31:T20fc,## Intro: why base maps One rarely creates a map from a blank state, with just a pen, a keyboard, or a tablet. Over the years, the professional cartographer will have gathered and developed her own collection of “shapes” to use as base maps, to tweak and adapt—depending on the topic and the data. These shape files (in the general sense—they often are in the GeoJSON or TopoJSON format rather than the shp format of ESRI) must include, for example, borders for every county, country, region, or continent, that she wants to draw. Files for base maps must have a level of detail (or precision) that matches the way we want to treat them visually. They must have the same granularity as the dataset (you can't draw city population on a map of countries, and vice-versa). Maintaining these files is a lot of work, since the actual shapes change every so often in time. As history marches on, borders fluctuate, countries are liberated from their colonizers, fall under another country’s domination, split, disappear, merge… In some parts of Europe, a person might never have moved out of the village in which they were born, yet have lived in four or five countries. The wheel of history also turns in a more pacific way when the census gives an identity to new settlements. Or, when agencies that collect statistics recompose existing boundaries to fit changing patterns of population distribution. Or, when politicians manage to [gerrymander](https://en.wikipedia.org/wiki/Gerrymandering_in_the_United_States) electoral districts. Many a cartographer—from beginner to pro—now relies on various files found around the Web. More savvy cartographers will be part of online communities and help each other tracking the correct file for the exact problem. But they know that any of the base files that you can find this way have quirks and issues that you might not be aware of, but that will make your map inaccurate, unbalanced, unfair, insensitive… And that is… unavoidable. There is no way that such files can be all-purpose accurate. _The map is not the territory_ (Alfred Korzybski), they are only a certain interpretation of it. One cause for interpretation is that the shapes need to be generalized (simplified), since we just need to represent them at a reduced scale on a tiny screen or sheet of paper. The cartographer might want this generalization to convey a certain aesthetic, [emotion](https://www.nytimes.com/interactive/2014/09/19/travel/reif-larsen-norway.html), and maps are more or less beautiful depending on how and by which person or what algorithm, and with which intent they were generalized. The [Natural Earth Data](https://www.naturalearthdata.com/) project has done a great job with this, and provides shapes at three levels of generalization (110m, 50m and 10m) with lines that render very neatly. Furthermore, they are shared and distributed under a good license, which explains why they have become the go-to for a lot of examples (in particular with TopoJSON’s derivation as the [world-atlas](https://github.com/topojson/world-atlas) package). However, in many cases these are not suitable to create maps, because of technical glitches, or because of certain decisions baked in over so-called _disputed_ political boundaries, that you might not share as a cartographer. _Cartography is inherently political._ “Political maps,” after all is the expression that designates maps “which show state and national boundaries and capital and major cities”, as opposed to physical maps which show mountains, lakes, etc. _Statistics_, in turn, are a distinct prerogative of the _State_—that’s even the origin of the word. With this, conflicts about territory become conflicts about maps. Some of these situations are “hot”, with ongoing military activity, and fast moving delimitations between influence zones; others seem frozen in time. The political frontier, as a single “line in the sand” that entrenches a clear-cut vision of territory and sovereignty, often clashes with the actuality of what people experience on the ground—through migrations, trade, work, marriage, studies, friendship, and all ways of living. It also often clashes with the law. As the anarchist geographer Élisée Reclus argued in 1868, _“borders are only artificial lines imposed by violence, war, the cunning of kings, and sanctioned by the cowardice of the people”._ Reclus’ ethical principles even made him reject mountains, rivers and seas as being “natural frontiers”: _“As for the so-called natural frontiers, those which are based on the topography of the land, we can understand them at a pinch: but even they have no more right than the previous ones to form an obstacle between populations, nor do they have the right to serve as a basis for the organization of society. There is no natural frontier; the Ocean itself no longer separates countries.”