Improvements for next course

[andersfi]

Simple list of issues / improvements noted during Nordic OIKOS Workshop

Please feel free to amend or comment on the list so that we have something to grab hold of when keeping this course next time.

• There was quite a lot of time spent on getting the technicalities working. Either technicalities like working with git or github and installing various packages. Working through a cloud service like https://rstudio.cloud/ may speed up things. Docker is also a possibility, but would require students to install this first (maybe high user threshold?)

• Make sure the scripts (to as large extent as possible) are system agnostic, i.e. works on both windows, mac and linux.

• A gentle introduction to standards (DwC, EML) at the start of the course may help users faster recognize the data and data-structures.

[ErlendNilsen]

• We might also consider including an introduction to the ethics of open/big data - including licenceces, duties of data publishers, responsibility of data users, including a short introduction to current debates on these issues. Basically some research ethics. This should include introduction to terms like RRI and FAIR-principles.

[andersfi]

• Add package name before every function used, good practise + it helps student see from which package the different functions in scripts come from - e.g. rgibf::occ_search(*)

[mskyttner]

• More examples using leaflet? Perhaps using mapedit and mapview to get wkt data interactively? Then use that in subsequent rgbif-call.
• A simple example involving working with and validating tree-structured data (site - plot - subplot?) - could use packages igraph, tidygraph, ggraph

[ErlendNilsen]

• At the risk of becoming excessively long, I think it would be nice also with an event-core example focusing exclusively on site-occupancy modelling. That is - we need a data set with presence-absence data, so that we can show the benefits of that. This can be included before the line transect case, which focus on abundance.