Microgrants/Microgrants 2020/Proposal/Map Maintenance with StreetComplete

From OpenStreetMap Wiki
Jump to navigation Jump to search
Map maintenance with StreetComplete
Extend StreetComplete with functionality that makes its users actively and continuously keep contributed map data up-to-date
budget (Euro)3000
granteeTobias Zwick
contact(s)• osm@westnordost.de

About your Mapping Community

The community in Hamburg is quite active, there is a lot of activity on Telegram, monthly Stammtische and irregular common activities. Recently, there was an attempt by Hauke Stieler and others to launch a coordinated effort to re-map and re-survey areas that have been determined as "should probably be re-surveyed". He shared a public umap in which everyone could mark, describe (and resolve) places that should be checked again. Another Hamburgian mapper regularly generates routes to postboxes around Hamburg and exports them to his navigation app to check if their collection times are still correct on bicycle tours. These two examples are relevant because they are an examples how proper map maintenance is time-consuming, require coordination and thus an active and alive community and they are an example how local communities struggle to find suitable tools for assisting in efficient and effective map maintenance.

What do you need the grant money for?

The more detailed and complete our map is becoming, the more important does the topic of map maintenance get. The world is changing all the time and thus our map must be updated all the time. The more and the more detailed data we have, the more work it becomes to keep it up-to-date. In the future, map maintenance will consume most of our contributor manpower while at the same time it is an intrinsically less rewarding, often harder and a more time-consuming endeavour than virgin mapping new things and creating and adding data.

In gig-based mapping for example, a lot of useful data is added, but who will maintain it afterwards? For some data, it is not that bad if it is outdated, but for other data, it is! It can be very frustrating to find out that opening hours were wrong or that a clinic does not exist any more. In the end, map data is a liability, one that requires a continued effort.

Keeping map features that are visible on the map (like roads) up-to-date is still relatively straightforward. But not so for properties of map features that are not directly visible on maps, such as surface quality, lanes, opening hours, cycleway infrastructure and quality, tactile paving for the blind or wheelchair accessibility etc. There are also only very few tools that facilitate keeping the map up-to-date, and they are not for beginners. Map maintenance of this kind is currently up to local communities undertaking time-consuming coordinated re-survey efforts or particularly engaged power mappers that care about their particular neighbourhood.

I want to extend StreetComplete with functionality that makes its users aid in the maintenance of map data. And thus, supporting power mappers and local communites in their struggle to maintain the map by both giving them a tool and giving them the support of the 7000+ StreetComplete users (which may otherwise not be that connected with the local community, see my other proposal).

If you receive a grant, what do you aim to achieve?

The idea is simple: StreetComplete should just repeat the questions asked to its users in certain intervals, depending on the expected lifespan of a property of a feature. Implementation should happen in a generic way, so that to enable a question to be asked again in certain intervals does not require a big implementation effort for each question type, but only once.

If there is time left, I would finish the resurvey-opening-hours Pull-Request that Mateusz Konieczny started but did not finish (so far) (#1676)

Budget breakdown

The estimation of the budget (3000€) is based on an estimation how long I need to implement this feature. My ~optimistic estimation is:

  • ~6 days for adapting the overpass queries for the ~35 question types that are eligible for re-survey
  • ~3 days for related UI changes and settings
  • ~6 days for the implementation of a generic approach to asking-the-question-again periodically

So in total, 15 days (optimistically). I'd roughly work half-time on it and could start anytime. Since I do not plan to make a profit, if we assume an hourly rate of 25€, we are at 3000€. I may be willing to work for even less for the benefit of the OSMF, but then we'd need to negotiate that. (Hourly rates for freelancers in Hamburg in my profession roughly range somewhere between 50-90€/hour)

Do you receive funding from any other sources at the moment?

I have profiles on Liberapay, GitHub and Patreon and get monthly donations from some StreetComplete users who want to support my development and maintenance of the app. Currently, it's 60€ per month.

I applied to another grant on which I still await an answer. One part of that application was to implement the feature as described here plus quite a lot of more features related to map maintenance and easy OpenStreetMap survey. These features would go beyond the scope of this grant in terms of effort (not in terms of mission) so that's why I am not proposing them here. I will update this section if I get news from that other application. (Edit: No news yet.) Generally though, the features enumerated for that other grant were quite broadly worded, so it would be possible to adjust the goals for that other grant in case I receive this grant.


By submitting this form to the OpenStreetMap Foundation, you certify the information contained in this application is correct, and that if you are awarded a grant, you will use it only for the purposes described above. You will provide written documentation and receipts for all of your expenses to the OSM Foundation to demonstrate this. You understand that the decisions made by the OSM Foundation Microgrants committee are final.

I, Tobias Zwick (westnordost), agree to follow the above declaration.


Community members are encouraged to endorse your project request here!

