It seems there is an issue with publication. https://www.r-spatial.org/book/ is not showing anything.

Does it work for you ?

Just in case you are not aware of this. I recall it was working fine a few weeks ago.

See quarto-dev/quarto-cli#4294

now running, still todo:

• render python branch somewhere else (netlify?)
• run on PRs
• Python branch in cron job
• add to README that this is a daily built, unchecked
• install spatialreg from CRAN, once released
• put errata somewhere, with link

How should I update/re-generate the netlify rendering after edits locally? Through PRs, the main site (_book?) will not update until you change/merge, right? I shouldn't mess with _book, I think, because then PRs will pick up my changes ...

See https://github.com/edzer/sdsr_exercises/actions/runs/7824811792/job/21347983478#step:8:506

Is this caused by the new spdep on CRAN? @rsbivand

I think it appease in section 6.2.1 and is never really defined. But it might be obvious to this audience, and not need defining.

"With two affine parameters..."

(Don't tell anyone, but I might have had to peek at the definition to remind myself.)

Should I download my own aq files to be able to rebuild the whole book to check the formatting in ch 17-18?

In the following bullet point in the first section of Ch 11 there's a small typo, I guess it should be "they".

second order properties, e.g. pairwise interactions: given a constant or varying intensity function, are points distributed independently from one another, or do the tend to attract each other (clustering) or repulse each other (appear regularly distributed, compared to independence)

I have the vector PostgreSQL driver but not PostGIS. I do not want to install PostGIS (because it depends on GDAL/GEOS/PROJ which I install from source and change frequently). The has_PG test doesn't check that the driver works. How to skip these sections?

Have you considered adding a short chapter, on the same level as "R data structures" chapter?

I think the installation of geospatial packages has made a huge leap forward, but there still might be times or reasons where readers of the book would profit from a quick introduction in the the `rocker/geospatial´ images.

This section could also extend a bit on the idea of reproducibility, which so far is only mentioned in the preface.

I studied a few pieces of puzzle of 'Spatial data science' from here and there

( It's always not enough for me. )

and found your draft by accident.

It is to me something like '101 stuff for practical spatial data science'

I really enjoyed reading your draft.

Thank you for sharing such a great work!

Do you have any plan to publish the book? (when?)

I am eager to know it and buy one.

Both pdf and ePub download buttons lead to broken links.

Hello,
I'm here just wondering the why of:

r <- lapply(files[-1], function(f) read.csv(f))

in particular I'm referring to the subsetting files[-1];

if I understand well the purpose (reading all files), shouldn't be something like this:

r <- lapply(files, function(f) read.csv(f))

or do I miss something?

Thank you so much for your excellent work.

Best

Should we mention for example gridGraphics:: grid.echo () and grid::grid.grab() ? Should we point to geocompr section 8.2.7 on viewports and inset maps? Could tmap be mentioned in 9 in the grid part? Or point to geocompr and tmaps JSS article?

My ch 17 is now 35K, and could happily be two shorter chapters, 17 Spatial weights, 17a Spatial autocorrelation, followed by 18 Spatial regression. How do I do that? Bump 18, 19, 20 with git mv?

How can I build the output of one or more chapters? Run make book? That builds everything, doesn't it? Can I build out-of-repo so that I avoid generating a PR including the built book? It is hard to write without control over the output.

12.7.1 A population grid refers to:

v = vroom::vroom("aq/pop/Zensus_Bevoelkerung_100m-Gitter.csv")

And 13.1 Preparing the air quality dataset refers to:

library(tidyverse)
read.csv("aq/AirBase_v8_stations.csv", sep = "\t") %>% 
    as_tibble()  %>% 
    filter(country_iso_code == "DE", station_type_of_area == "rural", 
                 type_of_station == "Background") -> a2

But they are not in here: https://github.com/edzer/sdsr/tree/main/aq

Ch 16 is a bit unpolished; some figures deserve more explanation, some missing punctuation, etc. But OK to send to reviewers. I see that Ch 17-18 have a higher reference density, maybe that doesn't matter.

