JOSS manuscript review

[PieterjanRobbe]

This issue entails my review comments for the JOSS manuscript submission at openjournals/joss-reviews/issues/7047.

JOSS requires a "State of the field: Do the authors describe how this software compares to other commonly-used packages?". I couldn't explicitly find this in the paper. Does that mean that 4DModeller provides unique functionality that can't be found elsewhere?

Also, if I'm really nitpicking, I noticed that fdmr software package is written in typewriter font, but other software packages, e.g., inlabru, are written in normal font. Perhaps you could use typewriter font everywhere?

Thanks!

[richardwestaway]

We have updated the manuscript with additional text about state of the field:
"There are number of packages within the R ecosystem that perform various spatio-temporal functions (e.g., spTimer MCMC modeling for space-time data, spBayes MCMC modeling of random effects for space-time data, and bmstdr a model comparison tool that uses MCMC, INLA, and other methods to fit to data and then compare between models). A critical expectation of each of these packages is a deep background in both bayesian modeling and understanding bayesian data. Users are expected to build their own fit equations, have a deep understanding of random effects, and are expected to be able to interpret predictions and uncertainties. However, none of these packages offer much if any scaffolding to do so."

typewriter font has now been used consistently for all R package names.