Dear Working Group II members,
The requirements for a next generation format (NGF)
https://docs.google.com/document/d/1ymAe9v1rUuucpY7ai5ilKsD7V1ejwt6GxQQmJ5IevDI/edit include "A new method of identifying a time series is required for NGF". Any group wishing to develop a prototype for a NGF must use such an identifier. Furthermore, the recently adopted StationXML 1.1 specification contains a "sourceID" attribute that may contain the same identifiers.
At the IRIS DMC we have developed a specification for a URN-based identifier, which is attached. We invite comments in preparation for a proposal to the WG, which can be sent to the WG mailing list or directly back to me. Ideally this would be accepted by WG II and the FDSN at the meeting in July.
We believe our specification is a good first step for the FDSN to transition from current SEED codes to a single string, URN-like identifier, while allowing for longer codes to be utilized when needed. Use of these new identifiers has a lot of potential and we expect that the FDSN will create other identifiers in the future. For example, this specification uses an "XFDSN:" (for eXtended FDSN) prefix or namespace and the FDSN can easily define more prefixes for different identifier schemes.
Note that this specification proposal is primarily about the form of the identifiers and we have not tried to address creating new channel codes for synthetic data, state-of-health, derived data, etc. yet. We hope this can be part of broader WG discussion.
Following the guidelines in the requirements for NGF, this specification defines a single, name-spaced string identifier. One of our primary goals was a scheme that allowed trivial mapping of SEED codes to the new identifier, and, when possible, mapping from such an identifier back to SEED codes. By incorporating the SEED codes, these identifiers are also highly recognizable for those used to working with the codes.
regards,
Chad
IRIS DMC
Chad Trabant