_ However, once that’s said and understood, the cartographer is still facing the same concrete problem. Loathe those frontiers all she wants, she still needs to draw some _areas_ for the statistical units that bear the numbers that she needs to represent. The role of an independent cartographer is not to decide where frontiers between countries lie—this, at best, is for diplomats—, but to help readers get a sense of how a phenomenon is spatially distributed. And to do so, she needs shapes, and these shapes define frontiers. And frontiers will be contested, sometimes with harsh words and sometimes with heavy artillery. So, even when she wants to draw a map of health issues, or sports, or musical tastes, or natural seasons… the cartographer will be caught between warring parties. Newspapers will be censored, books will be trashed, ambassadors will be summoned and trolls will troll if maps don’t follow this or that guideline. There is no way to escape it, because it’s the world we live in. Fixing this mess requires fixing the world. (One day will come!) What we can do in the meantime, is to “do no harm”, or at least try and do as little harm as possible. Back to our shape files problem: the approach to that question adopted by Natural Earth is to follow [a policy on “disputed boundaries”](https://www.naturalearthdata.com/about/disputed-boundaries-policy/) that we can summarize as the “boots on the ground” principle (BOTG). In short, when two countries lay opposite claims over a region, that region is shown as a part of the country that controls it—militarily. (Natural Earth now tries to address the issue with “countries [points-of-view](https://www.naturalearthdata.com/downloads/10m-cultural-vectors/)” that you can download separately, but these are only available at the high-precision scale 10m, and sparsely documented.) Of course, unless we want to make maps of a fantasy world, we still have to follow some principle of reality. But the BOTG principle, in my view, is not doing “as little harm as possible”. It is unfair, people who have no power have no voice, irrespective of justice and other principles. And since at the end of the day it yields to the strongest army, BOTG gives precedence to [military force](https://github.com/nvkelso/natural-earth-vector/issues/812) over diplomacy. This is why for our practice as cartographers (with [visionscarto](https://visionscarto.net/)), we’ve had to fork and made amendments to Natural Earth’s shape files. Some of the changes are of a technical nature (such as fixing topology issues); others are of a more artistic nature adding a few islands that were generalized out too sharply to our taste. More importantly, we tried to follow the UN conventions instead of boots on the ground… except when it didn’t make sense, which is why we’ve kept Natural Earth’s three self-governing entities with an undecided UN status: Somaliland, Taiwan, and Kosovo. _Making maps is hard_, and as much as we now have access to tools that make it a bit easier, the fundamental difficulty of drawing the correct shapes remains. By sharing and documenting our files (and our rationale), we hope to help some of you find a proper base for your “casual” mapping activities—for example, when you learn how to technically draw a map. We do not claim that these files are accurate, or better for everyone. They are one possibility, one choice, a mix of rationality and personal preferences, over many. 32:T44a,function map1({ countries, center = [0, 0], radius = 90, caption = "World", width = 230, highlight = [""], dx = 0, dy = 0, height = width === 230 ? width - 9 : undefined } = {}) { return Plot.plot({ marginRight: 9, projection: { type: width === 230 ? "mercator" : "equal-earth", domain: d3.geoCircle().center(center).radius(radius)() }, width, height, color: { domain: highlight.map(String), scheme: "pastel2", unknown: "#efefef" }, marks: [ Plot.geo(countries, { fill: (d) => String(d.id), stroke: "#ccc", title: (d) => `${d.properties.name} (${d.id}). ${d.properties.a3 ?? ""}` }), countries !== countries110m ? [] : Plot.text( countries.features, Plot.centroid({ filter: (d) => highlight.find((h) => h == d.id) !