𝚖𝚊𝚌𝚔𝚊 𝚜𝚣𝚎𝚌𝚑𝚗𝚘
LightNews LightNews
banner
mackaszechno.bsky.social
𝚖𝚊𝚌𝚔𝚊 𝚜𝚣𝚎𝚌𝚑𝚗𝚘
@mackaszechno.bsky.social
1.3K followers 620 following 3.3K posts
#rstats geospatial data analyst. GIS expert, of a sort. Sci-fi fan and would be writer. Black cat owner. Occasionally apocalyptical. Can't DM.
WORDS ARE QUICK AND WORDS ARE VAIN, THE SINGLE SURE AND FINAL ANSWER MUST BE PAIN. -Dan Simmons, Olympos
Posts Replies Media Videos
mackaszechno.bsky.social
Another pic of Alice. She likes to wait at the door of my home office so she can follow me downstairs and demand treats.
Black cat patiently loafing #blackcat
November 22, 2025 at 10:00 PM
mackaszechno.bsky.social
My work setup. My process is constantly referring back to saved snippets of code that worked, reviewing help in RStudio and online, and deleting off all the non-working stuff until I get something I think is acceptable 😜
#30DayMapChallenge | Day 23 - Process
My setup with black cat and hopefully nothing to incriminating.
November 22, 2025 at 8:10 PM
mackaszechno.bsky.social
Ice shelf collapses in Antarctica using {mapgl}.
#30DayMapChallenge | Day 22 - Data challenge: Natural Earth #rstats
(code in reply)
Antarctica in {mapgl} with "WGS 84 / Antarctic Polar Stereographic EPSG:3031" shown on a globe as a circle. "WGS 84 / Antarctic Polar Stereographic EPSG:3031" viewed projected in {mapview} rather than as a globe.
November 22, 2025 at 7:47 PM
mackaszechno.bsky.social
Crawley and Gatwick...
Zoomed in to reveal more bus stops and Gatwick Airport train station and its 16.5m annual passengers.
November 21, 2025 at 3:09 PM
mackaszechno.bsky.social
Icons, revisiting Day 7.
#30DayMapChallenge | Day 21 - Icons #rstats
(code in reply)
November 21, 2025 at 3:07 PM
mackaszechno.bsky.social
Natural Earth to the rescue with rivers_lake_centerline, lakes, ocean, and bathy from 0-10000m.
#30DayMapChallenge | Day 20 - Water #rstats
(code in alt)
bathy_all_simp <- bathy_all |> ms_simplify(keep = 0.05) maplibre(style = carto_style("dark-matter-no-labels")) |> add_fill_layer(id = "ocean", source = ocean50, fill_color = "#008b8b") |> add_fill_layer(id = "lakes", source = lakes50, fill_color = "#e0eeee") |> add_line_layer(id = "rivers", source = rivers50, line_color = mapgl::interpolate( column = "scalerank", type = "linear", values = c(1, 12), stops = c("navy", "lightblue") ), line_width = mapgl::interpolate( column = "scalerank", type = "linear", values = c(1, 12), stops = c(2, 1) ), popup = "name") |> add_fill_layer(id = "bathy", source = bathy_all_simp, fill_color = "#f0fff0", fill_opacity = 0.4, fill_outline = mapgl::interpolate( column = "depth", values = c(0, 10000), stops = c("lightblue", "darkblue") ) ) |> add_control("<div style='padding:5px'> <b>Oceans, Rivers, and Lakes of the world</b><br /> with bathymetry from 0 to 10000m<br /> data: <a href='https://www.naturalearthdata.com/'>naturalearthdata.com</a><br /> macka szechno<br />20 November 2025 </div>") North Atlantic zoomed in
November 20, 2025 at 6:10 PM
mackaszechno.bsky.social
A day late due to man flu. West Sussex in one correct project (BNG 27700) and 3 inappropriate projections with area calculations showing how the deformation affects the total.
Note: I don't use {ggplot} much so plots are fairly basic.
#30DayMapChallenge | Day 19 - Projections #rstats
(code in alt)