Something with the double commas was not interpreted correctly by the copy editing process for the hardcover. Instead of successive commas, it looks like a uncurly/straight double quote mark (that is also lowered like a subscript).

We see a new error in this action

Quitting from lines 180-184 [unnamed-chunk-12] (16-SpatialRegression.qmd)
1080 Error in `initializePtr()`:
1081 ! function 'cholmod_factor_ldetA' not provided by package 'Matrix'
1082 Backtrace:
1083   1. lme4::lmer(...)
1084   3. lme4 (local) `<fn>`(...)
1085   5. methods (local) `<rfMthdDf>`(...)
1086   6. methods::new(def, ...)
1087   8. methods::initialize(value, ...)
1088   9. .Object$initialize(...)
1089  10. lme4 (local) initializePtr()

@rsbivand is this something you could look at?

Are:

https://lists.osgeo.org/pipermail/proj/2019-July/008748.html
https://lists.osgeo.org/pipermail/proj/2019-August/008750.html

useful for ch 2 and section 7.3? Should we consider moving to coordinate_metadata with NA for unknown epoch (S3 easy, is.null(epoch) as the test)? Should sp::CRS also move (S4 harder, breaking change in CRS object with new slot)?

Hi, there is an error with the last line here. Not sure what the correct code should be.

In this section: 7.4.2 Example: Bristol origin-destination datacube

library(units)
a = set_units(st_area(st_as_sf(o)), km^2)
o$sum_km = o$sum / a
d$sum_km = d$sum / a
od = c(o["sum_km"], d["sum_km"], along = list(od = c("origin", "destination")))
plot(od, logz = TRUE)

Error in Ops.units(at, br[1]) : both operands of the expression should be "units" objects

> sessionInfo()
R version 4.0.3 (2020-10-10)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.6 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.7.1
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.7.1

locale:
 [1] LC_CTYPE=C.UTF-8       LC_NUMERIC=C           LC_TIME=C.UTF-8        LC_COLLATE=C.UTF-8     LC_MONETARY=C.UTF-8   
 [6] LC_MESSAGES=C.UTF-8    LC_PAPER=C.UTF-8       LC_NAME=C              LC_ADDRESS=C           LC_TELEPHONE=C        
[11] LC_MEASUREMENT=C.UTF-8 LC_IDENTIFICATION=C   

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] units_0.7-2       spDataLarge_2.0.6 xts_0.12.1        zoo_1.8-8         spacetime_1.2-8   stars_0.5-2       abind_1.4-5      
 [8] forcats_0.5.0     stringr_1.4.0     dplyr_1.0.8       purrr_0.3.4       readr_2.0.2       tidyr_1.2.0       tibble_3.1.6     
[15] ggplot2_3.3.5     tidyverse_1.3.0   sf_1.0-8         

loaded via a namespace (and not attached):
 [1] httr_1.4.2         jsonlite_1.7.2     modelr_0.1.8       assertthat_0.2.1   sp_1.4-5           cellranger_1.1.0   yaml_2.2.1        
 [8] pillar_1.6.4       backports_1.2.1    lattice_0.20-41    glue_1.6.2         digest_0.6.27      rvest_1.0.1        colorspace_2.0-2  
[15] htmltools_0.5.2    pkgconfig_2.0.3    broom_0.8.0        haven_2.3.1        s2_1.1.0           intervals_0.15.2   scales_1.2.0      
[22] tzdb_0.1.2         proxy_0.4-26       generics_0.1.2     ellipsis_0.3.2     withr_2.5.0        cli_3.3.0          magrittr_2.0.1    
[29] crayon_1.4.2       readxl_1.3.1       evaluate_0.14      fs_1.5.0           fansi_0.5.0        MASS_7.3-53        xml2_1.3.2        
[36] lwgeom_0.2-5       class_7.3-17       tools_4.0.3        hms_0.5.3          lifecycle_1.0.1    munsell_0.5.0      reprex_0.3.0      
[43] packrat_0.7.0      compiler_4.0.3     e1071_1.7-9        rlang_1.0.2        classInt_0.4-3     grid_4.0.3         rstudioapi_0.13   
[50] cubelyr_1.0.1      rmarkdown_2.11     wk_0.6.0           gtable_0.3.0       DBI_1.1.1          R6_2.5.1           lubridate_1.7.10  
[57] knitr_1.34         fastmap_1.1.0      utf8_1.2.2         KernSmooth_2.23-17 stringi_1.7.4      parallel_4.0.3     Rcpp_1.0.7        
[64] vctrs_0.3.8        dbplyr_2.1.1       tidyselect_1.1.2   xfun_0.25  

