In the news: South Pacific Sandy Island 'proven not to exist'

BBC: South Pacific Sandy Island 'proven not to exist'. It still shows on my Google Maps though.

Head and shoulders above the rest... well, one shoulder anyway

(This is an expanded version of a review for a Fujian road atlas and other atlases from the same series on Amazon.cn)

If you travel, you need maps. Even more so, if you travel "off the beaten track", bicycling or hiking. If you like maps in general, traveling also gives you pretext to shop for maps, and to learn what kind of maps are published in each country you visit.

China travelers are actually supplied with maps pretty well. Every big book store (what they'd call a 书城, "Book City" - there are at least a couple in every provincial capital) would have a fairly well-stocked map department; smaller versions are found in local book stores (mostly, the Xinhua chain) in prefectural and most county-level cities. Besides the city maps, of which I am not talking now, what you need for the countryside travel are provincial atlases. A typical atlas, selling for about CNY10 (less than $2), would have perhaps 30 pages, with one map (a two-page spread) per a prefecture-level unit (there are about a dozen or more in each province). The scale is usually around 1:500,000 - 1:1,000,000 in the more densely populated areas, or 1:2,000,000 or even less in the desert western areas. These are basically detailed road maps, primarily targeted to motorists; but for a bicyclist they are more convenient (and usually more detailed, too) than thick glossy national or regional highway atlases. They show all towns and townships (镇 zhen, 乡 xiang), and a few more major villages (村 cun ); railways and stations; all National (G) and state (S) and some county (X) roads, as well as major tourist attractions. The legend is usally all in Chinese, of course; but even if you don't speak the language, the map vocabulary is not that hard to learn. (You probably will need most of it to read road signs anyway...).

Some of the same publishers also produce provincial wall maps with essentially similar scale and content. Some travelers like them more (see Marian Rosenberg's article on Star Maps Publishing), but I'd rather pack an atlas into my bag, and get a map to hang on a wall.

Getting an atlas for the province you're currently in usually is not problem, as most good bookstores in the province would carry at least one edition. For out-of-the-province atlases, it is more of a hit and miss: a major "Book City" may have 2-3 competitors' atlases for the "home" province, and maybe also atlases for a dozen or more other provinces, but it's somewhat hard to predict which ones. If you have time and a fixed address in China, ordering from Amazon.cn is often your best bet. As of the early 2012, Amazon.cn take cash on delivery for many books, so you don't even need a Chinese credit card.

Most publishers update their atlases every year or two; still, some may be rather out of date, despite the purportedly recent publication date. And yes, occasionally you may encounter a "fantasy road" (in reality, a river of mud, or a construction site) or a "fantasy bridge" (in reality, a ferry). Occasionally, there are bigger bloopers: I've seen a few maps of this kind that manage to show Lüshun, Liaoning without its famous harbor. Hmm, maybe fighting the Battle of Port Arthur in was a mistake?

There are several competing publishers some with multiple maps series - Star Map Press (星球地图出版社)， SinoMaps 中国地图出版社， Dizhe Chubanshe 地址出版社 (i.e., Geology Publishers) with their Dipper （北斗） series, and a few others. Which ones to buy, if you have choice?

Hey, my cheaper atlas said this was National Highway 209!

The best series of provincial atlases that I have seen is the one published by Star Map Press, and labeled 军民双用 (Jun-min liang yong, "Civil and military use"). They of course are nothing like real military topographic maps, but have better scale than other similar maps (1:300,000 for most provinces) and correspondingly better level of detail.

They may not be as good as you can see on an on-line maps, such as on maps.google.com or ditu.sogou.com - but that assumes that the online maps have a good coverage of the region of interest - which is not always the case! Still, the rivers and the switchbacks on the roads allow you to get some idea of what the terrain is like. These atlases are a bit thicker and more expensive than those of the competitors, or than Star Map's own "lightweight" provincial road atlases series (中国分省公路丛书) (maybe CNY15-18 instead of CNY9-12), but are certainly worth the trouble getting if you're cycling, hiking, hitchhiking (usual disclaimer: not recommended!), or making use on "very local" buses.

