Small issue, but I think it would be cleaner to rename some of the fields.
Data publication version -> Publication version: it is not just relevant to the data part, a header change could also trigger a version increment.
Length of time series identifier -> Byte length of channel identifier Bytes makes clear the length is byte count. Maybe potential for unicode in future, in which case string length is different. I am not advocating unicode by the way. Also this identifies the channel the time series came from, not the time series itself.
Length of extra headers -> Byte length of extra headers Bytes makes clear what the length is, this is not the number of extra headers.
Length of data payload -> Byte length of timeseries Bytes makes clear what the length is, not number of samples. Also call it timeseries instead of data payload.
Time series identifier -> Channel identifier identifies the channel the time series came from, not the time series itself.
Data payload -> Timeseries Clearer as to what this is, the data part is a timeseries, not some generic data.
I would also use the word "Channel identifier" instead of "time series identifier" in the FDSN Identifiers document. To me a "time series" is the output data samples, not the source of the data samples.
Small issue, but I think it would be cleaner to rename some of the fields.
Data publication version -> Publication version: it is not just relevant to the data part, a header change could also trigger a version increment.Length of time series identifier -> Byte length of channel identifierBytes makes clear the length is byte count. Maybe potential for unicode in future, in which case string length is different. I am not advocating unicode by the way. Also this identifies the channel the time series came from, not the time series itself.Length of extra headers -> Byte length of extra headersBytes makes clear what the length is, this is not the number of extra headers.Length of data payload -> Byte length of timeseriesBytes makes clear what the length is, not number of samples. Also call it timeseries instead of data payload.Time series identifier -> Channel identifieridentifies the channel the time series came from, not the time series itself.Data payload -> TimeseriesClearer as to what this is, the data part is a timeseries, not some generic data.I would also use the word "Channel identifier" instead of "time series identifier" in the FDSN Identifiers document. To me a "time series" is the output data samples, not the source of the data samples.