The Databox platform is an open-source personal networked device, augmented by cloud-hosted services, that collates, curates, and mediates access to an individual’s personal data by verified and audited third-party applications and services. The Databox will form the heart of an individual’s personal data processing ecosystem, providing a platform for managing secure access to data and enabling authorised third parties to provide the owner with authenticated services, including services that may be accessed while roaming outside the home environment. Databox project is led by Dr Hamed Haddadi (Imperial College) in collaboration with Dr Richard Mortier (University of Cambridge) and Professors Derek McAuley, Tom Rodden, Chris Greenhalgh, and Andy Crabtree (University of Nottingham) and funded by EPSRC. See http://www.databoxproject.uk/ for more information.
These instructions will get a copy of the Databox up and running on your local machine. For development and testing purposes, see Development section below.
- Requires Docker. Read here for docker installation.
Note: currently supported platforms are Linux and MacOS. Running on other platforms is possible using a virtual machine running Linux with bridge mode networking. Also note that more than one CPU core must be allocated to the VM.
Make sure Docker is installed and running before starting Databox. Run the following to get your databox up and running.
mkdir databox
cd databox
docker run --rm -v /var/run/docker.sock:/var/run/docker.sock -v -t databoxsystems/databox:0.5.1 /databox start -sslHostName $(hostname)
Note: arm64v8 Platforms must be running a 64 bit version of linux (Alpine 3.8 aarch64)[https://alpinelinux.org/downloads/] or (HypriotOS/arm64)[https://github.com/DieterReuter/image-builder-rpi64/releases]
The above starts Databox using pre-build images published on Docker hub and runs Databox on your local machine.
Once it's started, point a web browser at http://127.0.0.1 and follow the instructions to configure your HTTPS certificates to access Databox UI securely (using a web browser https://127.0.0.1, or the iOS and Android app).
Note: Using the databox iOS and Android apps with MacOS may require you to modify your firewall to enable external access to port 80 and 443.
To stop databox and clean up,
docker run --rm -v /var/run/docker.sock:/var/run/docker.sock -v -t databoxsystems/databox:0.5.1 /databox stop
The graphical SDK will allow you to quickly build and test simple databox apps. To start the SDK run:
docker run --rm -v /var/run/docker.sock:/var/run/docker.sock -v -t databoxsystems/databox:0.5.1 /databox sdk -start
The SDK web UI is available at http://127.0.0.1:8086
To stop the SDK run:
docker run --rm -v /var/run/docker.sock:/var/run/docker.sock -v -t databoxsystems/databox:0.5.1 /databox sdk -stop
It is possible to develop Databox apps and driver without the SDK. Currently, Python, Golang and NodeJs all have support libraries. Building outside the SDK allows you to make smaller more efficient containers and use more third-party libraries.
To get started all you need is a Dockerfile and a databox-manifest.json examples can be found in the libraries '/samples' directories. To make your app available to install locally on your databox you will need to upload the app-store driver and use docker build -t [your-app-name] .. Once the manifest is uploaded and the image has built then you should be up to install the app on your local Databox.
A good place to get started is the databox quickstart repo which has all you need to develop apps and drivers and a small tutorial.
Images must be post fixed with -amd64 or -arm64v8 respectively. The image must have the version tag that matches your running version of databox :0.5.1 or :latest for example.
If you would like to modify one of the currently available actual drivers you can do so by doing the following:
./databox-install-component driver-os-monitor
This will download and build the code on your machine and upload the Databox manifest to your local app store. You can also use this with your repositories and forks using:
./databox-install-component [GITHUB_USERNAME]/[GITHUB_REPONAME]
To develop on the platform and core components the databox start command allows you to replace the databoxsystems core images with your owen. For example to replace the arbiter.
docker build databoxdev/arbiter . # build your updated arbiter image
make start OPTS=--release 0.5.1 --arbiter databoxdev/arbiter # start databox using the new code
Databox has a number of platform components, divided into two parts: Core and Other components. Core components are required for Databox function. Other components of things like apps and drivers to demonstrate Databoxes functionality.
- Databox-container-manager Container manager controls build, installation and running functions of the other databox components.
- databox-arbiter Arbiter manages the flow of data by minting tokens and controlling store discovery.
- databox-export-service This service controls the data to be exported to external URLs.
- core-store This is a data store used by apps and drivers to store and retrieve JSON data or JPEG images.
- core-ui This is the databox default user interface.
- driver-app-store This is a driver for retrieving manifests and making them available to your databox.
- driver-sensingkit This driver provides SensingKit mobile sensor data.
- driver-google-takeout This driver supports bulk import of google takeout data.
- driver-phillips-hue This drivers allows connection to Phillips Hue Platform.
- driver-os-monitor This driver monitors the databox hardware by fetching Memory consumption and CPU load.
- driver-twitter This driver streams data from a twitter account into a datastore.
- driver-tplink-smart-plug This driver collects data from TP-Link smart plugs.
- app-light-graph An app that plots mobile phone light sensor data.
- app-twitter-sentiment An app that used data from driver-twitter to calculate tweet sentiment.
- app-os-monitor An app to plot the output of the data generated by driver-os-monitor.
For writing a new driver or app for Databox, one needs Databox APIs. To make app/driver development easy, we have wrapped Databox APIs in nodejs, python and go. Using any of these libraries, a developer can build their databox app/driver.
- lib-node-databox: Databox Nodejs API library for building databox apps and drivers.
- lib-python-databox: Databox Python API library for building databox apps and drivers.
- lib-go-databox: Databox Go API library for building databox apps and drivers.
Databox System Design document can be find here and general API specifications are here.
Multi arch builds only work on Docker for Mac experimental enable docker cli experimental features "experimental": "enabled" ~/.docker/config.json
make all ARCH=amd64 DEFAULT_REG=[your docker hub reg tag]
Or for amd64v8 platforms ''' make all ARCH=amd64v8 DEFAULT_REG=[your docker hub reg tag] '''
make test
For more details, have a look here.
The databox project welcomes contributions via pull requests see CONTRIBUTING.md for more information. A good start is to look at the current issues and forking the databox repo and fixing bugs/issues and submitting a pull request. Read more on Fork and Pull here.
This documentation is up-to-date till this commit. The master branches on all components point to the current release and are tagged in git using semver.
The list of contributors who participated in this project.
MIT Licence, See here.
The Databox project welcomes contributions via pull requests see CONTRIBUTING.md for more information.
EP/N028260/1, Databox: Privacy-Aware Infrastructure for Managing Personal Data
EP/N028260/2, Databox: Privacy-Aware Infrastructure for Managing Personal Data
EP/N014243/1, Future Everyday Interaction with the Autonomous Internet of Things
EP/M001636/1, Privacy-by-Design: Building Accountability into the Internet of Things (IoTDatabox)
EP/M02315X/1, From Human Data to Personal Experience