value equal to: 'any', 'numeric', 'text', 'datetime' The data-type provides support for per column typing and allows for data validation and coercion. 'numeric': represents both floats and ints. 'text': represents a string. 'datetime': a string representing a date or date-time, in the form: 'YYYY-MM-DD HH:MM:SS.ssssss' or some truncation thereof. Years must have 4 digits, unless you use `validation.allow_YY: true`. Also accepts 'T' or 't' between date and time, and allows timezone info at the end. To convert these strings to Python `datetime` objects, use `dateutil.parser.isoparse`. In R use `parse_iso_8601` from the `parsedate` library. WARNING: these parsers do not work with 2-digit years, if you use `validation.allow_YY: true` and do not coerce to 4-digit years. And parsers that do work with 2-digit years may make a different guess about the century than we make on the front end. 'any': represents any type of data. Defaults to 'any' if undefined.