Tag Archives: Google Earth View

Local Landscapes in OSM: the Shifting Economy of Mapping Place

A battle-cry of the ages goes out over the internet–Crowdsourced OpenStreetMap Trounces Google Maps!–in a banner New York Times headline so shocking it evokes the battle of David and Goliath of the Information Age.  The battle-cry hinted at the victory of not a company, but of the members of a mapping community who slung myriad shots from the slings of multiple local mappers in Sochi,most of  whom we only know by their monikers, to topple trust in the corporate behemoth’s maps at the same time as the whole world was watching.  And the basis by their cartographical diligence is evident in the juxtaposition of two visuals of the ski runs at Sochi that we’ve been watching on TV, since we can easily contrast it to the terrain on which we’ve already watched so many ski events–and use them to track the skiers whose slalom courses we might was better placed in space.  The on-site demand for maps is so acute that the divergence from the now nearly ubiquitous Google Maps engine is striking, and has caused a bit of a shockwave in the mapping micro-world–“In Sochi, Open Source Maps Beat Google”–as if this was a not-to-be-unnoticed Olympic event, when the “Wikipedia of Maps . . . has bested the corporate giant” at its own game.

The mapping of the Olympic games marked not only a symbolic victory, but a dent in what Adam Fisher aptly terms the “Google Maps-based ecosystem” that has grown out of the widespread reliance of a small and growing sector of the economy on the Google map making machine.  The reliance on map-providers is evidence that even now, in an age of satellite maps and extensive geovisualization, mapping is a marketplace and a business to best orient viewers to an image of the ground in some very interesting ways:  rather than exercising the dominance of organizing “all the world’s information” through a monopoly geo data, fast on the heels of some rather nasty accusations that some yahoos caught using Google IP addresses had set out to vandalize OpenStreetMaps from India, entering false information in their competing images to undermine trust in their accuracy by things from reversing the directionality of one-way streets and altering script in order to dissuade users of expectations for OSM’s accuracy.

As much as an act of random hooliganism, this is a sort of trade-war waged by undermining the credibility of the opposition–a huge change from the days when Google might have sponsored OSM’s annual conference.  Alas, no more, as the two generators of landscapes are at one another’s necks:  at the very time that Google is trumpeted as inevitably on its way to charting a road-map to world domination, cracks in its geo location armor appear.  And the interesting part was, in part, that rather than finding weaknesses or inaccuracies within the many photographs that Google engineers thread together to create a database of terabites that allow us to flip through an apparently seamless photograph of the world, the absences lay in the value of selectivity in labeling the sites, routes, and courses that athletes took, and their exact levels of elevation:  information absent from the outdated photographs Google used in its Earth View.

The two media of mapping provide essentially different landscapes–and a different access to the surroundings that they described.  These contrasting visions of landscape are so readily generated by the Map Compare tool designed by Geofabrik, which actively promotes the commercial use of open-sourced maps.  Designed to suggest the difficult selectivity and clarity that the Google offers on the environments it maps, the juxtaposition of the ski runs from the Google Earth view and OSM map are a triumph in the value of embedded signs and measurements of elevation, as well as potential routes of skiing on the slopes of Sochi.

.sochi-ski-satellite

Despite the authority of the Google Map view, the rhetorical power of this juxtaposition between these forms of mapping shows the extent to what Open Source allows in the recently built environment of the Olympic village:  one is immediately struck by the absence of the Skiing Pavilion on the less-often updated Google Maps views, where not only the routes of skiing on the local slopes are less often noted, but the structures built for the Olympics remain entirely absent, and the far limited points for attending to the landscape and its elevation. There is less data in the Google View, even when one goes outside Google Earth.

Indeed, the comparison and turned up many more signs of orientation than the blank spaces of Google Maps which so strikingly recall the “terra incognita” covered by cartouches in early modern maps of the New World or the icy uncharted regions of polar expanse.

