c3bottles

Please have a look at INSTALL.md on how to install and build c3bottles. All the scripts and configuration needed to run c3bottles in Docker are available as well, please see DOCKER.md for the details. If you want to get involved and improve c3bottles or adapt it to your specific needs, please have a look at DEVELOPMENT.md.

What is this about?

c3bottles is a bottle drop and collection management system for chaos events. Collecting all the empty bottles is a hard task at every event since they appear in large numbers at places and times you cannot always plan beforehand. Bottle collectors need to roam the area all the time and check periodically for any left empty bottles.

To make this job a lot easier, bottle drop points were invented: Empty crates are placed at spots where a high level of empty bottles is expected, e.g. exits of lecture halls, lounges or the main exit/entrance of a building. The participants are advised to leave their bottles not at random but rather only at the designated bottle drop points. The bottle collectors then only need to visit these points to do their job.

However, there are some caveats: In a large building nobody is able to memorize all the drop points and some of them may be forgotten and overflow without anyone noticing. In addition, if the planning in advance was not perfect (as always), some drop points in popular locations may overflow very quickly making the job of the collectors a lot harder. At an overflowing drop point, they can't simply exchange a full crate for an empty one but instead have to put all the bottles laying on and around the more than full crate into their fresh ones. In the worst case they then have lots of full but no empty crates left in locations with a high bottle drop rate. If that ever happens, the bottlecalypse is inevitable.

c3bottles aims to solve this problem using modern web technology combined with crowdsourcing: A database tracks the state of all the drop points and participants noticing full or overflowing drop points can report that fact. The bottle collectors will be notified and can react promptly. They track their tour and the system can notify them of drop points not visited in a long time or needing immediate action. Statistical analysis allows for continuous optimization of drop point placement by scaling drop points according to the demand or abandoning drop points in remote locations which are not used much anyway.

How does this work?

During placement of bottle drop points, every drop point is registered in the database with an identifier and its position. A label to be printed directly from the web interface tells the participants to report full drop points using a URL specific for every drop point. On the website, the participant can report the drop point as needing attention by simply clicking a button. Optionally, they may inform the bottle collectors about the filling level of the drop point.

The other end of the web interface shows a list of drop points to visit ordered by priority: Drop points reported a lot lately get a higher priority. The priority of drop points not visited for a long time grows over time as well, so the collectors will check their state eventually.

A map shows the location of all the drop points so new members of the bottle collector team can find them easily and participants can always find the nearest drop point from their current location.

The bottle collectors track their actions when visiting drop points by clicking a button whenever they empty a drop point. In addition, they can change the location of drop points, remove unused ones or create new ones on the fly as needed.

Evolution of features

The system has been used for bottle collection on 32C3, 33C3 and for trashcans on SHA2017. During 34C3, bottle drop points as well as trashcans needed to be emptied, so trashcans were added ad-hoc. After 34C3, categories were added, so the system is now suited to track different categories of targets, be it bottle drop points, trashcans or other locations that need to be visited more or less frequently based on user feedback.

The map started as custom tiles rendered from a large PNG file and over time has been extended to be compatible with a number of different map sources that can be used with Leaflet, e.g. OpenStreetMap or c3nav. The details regarding different map configurations can be found in MAP.md.