The reader should be warned that atlases from this Jun-min liang yong series are pretty hard to come by in shops, though, especially outside of the province in question. Exceptionally, I saw an almost complete selection of them in Quanzhou's "Book City" (the underground one, in the park near Quanshan Gate), but elsewhere you'd only see a couple of them at best, often for some random province you don't need. Amazon.cn, however, sells pretty much all of them (and at a discount to the cover price, too!). For example, here's one for Yunnan， or for Fujian. And here's Tibet and Xinjiang

Star Map Press' Jun-min liang yong atlases are to be commended for their economical use of space: instead of allocating one page per prefecture, and wasting a lot of paper this way, their maps correspond to squares of a province-wide grid. (There is still a table of content, which identifies pages by the major cities located there). The publisher also avoids the annoying "feature" present in the Dipper provincial atlases, which are padded with rather unnecessary single-page maps of adjacent provinces and their capital cities. (Which results in a huge duplication of space if one buys several provinces' atlases).

Overall, Star Map Press' Jun-min liang yong product truly stands out head and shoulders above the competition. Well, maybe one shoulder: despite generally better quality than the rest of the breed, this series has some minor, but annoying, shortcomings, even in comparison with more commonly available atlases:

It sort of helps if your map shows the same villages that the local bus schedules do, doesn't it?

• Somewhat strange choice of populated places to show. It is actually a fairly common thing with online maps (like Google Maps): when you zoom in, the map shows smaller villages, but sometimes stops showing labels for some larger populated places. This is not much of an issue with an online map, but when you see it happens when you switch from a "worse" to a "better" paper map, it's annoying. For example, on the road (S207) from Xiazhai to Xiaoxin in Pinghe County, the less detailed SinoMaps atlas shows five villages (彭林 Penglin，长汀 Changting，枫埔 Fengpu，旧楼 Jiulou，厝丘 Cuoqiu）, pretty much all of which are indeed landmarks: they are signed on the ground, and listed in the schedule of a local bus (which, helpfully, is posted at each stop along the road). On the other hand, Star Maps' more detailed Jun-min liang yong atlas choose to show a completely different set of villages along this route - of which only one appears on the locally posted bus schedules!
• Similar to the competitors' products, these maps do show some of the railways and highways that are under construction (labeled as such). However, they certainly could show more of them - surely such construction projects usually take a few years, and detailed plans ought to be available to the mapmakers. It is particularly annoying that even when these maps show railways under construction, they don't indicate the location of the stations - which is something that the traveler need the most. The Fujian atlas doesn't even mark stations on the new Fuzhou-Xiamen high-speed railway, even though it must have been about to open when the atlas was printed.
• Rather strangely for mass-market maps (perhaps, taking the series name too seriously?) the Jun-min liang young series is quite stingy with showing the location of various tourist attractions. For example, out of the thousands of the Fujian Tulou, only 10 individual sites are specifically entered on the UNESCO World Heritage list. Surely it is appropriate to mark them all (as some of the cheaper maps do)? But Jun-min only show three of them. Similarly, among Fujian's famous bridges, Jun-min shows the Anping Bridge but not the Luoyang Bridge - even though most competitors usually show both. Such omissions are unfortunate, because many road signs show directions and distances to such objects, so being able to locate them on the maps is quite beneficial even for travelers who aren't interested in these attractions per se.
• The Jun-min liang young atlases probably show more of the county routes (the "X-series" roads) than other atlases, but they never label them. Some writers (see e.g. Marian Rosenberg's article) claim that no maps label them, but this is incorrect: for example, SinoMaps' Fujian road atlas (福建省公路里程地图册, from their 中国分省公路里程地图册系列 series) clearly shows and labels them all. These roads are important for country side travel, and at least in some counties they are signposted (even if not always consistently). There is no reason why a 1:300,000 map should omit these labels!

Printing Google / Sogou maps

Have you had this experience? You've centered and zoomed the map of an interesting area on Google Maps, or a similar site (e.g., maps.sogou.com) and wanted to print it. Some map sites actually have a "print" button, or you can just use the "print screen" / "screen capture" button of your operating system's window manager to save the image into a PNG file to print it later (directly, or after including it and the caption text into an HTML / PDF / LaTeX file). But, whatever you do, chances are that the printed image will be almost un-readable: Google, Sogou, and other map sites optimize the image for on-screen viewing, and printing is just an afterthought.

There is, however, an easy way to improve the legibility of the map image. The idea is, an image is hard to read because there is not much contrast to it: both the brightest pixels (the background) and the darkest pixels (text and lines) are merely shades of grey (or, generally, some intermediate RGB colors), rather than good old printable black and white.