ab4561346

Sochi Ski Center Mapped

What seem open areas marked by the faintest of trails or icy frost engage viewers in concrete ways in the OSM maps, raising questions about how they effectively invite us to see, and what constructions that they use to invite us to contemplate space from eight models of the same landscape.  In an age of the huge expansion of Geographic Information Systems and geovisualization, it is amazing not that different modes of mapping circulate–that’s to be expected–but that their contents will continue to be so diverse, or that the very multitude of information that they’re designed to visualize are available in so many competing models.  The ‘Map Compare’ function devised, in ways that recall the classic art history course’s comparison of two slides side by side each other, provide a slippy-screen template to compare any regions with the boast that the open-sourced OSM version will both be more complete and inclusive in its details, and a better commercial model for anyone interested in mapping any city, anywhere, on demand, on account of the multiple modifications OSM users have made.  The story of the more complete coverage of OSM is anything but new, but the recent focus on the demand for better maps in the Olympic games is a great news story, making the lack of information on Google’s map browser comparable to the shoddy quality of the ready-made rudimentary hotels in Sochi in quite potentially embarrassing ways.  Despite the copious Street View detail, Google’s maps of Sarajevo were lacked in the information and visual detail that OpenStreetMap could readily provide to its users.

open-versus-closed

Sarajevo OSM Google

For all the innovation of push-pins mapping cities, Google seems to have neglected the Bosnian and Herzegovinian landscape, and the very elements of regional mapping we need for detailed spatial orientation.   The mapping of green space,  rather than the Olympic village and the architecture of the skiing slopes, that OpenStreetMap provides a distinctly different approach, which makes it more valued so often by hikers or outdoorsfolk, rather than the streamlined images of roads of Google Maps that so often cast geographical surroundings only as lightly colored muted blocks.  Is OSM a more geographically ethical mapping of space, in ways that reflect how its composite character derived from a community of mappers, as much as a collective crowd-sourced medium whose users have championed it as an anti-corporate mapping model of map making?

The differences in local mapping are evident in the sextet of views that  Map Compare function offers.  Starting from Geofabrik’s local town, Karlsruhe, which seems the default starting place for Map Compare, one can scan the different levels of information they supply, in a massive time-suck and complex compare-and-contrast exercise, moving a nice view of the area around the town’s central castle, that invites visitors to compare what sort of map they’d rather use to navigate the city’s groundplan and to do so with the grain and detail that best illuminate or shine a light on the fabric of its urban planning:

Central Karlsruhe

Moving around Karlsruhe, away form the castle, one can compare alternate mapping views, which offer their own alternative glosses on the fabric of urban space, and their own points of entrance to it:

8 Maps Karlsruhe

There’s a neat abundance, underscored by a healthy pinch of relativity, in this crowding of a variety of perspectives.  Although there is an association of certainty with the map, each of the above images, using different databases which are often protected by copyright, offer different tags to recognize and navigate exactly the same environments, some focussing on the greenery and paths through it, or the road maps and the presence of the national border near Durmersheim, and others letting the national borderline slip into barely detectable gray.  There is a certain healthiness in this plurality–a plurality underscored, in the case of OpenStreetMap, by the varied contributions individuals have made over time, a la Wikipedia, to its contents.  There is a crossroads at which each stand, between data and design, that reflects an attempt both to give and to parse the most useful information in attractive form, and to create a selective map that give each meaning:  as far as selectivity of its record of urban space, the Stamen Toner map in the lower left gives it the most prominent definition by far:  it is something like a bleached version of a diagram of urban design.  The notation of walking-friendly regions of the city in the “Hike and Bike” map offer something like an index of walkability; the OSM De map, made by local German mappers, provides the clearest model to navigate the network of the largest driving streets in Karlsruhe.

And we can follow each into the nation, at a similarly close scale, toward those regions on the French-German border, near Durmersheim.  The maps foreground the different natures of indicating not only country roads and trails, but the nature of national boundary lines, suggest fairly radically different selective views of the local landscapes.  Are roads more important, or is green space?


8 Veiws Maps of Durmersheim

Or, in a clearer juxtaposed context, closer to home, but with similar concerns for the mapping of green space, contrast the highlighting of lakes, freeways, or greenery of the countryside, which the German OSM details in its lakes and countrysides, whose rather picturesque palette of lakes and greens that contrasts with the blah matte of Google maps, in whose flattened 2D color scheme the lakes stand out, but paths to navigate the landscape are annoyingly muted:

Comparing Info Foregrounded in Mapped Landscapes in Germany

Even if the map in Stamen Toner offers the sharpest contrast, as a strictly road map, the German OSM offers a clearer–or crisper–reading of the autobahn’s highway system and its levels of classification–important to drivers, but mute on Google Maps.  The relatively unprocessed nature of the OSM platform, which after all privileges the local detailing of a landscape in ways that are argued to recover the craftsman like nature of remapping space, albeit in a digital format, and after all process the viewer’s relation to place in ways that champion the individual agency of the locally situated mapper’s techniques.  Rather than deriving form LandSat imagery, even if including the backdrop deriving from Bing to ensure its global coverage, thanks to the new friends it gained at Microsoft, or satellite imagery, the structure of OSM uses a form of illustrator that seek to rehabilitate the familiar values of accuracy and open debate in the creation of a local map:  we are all, OSM users say, digital mappers, and can take back the overdetermined datasets all too often passively read and interpreted via GPS.

