View source for Item:OSW70b4d6464c1d44a887eb86e3b39b8751

== State of the art regarding data standards ==

* Documents
** Formats themselves are highly standardized (pdf, MS Office, OpenDocument family)
** Support for metadata, but implementation in the UI mostly limited to generic information
* Images
** Formats themselves are highly standardized (png, jpeg, tiff, etc.)
** Metadata beyond generic information only supported by special formats like tiff
* Tabular data
** CSV widely used, supported by almost all software systems
** Extensive degrees of freedom almost always require prior knowledge for correct interpretation
* Domain-specific data (chemistry, engineering, computer science, etc.)
** Depends strongly on the use case
* How can we generate arbitrary structured (meta)data in a well-defined format?
** e. g. for processes running in different labs around the world?