So the solution for improving the printable quality is, basically, as follows: figure what color you want to convert into black, what color you want to convert into white, and linearly transform all colors in between.

There are of course plenty of image manipulation tools on the market. The one I happen to use is ImageMagick, and sure enough, it has a command for it.

To find the colors to be made into black and white, you probably can zoom in on your screenshot image in some image editing program, and take a look at what the RGB values for the darkest pixels (parts of text, or lines showing roads etc) are, and what the RGB values for the brightest pixels (the not-quite-white background). In my case, I did it with cropping out a small section of the image with "convert -crop", and then querying the range of colors with "identify -verbose". Having chosen the colors this way, all you need to do is a single command:

convert original-map.png -level-colors 'rgb(144,144,10),rgb(241,238,232)' printable-map.png

Enjoy!

Before	After

Emperor Zheng Ho, meet Comrade Kirov!

This review is from: Fra Mauro's Map of the World (Terrarum Orbis)

This product consists of two parts: a CD-ROM that contains a high-resolution scan of the famous Fra Mauro Map, and a huge book that contains transcripts of the labels on the map, their English translations with comments, and a couple hundred pages worth of articles about the various aspects of the map.

This map, created by the Venetian monk Fra Mauro ("Brother Mauro") ca. 1450 is, of course, famous, being perhaps the best example of the European map-making on the cusp of the Age of Discovery. A copy of this map was actually made for the Portuguese king of the day, so you could imagine Bartolomeo Dias and Vasco da Gama studying it in preparation for their pioneering voyages to[ward] Asia; Columbus may have looked at some similar map, wondering how far its western end really was from its eastern end.

Basically, it may be viewed as the last great maps created before two great revolutions - one in maritime technologies, allowing the Europeans to sail all over the world, and the other in book reproduction (the [re-]invention of printing in Europe) - revolutionized European cartography and led to the appearance of works by Mercator, Ortelius, etc. in the 16th century.

Fra Mauro's map is famous, in particular, because it appears contains a lot of geographic information that is (more or less) correct (i.e., not entirely fantastic), but about whose provenance we don't know anything. That is, the good Brother Mauro must have gotten it *somewhere*... but whatever his sources were, they just have not survived to our day. As one example, it shows a fairly reasonably-looking continent of Africa, one you can sail around, even with a Madagascar of sorts (called "Diab") south-east of it... all this a generation before Dias and da Gama.

Now, what do this book and CD give you? There is actually a small printed map coming in a pocket with the book, but since the original map was *huge*, you can't read much of anything on that small map, beyond the names of countries and major regions. The CD has of course high-resolution data - much better than a freely downloadable scan file available e.g. via Wikimedia Commons. But the data are not in some standard portable format such as JPEG or PDF; rather, they are only readable by some specialized program, for the Microsoft Windows operating system, included on the disk. So this is rather inconvenient, if you are used to view images in some general-purpose image file viewer. For example, the program's interface allows you to scroll through the image, or to zoom in on a small piece of it, but you can't zoom out - which means that you can't see more than a very small part of the entire map on the screen at once (unless, that is, you have a really huge screen, or can control your monitor's resolution to be 10,000 by 10,000 or something...) The "transcription" feature, however, is quite convenient: you can click on any text on the map (label, caption) and to see the transcription of the Italian original, or its English translation, as plain text in a pop-up window. A great help for someone not too comfortable with Mediaeval handwriting. (Although I'd say that the text on the map is fairly legible as it is).

The luxuriously printed book itself is... useful, although one wonders if the publisher could come up with a more compact format for what is, basically, a huge dictionary of transcribed and translated map labels, which now happens to be formatted so that most of the pages are 80% blank space. (This is why the thing probably weighs 5 pounds if not more).

The articles, written mostly by the curator of the maps at the Venetian Library, with a contribution from an art historian, are certainly very valuable for anyone interested in the "life story" of this particular map, and in Mediaeval cartography in general. The author tries to probe the problem of "missing sources", and come up with interesting ideas here and there. (Arab manuscripts; interviewing visiting Ethiopian and Russian churchmen...). Overall, he feel that "no longer identifiable oral accounts", mentioned by Fra Mauro himself (p. 33), were an important source in the creation of the map, so, in a sense, many critical "sources" of it probably never existed as written documents; and we probably should agree with him.

