Feedback on rgbif manuscript

[sckott]

If you’ve used rgbif - I’d love to get your thoughts/feedback on the manuscript in draft:

• .Rmd - https://github.com/sckott/msrgbif/blob/gh-pages/manuscript.Rmd (this is where the writing happens)
• .pdf - https://github.com/sckott/msrgbif/blob/gh-pages/manuscript.pdf (download pdf link: https://github.com/sckott/msrgbif/raw/gh-pages/manuscript.pdf)

If you want to build locally

git clone git@github.com:sckott/msrgbif.git
cd msrgbif
make

I only have plans to put up as a pre-print - not going to submit to a journal most likely

Can leave comments here, or comment on the repo via commit comments, or PR’s or issues - whatevs

UPDATE: I’m working on a Python client as well, and may include some discussion of the Python client in the MS, though maybe not, haven’t decided

[rmounce]

The use cases section looks really useful. I will probably be using something similar to the ‘Get Distribution map of Quercus lobata’ use case to map GBIF data for IUCN RedList-assessed plant species.

I love the line: “In fact, the rgbif package has already been used in 22 scholarly publications” – fantastic. Perhaps you should cite those papers though. It would be useful for readers to be able to see how others have used rgbif in practice.

Minor things:
(All the line numbers are with reference to the PDF draft)

line 122 “When you hit a “large dataset” will…”

Although it’s convenient shorthand slang I don’t think it’s clear what ‘hit’ means in this context. Perhaps use clearer, less informal language i.e. “If you execute a query for which more than 200,000 records match…”

line 136 “…whereas you can not do that in the occurrence search API”

suggest ‘cannot’ in place of can not.

"Cannot" Or "Can Not"? | Lexico

line 177 “Get only those occurrences with spatial issues”

What is a spatial issue? I can guess but honestly I’m not sure. Some explanation for those not intimately familiar with GBIF would be helpful here.

line 188 “Another package, spoccutils”

??? Incomplete / leftover sentence?

[sckott]

Thanks so much for the feedback @rmounce ! This is great.

I will probably be using something similar to the ‘Get Distribution map of Quercus lobata’ use case to map GBIF data for IUCN RedList-assessed plant species.

Cool! See also https://github.com/ropenscilabs/rredlist

Perhaps you should cite those papers though. It would be useful for readers to be able to see how others have used rgbif in practice.

Good idea, will do

Although it’s convenient shorthand slang I don’t think it’s clear what ‘hit’ means in this context. Perhaps use clearer, less informal language i.e. “If you execute a query for which more than 200,000 records match…”

Thanks, will fix

suggest ‘cannot’ in place of can not.

will do

What is a spatial issue?

Right, will explain that. They are occurrences that have some sort of note from GBIF about the occurrence, e.g., see the issues key in http://api.gbif.org/v1/occurrence/1227769194 [quote=“rmounce, post:2, topic:290”]
??? Incomplete / leftover sentence?
[/quote]

thanks will fix