library(sf) library(units) library(dplyr) library(ggplot2) library(patchwork) d <- read_sf("./OS.WESTSUSSEX_DISTRICT.shp") d <- d |> select(NAME) plot_projection <- function(d, crs, text){ dt <- d |> st_transform(crs = crs) ggplot(dt) + geom_sf(color = "white", aes(fill = NAME)) + coord_sf(crs = crs) + geom_sf_text(aes(label = NAME), size = 3) + theme(legend.position = "none") + labs(title = paste0("West Sussex - ", text, "\nArea calculation: ", st_area(dt) |> set_units("km^2") |> sum() |> round(), "km²")) } p1 <- plot_projection(d, 27700, "EPSG:27700 British National Grid") p2 <- plot_projection(d, 4326, "EPSG:4326 Latlon") p3 <- plot_projection(d, 54030, "ESRI:54030 Robinson") p4 <- plot_projection(d, 54099, "ESRI:54099 Spilhaus") p1 + p2 + p3 + p4
November 20, 2025 at 10:13 AM
mackaszechno.bsky.social
Left a message on the moon too as a bit of a running gag...
November 18, 2025 at 9:40 PM
mackaszechno.bsky.social
And a view of the moon in QGIS as a WMS layer courtesy of webmap.lroc.asu.edu. Instructions at lunaserv.im-ldi.com/getting_star....
November 18, 2025 at 9:18 PM
mackaszechno.bsky.social
Planning out interrelations between scenes, chapters, books, and associated notes.
#30DayMapChallenge | Day 18 - Out of this world #obsidian
full scale graph mid-level, I guess quite close
November 18, 2025 at 9:14 PM
mackaszechno.bsky.social
Okay, phoning this one in as got bored of all the necessary setup required to get Python spatial working, so kept to the absolute basics. Undersea cables in Python (image1) and with R (image2). Two lines of code in each.
#30DayMapChallenge | Day 17 - A new tool #python
(code in alt)
# https://py.geocompx.org/01-spatial-data import pandas as pd import shapely import geopandas as gpd gdf = gpd.read_file('./shapefile/cable.geojson') gdf.plot() library(sf) library(mapgl) cable <- read_sf("./shapefile/cable.geojson") maplibre_view(cable, column = "slug", legend = FALSE)
November 17, 2025 at 3:48 PM
mackaszechno.bsky.social
Zoomed out.
November 16, 2025 at 9:08 PM
mackaszechno.bsky.social
Reusing ONS Output Areas with pop density data from Day 8 to create cells colour-coded to pop density. Also could be considered a 10-minute map as used only maplibre_view.
#30DayMapChallenge | Day 16 - Cells #rstats
(code in alt)
library(sf) library(dplyr) library(mapgl) # read in ONS population densities by Output Areas d <- read_sf("./shapefile/Output_Areas_2021_WSx_pop.shp") # make border polygon border <- d |> summarise() # make grid of squares to cover border polygon grid <- st_make_grid(d, cellsize = 1000) |> st_as_sf(crs = 27700) grid <- grid |> mutate(id = row_number()) # get centroid for each grid square pts <- st_centroid(grid) # assign observation from ONS OAs to centroids ons <- d |> select(Obsrvtn) pts1 <- pts |> st_intersection(ons) pts1 <- pts1 |> rename(`Population density for West Sussex` = Obsrvtn) # produce map maplibre_view(pts1, column = "Population density for West Sussex", n = 10)
November 16, 2025 at 9:06 PM
mackaszechno.bsky.social
Now with megawatts as circle radius (still very basic).
November 15, 2025 at 10:11 PM
mackaszechno.bsky.social
Randomly found myself face-to-face with this culprit today following the wife's decision that we take the bus to Littlehampton during the unseasonably warm weather. Had to get a photo or two before leaving 🤣
Littlehampton Miniature Railway - end of the line (west-end) Littlehampton Miniature Railway sign
November 15, 2025 at 9:43 PM
mackaszechno.bsky.social
Okay in the end the simplest solution was to add in extra columns to the dataset, one for each decade ie S1920 for the 1920s, and then use set_filter to modify the original map.

