Metadata management solutions aim to solve this problem, and there are many promising developments in this space including Lyft’s Producers wonder why consumers aren’t satisfied, and consumers wonder why the data is never “right”. Producers may publish whatever data is accessible by the system or seems relevant, and consumers may be quick to assume tables or fields that “sound right” happen to be custom-fit for their top-of-mind question. This ambiguity can create a strange symbiosis between data producers and consumers. Published data tables, however, exist in an ambiguous gray area static enough not to be considered a “service” or “software”, yet too raw to earn the attention to user experience given to interfaces. Programs with graphical user interfaces form such contracts by labeling and illustrating interactive components to explain their intent (for example, the “Save” button typically does not bulk-delete files). Mature code packages and APIs document expected inputs and outputs, check adherence with unit tests, and transparently report code coverage.

Software products use a range of strategies to make promises or contracts with their users.

Free cat tool metacat