Storing and processing imprecise data

Purpose

This is an original Computer Science research program to create standards, algorithms and open source library software for storing data with varying levels of precision and granularity.

For example, in an application containing family history, some information is, of necessity, anecdotal. A picture may be known to include "one of the Jones twins, but I can't remember if it's Bobby or James". The date of the photograph may be known (July 4th, say) but the year may only be approximate.

A Plan

1. Decide what we should be able to store
2. Create test cases, including performance requirements for a minimal set of features as release 0.1
3. Implement release 0.1
4. Evaluate the release 0.1 including refining overall goals
5. Repeat steps 2-4 for subsequent releases until overall goals have been met

Participants

• User:Dmclean (principal investigator)

Please add your name here if you are interested in participating as a co-investigator or reviewer.

Categories of data

• Dates and date ranges