Chapter 8:

• Gb or GB? Tb or TB? You use Gb and Tb, but (correct me if I am wrong) I think it should be GB and TB
• " In such cases it may be most helpful to either gain access to (typically: rent) virtual machines in a cloud where these data are available and nearby" - there is something wrong with this sentence...

Chapter 9:

• You use "color" in some sentences and "colour" in others...
• "Good colors scales are e.g. found in packages RColorBrewer [@R-RColorBrewer], viridis [@R-viridis] or colorspace [@R-colorspace; @colorspace]." - maybe it is worth to mention the hcl.colors and palette.colors functions as they have all of the palettes methioned before...
• A few figure captions are not completed, for example "fig captions tmap: using ..."
• @ref(fig:mapviewfigure) is broken

Hi @edzer,

I'm learning spatial statistics and I'd like to, first of all, thank you and @rsbivand for publishing this open access book. Currently, I've read more than half of it and I'd to ask the following questions:

1. What style guide are you following?
2. Why function arguments within the text end in =? For example, 'Durbin= argument' from Ch. 17.
   Would removing them create unnecessary confusion?

Regarding the style guide, I've noticed that dashes aren't used in the text. Following the New Hart's Rules: The Oxford Style Guide (2014), which I use as a reference, we can choose between en (–) or em (—) dashes.

4.11.1 En rule (p. 86)
The en rule (US en dash) (–) is longer than a hyphen and half the length of an em rule (—). Many British publishers use an en rule with space either side as a parenthetical dash, but Oxford and most US publishers use an em rule.

If we chose the en rule, this is how it would look like (Examples from Ch. 1):

• 'Figure 1.1: A first map: birth counts 1974–78, North Carolina counties' [1]
• 'When we have longer time series associated with geometries, neither option – distributing time over multiple columns, or stacking columns while repeating geometries – works well, and a more effective way of storing such data would be a matrix or array, where one dimension refers to time, and the other(s) to space.' [2]

Similarly, there are a few typos regarding linking chapters. Instead of [3]

is shown in @fig-tmapstars. More examples of the use of tmap are given in Chapters -@sec-area - -@sec-spatglmm.

it should be

is shown in @fig-tmapstars. More examples of the use of tmap are given in @sec-area–@sec-spatglmm.

It also happens in 10.2 Support and statistical modelling.

I can correct this punctuation issue if you want me to.

I had to look up affordances, and suggest properties instead (line 31).

Edzer,

I am not up to speed on git to fix this myself, but as you asked for comments on twitter, there is a misspelling in the Preface:

"coordinate pairs really are pairs, and loose much of their meaning when treated separately"

"loose" should be "lose"

Hello @edzer. I'm reading your book Spatial Data Science. It's awesome! I like doing spatial analysis in the sf way~

I have a little question In chapter 16, although the URL of air quality data is provided, and it is said:

It returns a text file with a set of URLs to CSV files, each containing the hourly values for the whole period for a single measurement station. These files were downloaded and converted to the right encoding using the dos2unix command line utility.

I'm still confused about how to get the data in the right place so that I can follow this chapter. Could you please provide more details?

Many thanks for your kind guidance!