Thread by @WeAreRLadies

Anyone wants to get started or has used #rspatial and learned from tutorials/courses using packages such as {rgdal} {rgeos} {sp} {raster}? Are you getting used to them already? You might want to reconsider 🤔
🧵 1/n

For the past few months the R Spatial community of developers and active users has been dealing with the news of the retirement of {rgdal} & {rgeos}
🧵 2/n

https://twitter.com/RogerBivand/status/1435490579006246914

For {sp} is a similar story, although still getting maintenance, a migration to {sf} should be mostly preferred for any new developments and implementations
🧵 3/

https://twitter.com/edzerpebesma/status/1272928917700165633

Finally, {raster} will also be slowly transitioning to {terra}, which according to the package documentation 'is simpler, faster and can do more.'
🧵 4/n
github.com/rspatial/terra

Let's break this down a little: Handling spatial data is a complex and ever evolving task. Great FOSS like GEOS, GDAL and PROJ have been the backbone of spatial for a lot of #gis software out there (QGIS, ArcGIS, etc.)
🧵 5/n

This was no different for spatial implementations in R, and hence initial R Spatial packages started adapting to these environments. For the full history, I recommended this recenly updated section in the Geocomputation with R book
🧵 6/n
geocompr.robinlovelace.net/intro.html#the…

Current #rspatial packages like {sf} do a direct link to these software, skipping the need of {rgdal} and {rgeos}, as shown in this slide from @edzerpebesma UseR!2021 keynote
🧵 7/n

@edzerpebesma But coming back to {sp}, what differentiates it from {sf} and why is a migration needed? A good overview of both packages can be found on this blogpost by @vivalosburros jessesadler.com/post/gis-with-…
A short summary of this and other sources below 👇
🧵 8/n

@edzerpebesma @vivalosburros As for raster implementations, {terra} and {stars} are the ones to lookout for. Both have different ways of representing and working with gridded data, and both can be useful in their own way. This is a short overview of their strengths 👇
🧵 9/n

@edzerpebesma @vivalosburros Of course many other packages are out there to work with R Spatial in R. If you are getting started or want to keep up with the new trends take a look a the Geocomputation with R website and its book being updated as we speak!
🧵 10/n
geocompr.github.io

@edzerpebesma @vivalosburros And yet another set of materials for spatial data analysis with R
🧵 11/n

https://twitter.com/jakub_nowosad/status/1467528027873628163

@edzerpebesma @vivalosburros Do you have more #rspatial materials to add to the list?
🧵 12/12

@edzerpebesma @vivalosburros Thank you @RogerBivand for the correction on CRS representation 🙌
Once again: how do {sp} and {sf} compare...