Loading Data
The Data tab is where you choose the dataset to represent. Umwelt accepts tabular data — rows of records with named columns — as JSON or CSV.
The data table
The top of the tab shows the current dataset as a table, one row per record. Values are formatted according to each field's current type and transforms, so this table is also a quick way to confirm that dates and numbers were parsed the way you expect.
Upload a file
Under Upload JSON or CSV file, choose a file from your computer. JSON files should contain an array of objects; CSV files should have a header row. The data is parsed, type-coerced, and loaded into the editor.
Uploaded data stays in your browser — nothing is sent to a server. If you later share your work as a URL, the dataset is embedded in the link itself so others can open it.
Load from a URL
Under Load data from URL, paste a link to a JSON or CSV file hosted on the web. Umwelt fetches and parses it the same way. Specs that reference a URL stay small when shared, since the link points at the data rather than embedding it.
Recently uploaded files
Datasets you've loaded before are listed under Recently uploaded files, so you can switch between them without re-uploading. They're stored locally in your browser; use the Remove button next to a file to forget it.
Example datasets
The Example datasets list offers a selection from the vega-datasets collection — including stocks.csv, cars.json, weather.csv, penguins.json, and gapminder.json — useful for trying out the editor before bringing your own data. When you first open the editor with no data loaded, the first example dataset is loaded automatically.
What happens after loading
Loading a dataset kicks off Umwelt's inference: it creates a field for each column, infers field types, detects a key, and proposes a default visualization and sonification. Your working state is saved in the browser, so reloading the page picks up where you left off.
Next
- Configuring Fields — review the inferred types and key, and assign encodings.