At the same time, it seems that quite a few silly statements, or at least unexpected omissions, are hidden in plain view among the books' annotations. For example, the famous seven 15th-century Chinese expeditions to the Indian Ocean are described as being "ordered by emperor Zheng Ho" (item 19, pp. 178-179). (Hint 1: the emperors at the time were Yongle and Xuande; Zheng He was the eunuch in charge of the fleet. Hint 2: with Chinese names, you either use the modern Pinyin transcription [e.g. Zheng He] or the more traditional Wade-Giles [Cheng Ho], but try not to mix the two, at least not in the same name). Mozhaisk (the center of a small Russian principality west of Moscow) is described as having become "famous for its resistance of the Mongols in 1237-40" (pp. 91-92). (Hint: the small city famous for its resistance to the Mongols during that invasion was Kozelsk; I don't believe there was anything ever said about Mozhaisk in that regard). While "Provincia Mordua" is absolutely correctly linked to the Mordvin (Mordva) people, it is quite amusing to read that the nearby "Provincia Quier" "suggests a reference to the city of Kirov" (p. 93). (Hint: Kirov was so named after the pseudonym of a Soviet Communist leader, S.M. Kirov, in the 1930s; the place was known as Vyatka or Khlynov during the preceding 1000 years.)

Neither of these issues is a major one, and one cannot expect them all to be visible to any single author or editor; but if they are noticeable even to a casual reader, I wonder what an expert in the historical geography of a particular region would say if asked to proofread the chapter on the area of his expertise. The book probably could have benefited from having such a proofreading done in advance of its publication. Nonetheless, even as it is, this weighty volume is a valuable addition for any research library with a budget to afford it and the shelf space to go with it.

The world's railways, all on one map

Bill Rankin, of Yale University, has a wonderful map of world's railways (2008) on his site http://www.radicalcartography.net , with some thoughtful comments on how it reflects on the economic history of the world's nations.

His site does not seem to provide a space for readers to comment, so I would like to do it here.

First of all, a good way to describe this map would be not simply as a map of world's railroads, but rather as a map of (almost) all railroads, of any type, that ever existed prior to ca. 1990.

There are two reasons for this situation.

(1) First, a map like this brings to the fore a major issue for creating any railway map: the adequacy of sources. It seems that sometimes it is not easy for a newly-built railway to get to maps (especially published in other countries); but it is even harder for a dismantled railway to disappear from maps! The most prominent example visible of Rankin's map is perhaps the Newfoundland Railway, closed for good in 1990, but still gracing many world maps with its presence. Less visible is the Kettle Valley Railway southern British Columbia. Its western section, abandoned in the early 1960s, is gone from most maps (including Rankin's), but the eastern section, physically dismantled in the early 1990s, keeps living in the world of maps, including this one.

As a more interesting example, a good section of Stalin's pet project, the Salekhard–Igarka Railway in the north of western Siberia - namely, an east-west section across the northern part of Tyumen Region - actually saw trial service for a short time before the incomplete rail line was abandoned in 1953, and some western maps at the time showed the western half of this Arctic railway as operational. (I own a 1950s-vintage Rand McNally atlas which does so). Although that railway did not take root in the world of cartography, it somehow made its way into Rankin's masterpiece - presumably because the author tried to work with the most comprehensive sources available.

On the other hand, some of the more recently built railways - e.g., the China-Kazakhstan connection and the Turkmenistan-Iran connection, both activated in the early 1990s - aren't on the map, nor is the Kashgar railway in China's Xinjiang (also built in the 1990s; and it now has branches that only became operational after 2008...)

