Without restricting it to being machine readable, I think we would all agree there are three ages of data:
- Past data
- Present data
- Future data
And we have common goals for data (or parts of it):
- Past data – To understand past data.
- Present data – To be understood by others.
- Future data – For our present data to persist and be understood by then users.
Common to those ages and goals is the need for management of identifiers for our data. (Where identifiers may be data as well.)
I say “management of identifiers” because we cannot control identifiers used in the past, identifiers used by others in the present, or identifiers that may be used in the future.
You would think in an obviously multi-lingual world that multiple identifier identification would be the default position.
Just a personal observation but hardly a day passes without someone or some group saying the equivalent of:
“I know! I will create a list of identifiers that everyone must use! That’s the answer to the confusion (Babel) of identifiers.”
Such efforts are always defeated by past identifiers, other identifiers in the present and future identifiers.
Managing tides of identifiers is a partial solution but more workable than trying to stop the tide.
What do you think?