It’s well known that the detail put into the OSM maps offer a less synoptic view point on areas without roads–or where one might be more likely to travel as a pedestrian or hiker on a dirt path.  Close to my home, OSM is widely favored by hikers in National, State or Regional Parks.  Moving to one of the world’s most strikingly beautiful areas, around Mt. Tamalpais in Marin, the pronouncedly different views of space offer distinct ways of negotiating place and terrain, from the relatively blanched Google Map view of the terrain of the State Park to the mock-lithographic topography of MapQuest or Bing to the comprehensive detail of OSM–more busy, for some, but extremely relevant to orient oneself to the world-famous green space:  the density of the trails around Mt. Tamalpais in Marin are perhaps extreme, but this isn’t information that one would want absent from one’s world mapping system (or data) and suggests an erroneously vacant image of the park:  and the absence of points of elevation from most all mapping platforms, even if all GIS data is always “imperfect,” reminds us of the importance of finding criteria of selectivity that are comprehensive enough.

Mt. Tam Visualized

The question of cartographical comprehensiveness in a sense resonates with the perennial fantasy of mapping a complete view of place or region, trumping the difficulties of distortion with which mapmakers have perennially struggled.  But comprehensiveness–or accuracy–is less the point than the filters on data that exist in the structures (and databases) of certain GIS platforms.  Questions of accuracy are relative to the sort of point of view that one wants to measure, to be sure, and elevation points or nature walks might not be relevant to some–or ski runs.  But the features of the landscape surely are, and so are the role of maps as tools by which we attend to those features.  Something of the distinction that French theoreticians like Jean Baudrillard made between media of film and television in relation to the human imaginary seems to offer an apt point of distinction between the collective visualizations of OSM and the muted visuals in Google Maps, derived from LandSat photographs:  there is no trace of the imagined relation to the place or region in the platform, which offers far less of a basis to imagine one’s own relation to the places that it maps.

Moving to the greener space of the northeastern United States as a test case, I wanted to examine in some detail the different features of each platform a region that I know well, using the scalable functions of each to zoom into a specific place in the green space of central Vermont.

The distinct landscapes of different mapping media nicely foreground the benefits of Geofabrik’s own Topo map, and the OSM counterparts that suggest even greater detail and differences in the options of roads, paved and unpaved perhaps–for long an important local question–as well as variations of landscape green.  While MapQuest provides some important basic detail here, OSM offers a better view of the greenery and scenery, encrypting more information at a great density, especially in contrast to the generic light greens of Google or Bing.  (Sure, you have the Google Earth function to toggle to, but having a single sheet–either on a screen or paper print out, is an important navigational and orienting tool.)

Map Compare-  Vermont and NH

Moving to a local landscape that I know even better at first-hand, in greenish north-central Vermont, we can alternate among a range of mapped views to foreground or highlight distinct areas of the topography and roads that run through the flattened map:  between the topographic views of a cycle map to the routes of a Hike & Bike, the simple landscapes of an Open Cycle map, or the austere Bing and generic Google Map with its crowding of place names at odd angles, while the OSM offer views of the greenery that few others can beat.
8 Maps Near Montpelier

As we focus on a clustering of lakes further south on the Interstate, scrolling down at a greater scale, the clustering of three lakes offers a specific point to contrast mapping styles and the different data they embody and store, out of which we might focus on the somewhat notorious bridge across the lake that occurs in six of the following eight maps, but which I can confirm exists, a bridge which, while on the other six maps, it’s never noted that one cannot drive across:

