Dear All,
To acquaint myself with the Biological Collections Ontology I went through it and made some notes. John Wieczorek suggested I share these notes so that they can be responded to openly. They are as follows, I look forward to your comments.
1. Why are the dates clustered together in one branch, but day, month, year separate?
2. eventDate: regarding date ranges. Sometimes an event continues for a period such as a field trip, but often a date range is required because although the event took place in a certain time frame, the exact event date is not known. There should perhaps be a way of distinguishing between these two situations.
3. Vernacular name should be associated with a language and a geographic area where it is used?
4. verbatimSRS, footprintSRS and geodeticDatum are similar but not the same. They seem to overlap. The SRS has a clear definition epsg- registry.org, but geodeticDatum is a more general term. The definitions and usages could be stronger.
5. There is a need for local grid reference systems. Many countries have their own ways of writing grid references and verbatimLongitude and verbatimLatitude are not suitable for these. For example, the UK has grid references such as NZ28F, which mean something very specific in the country. It is a similar situation in Belgium and I imagine other countries.
6. georeferenceVerificationStatus & identificationVerificationStatus: It is not clear how these should be used. The names are similar, but the described usage is very different. I find it hard to imagine how they would be used in practice.
7. There are occasions within biogeographic databases where you want to indicate that an observation or collection is not trusted. I can’t see a place where a record can be verified as a whole. For example, if it is suspected that the label has been muddled up or the collector is suspected of fraud. With an observation there is no specimen to verify, but the record should be verifiable.
8. member, lithostratigraphicTerms, bed etc etc. I don’t know anything about geology, but these terms seem to overlap and could perhaps be thought of as synonymous. It would be valuable to find records where these terms are required to see how they are used in practice.
9. county & stateProvince: These are parochial terms. There are many nested subdivisions of countries and there is probably a better way to describe them. Perhaps there is already a hierarchical ontology to describe geographic areas.
10. Abundance estimates: There are several ways to describe estimates of abundance, such as DOMAIN, DAFOR, percent coverage, is there a way to accommodate these?
11. Controlled vocabularies: Many terms suggest using a controlled vocabulary, but most don’t have one. Is this an area we should work on?
12. Consider the importance of citation and providence the “according to” information seems rather weak. There should perhaps be a space for an identifier (ORCID). Also, biographic information such as date of birth, date of death, institutional association. These are useful for providence, but also for validation of records.
13. What is the difference between fieldNumber and recordNumber?
14. establishmentMeans: the description seems to conflate three concepts, the length a taxon has been established at a location (e.g. native); the way it got to the location (e.g. introduced) and the dispersive ability of the taxon (e.g. invasive). To my mind the latter is not required as it is an opinion about the taxon as a whole not the observation. The first two should be separated to make them useful. People need to distinguish between the native status of a species at a location and how it got to that location. Invasive species biologists are particularly interested in the vectors of introduction.
15. The health of an animal or plant can be noted on an observation and is useful for analysis. Is there some way of stating that the organism was dead when the specimen or observation was made?
Regards Quentin