Promoting an Extended Date-Time Format
For people that work with bibliographic data, you quickly realize that standard date-time formats formats often don’t go far enough. So it’s nice to see this effort from the Library of Congress. It seems to be a clean superset of the ISO 8601 date-time format, and to cover the most important missing pieces. This is a datatype that can and should be used anywhere that people need to represent bibliographic dates. I’m interested in using it in both RDF (Dublin Core) and in JSON, for example.
Yet the first paragraph of the description page presents the effort much more narrowly:
There is no standard date/time format that meets the needs of various well-known XML metadata schemas, for example MODS, METS, PREMIS, etc. For several years there has been discussion of developing a reasonably comprehensive date/time definition for the bibliographic community, and submitting it either for standardization or some other mode of formalization - a W3C note for example, a NISO Profile, and/or an amendment to ISO 8601.
- MODS, METS, etc. are not at all “well-known” except in the library world
- XML is not the only way to represent or move around data in 2009 (RDF and JSON are two I quite like, for example)
- XML Schema is not the only way to represent XML formats; many people avoid it like the plague
- formal standards are over-rated; better to first establish de facto standards through adoption
So my observation is simply this: the EDTF is little gem, and can and should be widely used in a variety of different contexts. The LoC should recognize this and adjust details (documentation, examples, namespace URI) accordingly to promote it as such.