pdf version of this lecture as delivered 12 Feb 2018

Metadata IS data

contains fields with values

Metadata IS NOT data

at least not the data you are interested in. It describes the data that will be of interest, but does not represent or contain that data.

Data ‘reporting’

• WHO created the data?
• WHAT is the content of the data?
• WHEN were the data created?
• WHERE is it geographically?
• HOW were the data developed?
• WHY were the data developed?

Who uses Metadata?

• you!
  • discovery of past projects
  • filtering (date, location, etc of datasets)
• Somebody else
  • meta-analysis (data citation)
    • This is the future currency in academia, and has been for some time…
  • web aggregators / engines (‘the semantic web’)
  • future scientists

You may be required to create metadata

• submission to archives
• submission to journals

Examples of applicable journal data submission statements

Nature

AGU

BMC Biology

Biological Conservation

Only slight difference between descriptive metadata and proscriptive metadata where the latter is required format to fit into some database or data sharing scheme

e.g. camera trapping metadata standard

General Metadata contents

Identification Information

• People
• project identifiers

Purpose

• original collection motivation
• associated papers or methods

Dates, times, and locations

• where and when are fundamental to utility of all data
• details vary based on data type
  • GIS: specific formatting for geospatial representation
  • weather data: date, time format important

Data dictionary

• table by table, column by column
• for each vector or element of information:
  • what is it called?
  • what is its type?
    • character (length? allowable values?)
    • numeric (integers? precision? decimal place limit?)
    • logical
    • null?
  • what is its description?

Constraints, Liabilities, and Citation instructions.

• What if any are the limits on use of the data?
• What responsibility do the named people in the metadata document have for the underlying data?
• How should users cite this data package?

Metadata Schema Examples

Darwin Core

• Example of “simple” Darwin Core

EML

• strong typing of allowable fields
• example
• Morpho
  • 110 page user guide

XML and EML are “human-readable” but only in the technical sense

USFS FIA metadata

An open standard for camera trap data

• Hierarchical nesting of images (within sequence) within deployment within project
• allows standardization of data storage & sharing
• using XML or JSON template allows data to fit into API web publishing

MN Geographic Metadata guidelines

Ad-hoc, institutionally consistent formats