot on 25 December 2021 at 01:50

Good call, will add some hyperlinks. I’m not 100% sure how the accuracy is as my two units generally have some variance however it would be a good topic for me to explore in future.

Comment from MattPoulidor on 22 February 2024 at 16:16

Hey there,

Thanks a lot for sharing that with us!

Concerning your 360pics, you can upload it on: https://panoramax.openstreetmap.fr

Log in to leave a comment