Each section then becomes...
on_section("map", "1920", { maplibre_proxy("map") |> clear_markers() |> set_filter("all", filter = list("==", "S1920", TRUE)) |> fit_bounds(d) })
November 15, 2025 at 9:06 PM
mackaszechno.bsky.social
Maplibre in storymap mode showing changes in fuel-types for power generation in the UK.

Kudos to @kylewalker.bsky.social for the {mapgl} package and examples 😁

Data: www.gov.uk/government/c...

#30DayMapChallenge | Day 15 - Fire #rstats
November 15, 2025 at 8:57 PM
mackaszechno.bsky.social
The exception that proves the rule?
November 14, 2025 at 10:36 PM
mackaszechno.bsky.social
Power Stations operational before or on 1920 in the UK.
November 14, 2025 at 9:21 PM
mackaszechno.bsky.social
Okay, sometimes it feels like that episode of Angel when Wes has to translate four languages at once to arrive at the vague answer but I will never tire of how beautiful maplibre is and how accessible it is because of {mapgl} ❤️
library(sf) library(dplyr) library(readr) library(units) library(mapgl) d <- read_rds("electricity1920_2020.rds") maplibre(style = carto_style("voyager")) |> fit_bounds(d) |> add_circle_layer(id = "bob", source = d, popup = get_column("start")) |> set_filter("bob", filter = list("<=", "start", set_units(1920, "year")))
November 14, 2025 at 9:16 PM
mackaszechno.bsky.social
Never really liked the old-style anonymous functions using ~ and . and .x. New style makes much more sense to me.
#rstats
# replace specific character with na # fecking bracket placement important # 1. old style mutate(across(site:fuel, ~ifelse(. == "[x]", NA, .))) # 2. new style mutate(across(site:fuel, \(x) ifelse(x == "[x]", NA, x))) # and repeated using set_units on character column d |> mutate(across(start:end, \(x) x |> as.integer() |> set_units("Year")))
November 14, 2025 at 11:44 AM
mackaszechno.bsky.social
Shinyapp using OSM data showing.
#30DayMapChallenge | Day 14 - Data challenge: OpenStreetMap #rstats
(key:values in alt)
{osmdata} Traffic Signals – key = "highway", value = "traffic_signals" Roundabout – key = "junction", value = "roundabout" Bus Station – key = "amenity", value = "bus_station" Train Station – key = "railway", value = "station" Education – key = "amenity", value = c("university", "college", "school") Medical – key = "amenity", value = c("hospital", "clinic", "doctors")_ Blue Lights – key = "emergency", value = "ambulance_station"; key = "amenity", value = c("fire_station", "police") Retail – key = "landuse", value = "retail" Commercial-Industrial – key = "landuse", value = c("industrial", "commercial") Leisure – key = "leisure", value = c("sports_centre", "stadium", "recreation_ground") Also… key = "highway", value = "bus_stop"
November 14, 2025 at 9:37 AM
mackaszechno.bsky.social
Ah, so different tags in OpenStreetMap
Tag:tunnel=culvert and, possibly, name includes 'beck'... ie 'Clow Beck' which appears as waterway:river.

Try this link for becks: www.openstreetmap.org/search?query...
beck, darlington in OpenStreetMap query
November 13, 2025 at 2:22 PM
mackaszechno.bsky.social
Used in conjunction with the EA's Northumbria dataset...

Any use? If so, I'll tidy up and email you a copy.
November 13, 2025 at 11:37 AM
mackaszechno.bsky.social
Prime Meridian displayed against backdrop of all time zones. Plus image of a magnet I brought back with me from Greenwich a couple of years ago :)
#30DayMapChallenge | Day 13 - 10 minute map #rstats
(code in alt)
library(sf) library(dplyr) library(mapview) prime_meridian <- st_sfc( st_linestring( rbind( st_point(x = c(0, 180)), st_point(x = c(0, -180)) ) ), crs = 4326 ) |> st_sf() # https://www.naturalearthdata.com/downloads/10m-cultural-vectors/timezones/ tz <- read_sf("./data/ne_10m_time_zones/ne_10m_time_zones.shp") mapview(tz, zcol = "time_zone", alpha.regions = 0.2) + mapview(prime_meridian, color = "red") Magnet from gift shop in Greenwich Observatory
November 13, 2025 at 9:23 AM