hettling
commented
5 years ago Good descriptions have been added for the main datatypes: Specimen, Taxon, GeoArea, Multimedia. Also the response and client classes are now properly documented. There is, still not a full description for each object model, since these are parsed out of code annotations from the NBA itsself. They would have to be added there by our developers.
However, by looking at the fields, the user can pretty much see what the class is about. Furthermore, these classes are mere containers for data - there is not much functionality that needs to be described.
In one of the vignettes, it was suggested I “Have a look at ?Response to see what this object contains.” -- yet the help file had little to no information. I think the level of information found in this help file is not too different from that of others.
I imagine that much of the documentation will have been generated with the code generator but i think it may be necessary to provide more information particularly for the more commonly used functions/classes. In particular, there are very few examples provided for the majority of the help files. I think it is required that all public functions/classes come with examples for any ROpenSci package.
With roxygen, examples can be stored in an examples/ folder and called in using roxygen syntax.