Skip to content

Latest commit

 

History

History
91 lines (62 loc) · 5.78 KB

README.md

File metadata and controls

91 lines (62 loc) · 5.78 KB

Introduction

Wazimap-NG is the next version of Wazimap. It provides a platform for users to bind tabular data to spatial boundaries in order create curated views of datasets. Yes - that's probably too vague a description to understand what it is. Hopefully the images below provide a better description:

Screen-Shot-2020-09-27-at-09-50-00 Screen-Shot-2020-09-27-at-09-50-33

Screen-Shot-2020-09-27-at-09-50-50 Screen-Shot-2020-09-27-at-09-51-30

See a link to the beta site here: https://beta.youthexplorer.org.za.

You can find the frontend code in this repository. The backend is available here: https://github.com/openupsa/wazimap-ng.

Development

Import Webflow exports using import-webflow-export

New features

The main new features are:

  • Admins now have more flexibility when it comes to loading data. This includes uploading massive datasets and then slicing and dicing in the backend rather than pre-preparing datasets beforehand.
  • Point data is now fully integrated as a first-class spatial object.
  • Choropleths built into the main view. These were hidden behind multiple clicks in the previous version.
  • One platform can host multiple profiles off the same database.
  • The Rich data view allows richer disaggregation of indicators.
  • The administrator can configure the view to use custom basemaps, colours, and other UI settings.
  • Arbitrary spatial boundaries and hierarchies can be loaded onto the same server.
  • Toggling of overlapping boundary layers such as switching between wards and mainplaces which typically cover the same areas.
  • Integration into third-party data sources for realtime data feeds.

Related software

There is no shortage of mapping software available, both commerical and open-source. Wazimap focuses on providing a platform for data custodians to showcase their datasets and mashing them up with public data. The most similar tool that we have found is the excellent GeoNode. We feel that approach to publishing data is significantly different enough to warrant a separate project.

Roadmap

Version 0.8 is due soon and will fix bugs that currently don't have workarounds. We'll publish the 1.0 roadmap soon.

Future features

  • WFS endpoint for publishing data to other GIS software
  • Pluggable data visualisations
  • Better handling of geography hierarchies.
  • Improved handling of temporal and other types of non-census-like data.
  • Speed improvements
  • A large standard database of public datasets.

Installation

yarn install
yarn start

Other commands

yarn build # compile the relevant bundles for production

Configuration

The build is configured using environment variables.

These could be defined in a .env file in the root directory, or on the command line, e.g. for development

HOSTNAME=youthexplorer.org.za yarn start`
Variable Default Description
HOSTNAME unset If set, overrides using the hostname in the address bar to determine the profile being requested. e.g. youthexplorer.org.za
API_URL unset If set, overrides the default production and staging backend URLs. e.g. https://api.wazimap.com
ENVIRONMENT unset E.g. production, staging, sandbox
SENTRY_DSN unset Required in production, staging and sandbox environment
GOOGLE_ANALYTICS_ID unset ID for Google Analytics. Usually only required in production environment

Known Issues

There is an issue with the parcel bundler that sometimes comes up. The easiest way to solve this is to delete the .cache folder.

Documentation

These are works in progress:

Contributions

Contributions are welcome - we are working towards making this process easier.

Shoulders of giants

This project is the next iteration of a number of excellent projects starting with CensusReporter and Wazimap that followed it. Special thanks to William Bird from Media Monitoring Africa whose initial idea (and funding) it was to build a tool to help journalists better understand areas they were reporting on. Also thanks to Chris Berens from VPUU who directed funding to help kickstart this new build. Finally, all of the amazing spatial software and tools developed by one of the most dedicated open source communities out there.