What can I build with the toolkit?

As long as you can dream it, you can build it with the Developer Toolkit (for the most part). Usually, your project will fall in one of 3 camps:

  • Integration: A fully integrated data product that will communicate back and forth with data.world
  • Application: A stand-alone application that may reference and access data, but users won’t interact directly with the data.world site
  • Automation: Tasks like importing/exporting data to/from data.world

Integrations

An integration is typically created with a tool that exists in tandem with data.world and communicates back and forth often. Users will be using both your product and the data.world site, and the usage will be interactive in a way that allows for sending and receiving of data from data.world to/from your tool.

Some examples are:

  • data.world’s Slack integration to explore data and get notifications on updates
    • Uses webhooks to trigger event notifications and interaction
    • Gives users notifications when updates are made to projects, datasets, or users
    • Allows subscription/unsubscription to/from users, projects, and datasets
    • Provides expanded links to offer jumping-off points for data exploration

  • An integration with R’s DBI to quickly query data from R
    • Uses the Retrieve a dataset endpoint
    • Allows for quick import of datasets to R for querying
    • Utilizes R Studio’s Connections Pane to allow users to explore full datasets in R Studio

Applications

An application, unlike an integration, doesn’t allow users to interact directly with the data.world site and features, but uses available data in a standalone application elsewhere.

If you wanted to create a tool that allows users to pull data from specific datasets through a dropdown menu on your website, but don’t want them to explore the data on data.world or use it in external tools to manage/query, an application is the most appropriate choice.

Automations

Automations are helpful when you want to get something done without having to manually recreate the task through a visual interface online. Say you wanted to add data to data.world or get data out of data.world for your work elsewhere. You could use API endpoints like Upload Files, Download a File, Retrieve a Dataset, Update a Dataset, or even query datasets with SQL Query.

Try sending a test request using the Download a File endpoint. You’ll just need the owner, dataset ID, file ID, and your API Token.

Use this information for your example:

owner: bryon
id: us-zipcodes
file: zipcode.csv

Or try using the following Shell command:

curl --request GET \
  --url https://api.data.world/v0/file_download/bryon/us-zipcodes/zipcode.csv \
  --header 'authorization: Bearer YOUR API TOKEN'

Getting Started

There are several easy ways for you to get started:

  1. Proceed with the following sections of this portal (next: Integration Toolkit, or jump right into our API docs)
  2. Check-out open-source integrations on GitHub and learn by example
  3. Reach out, pitch your idea, and get individualized support via our help portal or Community Slack