A peat railway, near Esterwegen, Germany

(2) Second, more detailed maps try to distinguish different types of railways, but in this comprehensive map it apparently was not possible. But a distinction we must make. For example, anyone used to looking at general-purpose maps of Russia and China can't but be amazed at some small but extra-dense clusters of rail lines which, according to Rankin's map, exist in the north-east of European Russia (roughly, the region between Arkhangelsk, Nizhny Novgorod (Gorkiy), and Yekaterinburg (Sverdlovsk), and in China's Heilongjiang and Jilin provinces (the northeastern part of what we used to call Manchuria). There is no mystery here, however: these are so-called logging railways (and, on rare occasions, peat railways). They are very different from standard, permanent, railways operating in these countries. A logging railway is typically a narrow-gauge railway, laid at the minimal cost, and only used to move logs from the woodlots to a mainline railway station or a sawmill. I understand that many sections of such railways are only used for a few years, and then, once that particular section of the forest has been cut, moved to a new location. A general-purpose Soviet or Russian map would never show them, but a topo map would... giving one probably an exaggerated idea of an area's rail density, as in reality many of the lines shown would not be actually operational anymore (see (1), above). Anyway, by now, I suspect, most of these logging railway live on cartographers' desks, as the timber industry has mostly switched to using trucks.

A sugar cane railway, near Proserpine, Queensland, Australia

A somewhat similar class of railways are sugar-cane railways, which apparently are responsible for the extraordinary visible density of railways in Cuba, Mauritius, parts of Taiwan, and even a rail line in Fiji. I am not sure if the map shows the sugar railways of Queensland - I think if it did, there would be a more dense network there. The sugar railways are, like forest railways, narrow gauge (sometimes, very narrow gauge), and only used to move a particular type of cargo. Unlike many forest railways, sugar railways are permanent, even if used for only a few weeks in a year, to transport cut canes to the nearby sugar mill. Their existence is made economical by the fact that sugar cane's yield, in tons per hectare, are much higher than those of most other crops grown on similarly large scales. I know that at least some such systems in Queensland are still very much in operation, and I've seen tracks for one of them in Fiji (no idea if they are ever in use, though), but I don't know about other countries.

(3) Dr. Ranking contrasts the American (meaning, here, US, Canada, and Argentina) railway network to systems elsewhere, by justly noting that "In the Americas, rail was the primary route by which crops reached global markets, and railways and settlement often went hand in hand." This is true in a sense - the remarkable dense networks covering the fairly sparsely populated agricultural regions of the Great Plains of the US and western Canada, central Argentina, as well as the grain-growing areas of Australia's Victoria and Western Australia are quite remarkable. But one should not forget, at the same, how extraordinary and short-lived phenomenon these networks were. They were, basically, brought to life by the particular situation that obtained during several decades in the second half of the 19th century through the early 20th century: the railway technology was much more efficient for moving grain and other cargoes overland than any other option available (which, in those areas mostly meant a horse and a cart).

It seems that soon after trucks became widely available and the governments subsidized highway construction, the less-used 50% (sometimes 90%) of this unique rail network became abandoned. The system reverted to a form more "sustainable" in the truck-and-road based economy: railways connecting major transportation hubs, with trucks moving grain and containers to the surviving high-volume stations. If one were to wipe all disused railways from the US/Canada/Argentina/Australia maps, and add all new railways that have been built over the last several decades in countries like China or Iran, I feel that the overall density of the operational railways today's in North America won't be that difficult from the comparable population-density regions of eastern Europe and Asia (Russia, Turkey, Iran, north-western China). And it will be only natural for high-population-densities regions in India and China to soon surpass the US and Canada with respect to the railway network density.

In a sense, I wish very much that all those rail lines that have been abandoned or converted to bike trails became active railways again, and all kinds of factories, warehouses, and other businesses - even post offices - moved back to the trackside districts, where they could ship and receive their freight directly by rail... but this isn't going to happen unless a very radical change in the different modes' transportation costs occurs.

As a comparison, in the late-19th century Russian Empire, the need to carry export grain to the sea ports was also a major stimulus for the railway construction boom. But the country - not even its most fertile areas in south-western Russia and Ukraine - never acquired as dense a system as Iowa or Saskatchewan had for a few decades in the early 20th century. Partly this indeed must have had to do with the availability of capital and the general level of the development of the national economy, partly with the simple transportation economics: with lower crop yields per acre, and much higher local population densities, I suspect that the net amount of grain to be exported from any 100-square-miles region of Ukraine or the Saratov Governorate was rather smaller than from a similarly sized piece of Iowa. As a result, they have not had to abandon all that much of railway trackage, except for in a few areas where the construction was moved by "mining or manufacturing", as Dr. Rankin says, and where that mining (or, less commonly, manufacturing) is gone. Among the lines still shown on the map, an example of this is a large part of the branch that goes east from Apatity and Kirovsk in Russia's Kola Peninsula.