  • chris_debian [1]: I fully endorse this project request. I have been using StreetComplete, since the initial release, and have added almost 3000 survey edits. This proposal is a great candidate for maintaining data. I and many others are guilty of updating a feature, then seeing a physical change, but not updating again. Any help to better maintain the OSM data, is a great initiative.
  • Nordpfeil [2]: I fully endorse this project request. I am using StreetComplete since a long time and after around 30.000 change sets in total (mainly with JOSM) it is still a perfect tool for me for underway mapping on a mobile device without brainstorming about the correct key value combination. Furtheron I consider it an excellent tool for newbys, because You do not even see keys and values you just answer questions or make click on a selfexplaning foto o graphic. For that reason it's most valuable to bring new mapper to the communitiy starting with StreetComplete.
  • Tordans [3]: I endorse this project request.
  • gislars [4]: I endorse this project request. It is so easy to map when waiting for a friend or just doing walks. The proposed feature would be a excellent addition. If a user can really maintain (not only) cycle infrastructure that would be a perfect timing right now as there is currently a lot of activity in shifting away from individual transport with cars to other, more sustainable kinds of transportation. Users, not familiar with the whole tagging schema in OSM, can still contribute because StreetComplete is really easy to use.
  • Fnordson [5]: I fully endorse this project request. This is a brilliant idea to use this great app. It's an easy way to update data and since many of us frequent users have already mapped their close environments, it would give us a chance to find new quests nearby also.
  • RubenKelevra [6]: I fully endorse this project request. I think we should really add some focus on validating informations after a while by making it possible to check how long some information has been up there, without tidious work in the history by individual users. Currently there's no editor featuring this (as far as I know), and an editor which is highly geared for easy small tasks is perfect for crowd sourced information gathering on information currency. --RubenKelevra (talk) 01:40, 11 May 2020 (UTC)
  • rugk [7]: I fully endorse this project request. StreetComplete is an awesome app that would be well suited for allowing easy map maintainance. Tobias Zwick has shown in the past to produce a great software and keeping user experience as one of his top priorities, which you can see in the results. --rugk (talk) 15:09, 11 May 2020 (UTC)
  • DenisJu [8]: I endorse this project request. StreetComplete is a good idea to improve the map quality. StreetComplete is in my agenda to use it with other contributors, if my proposal will be selected
  • Tuxayo [8]: I endorse this project request. StreetComplete is a good tool for many maintenance tasks that won't be done otherwise. That's a way that should keep being explored. --Tuxayo (talk) 17:32, 13 May 2020 (UTC)
  • eda_n [10]: I fully endorse this project request StreeComplete has changed my OSM mapping : mapping has a new fun dimension when you can do it while you walk and dicover places. Also teaching OSM with it is much easier. Also, some stuff are just much more easy to map/keep track/locate precisely using notes on StreetComplete while on the field, then deal with the notes when you return on a desktop env. This is a very important app for the OSM world --Eda n (talk) 00:03, 14 May 2020 (UTC)
  • Mapward [11]: I fully endorse this project request. StreetComplete motivated me to more than 7000 contributions and with this app I answered all "quests" in my quarter. Please support further development. It makes mapping much easier. --Mapward (talk) 08:11, 14 May 2020 (UTC)
  • buergi [9]: I fully endorse this project request. Are you pondering what I'm pondering ;) This request is exactly what I thought after using StreetComplete for quite a while now. It currently greatly helps completing the OSM but once an attribute has been set it is out of sight. Ths project request would easily allow verifying attributes like opening hours on a regular basis and would be a tremendous help for maintaining the OSM.
  • kmpoppe [12]: I fully endorse this project request. StreetComplete has brought me back to OSM and it is a wonderful tool to add information to the map with real ease. No knowledge on advanced tools is needed and so it's accessibility to new users is great. --Kai M. Poppe (talk) 07:58, 20 May 2020 (UTC)
  • I endorse this proposal on the condition that it will be translated into different languages, including Russian – Sildream
  • TheBlackMan [13]: I fully endorse this project request. StreetComplete is a very good entry point for new users - I myself encouraged several people to use the app to map in there neighbourhood. Map maintenance becomes more and more important and with the proposed feature users will be able to contribute on a permanent basis. --TheBlackMan 12:35, 20 May 2020 (UTC)
  • I endorse this proposal. – Craig Allan
  • kmpoppe [14]: I fully endorse this project request. This is the one feature I'm missing in StreetComplete. It would enable all users to further increase the accuracy of OSM with ease. --mkotzjan (talk) 11:56, 20 May 2020 (UTC)
  • mrey [15]: I fully endorse this project request. This feature would help a lot with keeping data in OSM updated. --mrey (talk) 13:25, 20 May 2020 (UTC)
  • dbf_ [11]I endorse this project. It is important to keep map up-to-date and StreetComplete is #1 tool for simple mapping on a ground.
  • I fully endorse this project request. Having an up-to-date map is important. We often encounter old data and streetComplete makes it very easy to contribute by answering simple questions. Using a simple tool to get up-to-date information is a good idea! - Amadeous 11:32, 21 May 2020 (UTC)
  • TSR [TSR]: I fully endorse this project request. This feature would help to keep tags up to date.
  • Fringillus [12]: I endorse this project fully. I've used this application quite a lot already and I also have been thinking about the possibility of using the app for this. Excellent idea! - Fringillus (talk) 15:21, 24 May 2020 (UTC)

Next steps

As you probably know, this proposal has been accepted by the OSMF! Yay! Thank you for your support!

Knowing that some people who endorsed this may have this page still on their watchlist, may I direct you to this mailing list discussion?


There, you can contribute and have a say in how some of the details of it are actually implemented. If you are interested in that. To post there, you need to subscribe to the mailing list on this page: https://lists.openstreetmap.org/listinfo/tagging

-- Westnordost (talk)


  1. Interim Report
  2. Final Report