== undefined, text: (d) => d.properties.name, dx, dy, stroke:"white", fill:"currentColor" }) ), Plot.frame() ], caption }); }33:[] 2f:{"title":"Visionscarto’s “World Atlas”","cells":[{"id":73,"value":"# Visionscarto’s “World Atlas”\n\n## Documentation for our modified topojson files\n\nInitially intended for [Visionscarto](https://visionscarto.net/), these files are available to all. We derived them from Natural Earth’s 110m and 50m cultural vector files, but added several modifications, in particular to reflect the United Nations’ position on the integrity of Ukraine and the non-self-governing territory of Western Sahara. Since nothing is simple in the world, these files also include three self-governing entities with an undecided UN status: Somaliland, Taiwan, and Kosovo. This notebook documents the changes we made, with respect to the original files published by Natural Earth.\n","pinned":false,"hidden":false,"mode":"md"},{"id":974,"value":"_This is a long intro, please feel free to jump directly to [the files](#files)!_\n\n","pinned":false,"hidden":false,"mode":"md"},{"id":0,"value":"$31","pinned":false,"hidden":false,"mode":"md"},{"id":90,"value":"---\n## The files\n\nThe shapes we share here are variants of TopoJSON’s [World Atlas](https://github.com/topojson/world-atlas), themselves a derivation of Natural Earth’s [110m](https://www.naturalearthdata.com/downloads/110m-cultural-vectors/) and [50m](https://www.naturalearthdata.com/downloads/50m-cultural-vectors/) “cultural vectors” base maps. You are free to use and modify them as you wish, since they are in the public domain, following [Natural Earth data’s terms of use](https://www.naturalearthdata.com/about/terms-of-use/). You can for example download them from the “Files” panel of this notebook, or import them into another notebook. Any mapping software that can read TopoJSON should be able to ingest them.","pinned":false,"hidden":false,"mode":"md"},{"id":1026,"value":"world110m","pinned":true,"hidden":false,"mode":"ojs"},{"id":1031,"value":"world50m","pinned":true,"hidden":false,"mode":"ojs"},{"id":986,"value":"The files have been optimized down to 49.3kB (gzipped, for precision 110m) and 274kB (gzipped, for precision 50m).\n\n","pinned":false,"hidden":false,"mode":"md"},{"id":1015,"value":"To use them in a notebook, you can copy them directly, or import them from our [geo notebook](/@visionscarto/geo)—which contains many other files, formats, and tools.","pinned":false,"hidden":false,"mode":"md"},{"id":417,"value":"import {world110m, world50m} from \"@visionscarto/geo\"","pinned":true,"hidden":false,"mode":"ojs"},{"id":448,"value":"The files are also available from NPM:\n~~~js\n const url = \"https://cdn.jsdelivr.net/npm/visionscarto-world-atlas@0.1.0/world/110m.json\";\n const world = await fetch(url).then(d => d.json());\n const features = topojson.feature(world, world.objects.countries);\n~~~","pinned":false,"hidden":false,"mode":"md"},{"id":988,"value":"And the project is tracked on Github:\n> [`https://github.com/visionscarto/world-atlas`](https://github.com/visionscarto/world-atlas)","pinned":false,"hidden":false,"mode":"md"},{"id":397,"value":"---\n## Examples\n\nThe examples below use the [topojson-client](https://github.com/topojson/topojson-client) library to convert the TopoJSON format to GeoJSON, which is then consumed by Observable Plot’s [geo](/@observablehq/plot-geo) mark. The files can be used by any other software that consumes TopoJSON.","pinned":false,"hidden":false,"mode":"md"},{"id":131,"value":"countries110m = topojson.feature(world110m, world110m.objects.countries)","pinned":true,"hidden":false,"mode":"ojs"},{"id":445,"value":"Plot.geo(countries110m).plot({projection: \"equirectangular\"})","pinned":true,"hidden":false,"mode":"ojs"},{"id":222,"value":"countries50m = topojson.feature(world50m, world50m.objects.countries)","pinned":true,"hidden":false,"mode":"ojs"},{"id":744,"value":"Plot.plot({\n width: 928,\n height: 928*0.45,\n projection: {\n type: d3.geoInterruptedHomolosine,\n inset: 2,\n domain: { type: \"Sphere\" }\n },\n color: { scheme: \"rdylbu\" },\n marks: [\n Plot.graticule({clip: \"sphere\"}),\n Plot.geo(countries50m, {\n fill: d3.randomLcg(42),\n stroke: \"currentColor\",\n strokeWidth: 0.5,\n title: (d) => `${d.properties.a3}. ${d.properties.name}`,\n clip: \"sphere\"\n }),\n Plot.sphere()\n ]\n})","pinned":true,"hidden":false,"mode":"ojs"},{"id":93,"value":"---\n## The changes\n\nHere is a list of changes we made with respect to the original Natural Earth files. They cover various categories: boundaries, technical changes, and aesthetics.","pinned":false,"hidden":false,"mode":"md"},{"id":101,"value":"#### Western Sahara\n\nMorocco occupies a large part of Western Sahara since the [green march](https://en.wikipedia.org/wiki/Green_March) in November 1975, and claims it as part of Morocco. The United Nations disagree. This file shows Western Sahara with the UN point of view.","pinned":false,"hidden":false,"mode":"md"},{"id":135,"value":"map({ center: [-9, 29], caption: \"Western Sahara\", radius: 9, highlight: [732, 504], dx: -5 })","pinned":false,"hidden":false,"mode":"ojs"},{"id":112,"value":"#### Crimea and Eastern Ukraine\n\nRussia has invaded Ukraine in 2014, then again in 2022. Moscow has made annexation claims over Crimea and the East of Ukraine. As I write this, the war rages on. This file shows Crimea as part of Ukraine.","pinned":false,"hidden":false,"mode":"md"},{"id":129,"value":"map({ center: [31, 47.5], caption: \"Crimea\", radius: 6.8, highlight: [804, 643], dx: 14 })","pinned":false,"hidden":false,"mode":"ojs"},{"id":178,"value":"#### Israel/Palestine\n\nWith Israel’s continued occupation of Palestinian territories, making a map of the Middle East has always been challenging; Natural Earth’s map defaults—I think?— to the borders of 1949 (!). Our only change is to make the Gaza strip a bit wider on the 110m file, so it’s not removed altogether by the simplification algorithm.","pinned":false,"hidden":false,"mode":"md"},{"id":176,"value":"map({\n center: [35, 32],\n caption: \"Israel/Palestine\",\n radius: 3,\n highlight: [275, 376],\n dy: 10,\n dx: 2,\n show: [\"110m\"]\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":144,"value":"#### Country codes and names\n\n- **Kosovo** hasn’t yet received an official ISO code. It is widely anticipated that it will be **XKX**.\n- In 2019, Macedonia accepted to change its name and is now called **North Macedonia**.\n- In the 2024 version, we also fixed the code for Palestine to be **PSE** (instead of PSX), and the code for Åland to be **ALA** (instead of ALD).\n\nA few other situations of note:\n- **Somaliland** has de facto been independent from Somalia since its referendum in 2001; we give it the code **SOL**.\n- **Taiwan** (“Republic of China”) is embarrassing to the United Nations, since only 13 countries recognize it as a country independent from Beijing (the “People’s Republic of China”), which it has been since 1945. The shape is separate since it’s an island, and it has the **TWN** code so people can choose to give it a specific color.","pinned":false,"hidden":false,"mode":"md"},{"id":259,"value":"#### Topology\n\nThe World Atlas files have several topology errors, which you only see if you try to do specific computations on the data. What we fixed:\n* In the 110m file, North Korea shared a border with itself; we removed that dangling arc.\n* The independence of South Sudan from Sudan introduced new tripoints at the borders—one with the Central African Republic (CAF), one with Ethiopia. For topological consistency, we added the tripoints to the arcs describing the CAF border and the Ethiopian border.\n* In 50m, the simplification of an island ended up giving it the wrong winding order.","pinned":false,"hidden":false,"mode":"md"},{"id":236,"value":"#### Aesthetics\n\nThe **Kuril Islands** run from the north of Japan to the south of the Kamchatka peninsula in Russia. Natural Earth 100m doesn’t draw them because of area simplification, but it seems to make more sense (cartographically speaking) to show a handful of the largest ones.","pinned":false,"hidden":false,"mode":"md"},{"id":226,"value":"map({\n center: [146, 48],\n caption: \"Kurils\",\n radius: 16,\n highlight: [392, 643],\n dx: 12,\n dy: -28,\n show: [\"110m\"]\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":414,"value":"---\n_supporting code_","pinned":false,"hidden":false,"mode":"md"},{"id":210,"value":"map = ({show = [\"110m\", \"50m\"], ...options}) =>\n htl.html`

\n${!show.includes(\"110m\") ? \"\" : map1({\n ...options,\n countries: countries110m,\n caption: `Revised file, 110m`\n})}\n${!show.includes(\"110m\") ? \"\" : map1({\n ...options,\n countries: wa202_110,\n caption: `World Atlas, 110m`\n})}\n${!show.includes(\"50m\") ? \"\" : map1({\n ...options,\n countries: countries50m,\n caption: `Revised file, 50m`\n})}\n${!show.includes(\"50m\") ? \"\" : map1({\n ...options,\n countries: wa202_50,\n caption: `World Atlas, 50m`\n})}`","pinned":true,"hidden":false,"mode":"ojs"},{"id":123,"value":"$32","pinned":true,"hidden":false,"mode":"ojs"},{"id":303,"value":"wa202_50 = d3\n .json(await require.resolve(\"world-atlas@2.0.2/countries-50m.json\"))\n .then((d) => topojson.feature(d, d.objects.countries))","pinned":true,"hidden":false,"mode":"ojs"},{"id":310,"value":"wa202_110 = d3\n .json(await require.resolve(\"world-atlas@2.0.2/countries-110m.json\"))\n .then((d) => topojson.feature(d, d.objects.countries))","pinned":true,"hidden":false,"mode":"ojs"},{"id":774,"value":"d3 = require(\"d3@7\", \"d3-geo-projection@4\")","pinned":true,"hidden":false,"mode":"ojs"},{"id":1053,"value":"Font: Atkinson Hyperlegible.\n\n\n","pinned":false,"hidden":false,"mode":"html"}],"files":"$W33","stdlib":"1","titling":"auto"}