Horizon's data collection features cover a wide variety of data types and use cases, ranging from unstructured text across the web (webscraping, social media), to structured data uploaded from files (csv, geojson), to manual data entry and drawing built ad-hoc by users within the app. The Horizon project aims to allow users to determine the extent to which they are inputting information manually, drawing from automated sources from within the application, or integrating Horizon inputs with other codebases.

In general, sources are RSS and social media feeds. The Horizon subspace (codenamed Thema) makes these sources available to you, doubling as a feed reader. For structured data you'll want to use the Horizon 'datasets' or 'maps' tab, uploading structured data that imply specific normalized data schemas and a more table-like experience.

The method of data collection you use will differ depending on your goals. Take a look at some of our End-To-End tutorials if you're not certain where to start. Otherwise, you can browse this section to learn all the ways to input information into the Horizon engine.