Lakes and Ponds near Brookfield VT

The contrast in mapping styles grows more evident around the smaller town of Brookfield proper, where the variety of map-signs offers a sharpened difference in perspectives on place:  the eight different conventions of noting the interstate are not only surprisingly different in color scheme to differentiate their source, but the mapped data seems surprisingly distinct in these images:  OSM Mapnik suggests a bridge, lined in black, and overpass, but both disappear in Google Map, and in Bing Sunset Lake disappears, while in none of them is the fact that the bridge is wood, floating, but mostly submerged, and closed much of the year to driving noted.  Not only is the coloration and breadth of Interstate 89 distinct in each, but so is the presence–or absence–of the small lake, the old wooden Floating Bridge that cuts across the Sunset Lake, and the foliage that surrounds Brookfield village itself.  But the inability to traverse that Floating Bridge, either in winter, when it is covered by snow, or in summer, since it has been blocked to all but foot traffic, made me smile at the multiple absences in the map engines arrayed below.  And, perhaps as important for motorists, which mapping renders the transformation of paved to unpaved roads?

Map Compare-  Brookfield VT x 8

What is the best way for a map engine to engage its viewers?  A slightly tweaked variety in another grouping of maps of the entrance to the floating bridge one can’t traverse by car, at magnified scope, suggests the range of arranging information in only one small intersection, and the need to constantly compare mapping forms for their different level of detail:  from the differences among dirt and paved roads, to the range of topographic detail, to the view that the so-called Floating Bridge is in fact perpetually sinking in Brookfield, VT.  At the end, it will all depend on what we want to see in maps, and the array is simply and increasingly boggling.

 

Dirt v. Paved in Brookfield VT x 8

For while Google has gobbled upwards of six million miles of streets for Street View, the interest in offering an accurate survey of the land surrounding seems to have eluded, as the aim of completing a complete set of photographs of place–as if to seduce us to allowing Google to maintain a system of location-awareness through it–may be removed from what we want to see when we trust the selectivity of the map. For a generation weaned on video, and gratified by the dazzling display of visuals, the sunny streets of Street View and panorama of Google Earth may be enough eye candy for some, but the need for selective filters and for improving semantic legibility in maps might well lead the best maps to be those that are most carefully iteratively refined.

Leave a comment

Filed under Compare Map, GIS, Google IP

Mapping Worldly Entrances to Hell

We are forced carry our very own hells with us, and to know their maps but even the routes of access to them.   If the location of Hell has been mapped and re-mapped as a personal experience since the Renaissance, defining fixed locations of Hell projects something of a state of mind to the world’s physical geography.  If, to quote Marlowe’s Mephistopheles, “Hell hath no limits, nor is circumscrib’d/ In one self place; but where we are is hell,/ And where hell is, there must we ever be,” the places Hell resides is almost a practice of mapping that reflects a culture’s cartographical imagination, hell has proliferated in the age of genocide on an unknown scale, as the atrocities of Armenia, Cambodia, Argentina, Bosnia, Iraq and Rwanda have created new landscapes from hell that have proliferated in the world as real places of a level of trauma that transcends any normalcy, including the normalcy of a map.

If we carry around our personal hells with us, the worldly entrances to hell came back with a vengeance in the Second World War and long before.  For the problems of these atrocities challenges on an ethical and moral scale the commensurability of our conventions of mapping, and with it any commensurability itself.  Perhaps the muted colors and odd grey zones by which earlier concentration camps in the Nazi era suggest that they are zones “off of the map,” indeterminate spaces ringed by green fields, all but exempts them from conventions of mapping, their primitive barracks, gates, transports, mess halls, work fields, ditches and crematoria all outside of “normal” space, and unmapped, left to strain credibility–even if they were viewed from space by reconaissance flights, these spaces and their modern proliferation cannot be adequately morally mapped in ways we know the world.

FARBEN_DWORY

This sense of places out of the normal, hardly part of humanity, has perhaps led to the proliferation of maps of hell online.  The problem of proliferating hells is a one good way to describe modernity.  From Samantha Powers’ attempt to map “problems from Hell”  as eventualities the United States government will be condemned to face to the problems of mapping atrocities that recurred in the terrifying landscapes of Hades, Argentina, worldly hells have proliferated in the world from Nazi concentration camps to sites of disappearing that ask us to map the presence of hell in the world, in a grim geography of devasttion that challenged pallettes and iconography to describe adequately.

grim geogrqphy of devastationaBefore these maps of spaces of dehumanization and devastation, we are really looking into hellish worlds we had not been able to see before.   But even these dots cannot capture the scale of the hellscapes that emerged for the accelerated loss of life within the industrialization of death that proceeded from Heinrich Himmler’s order of 19 July 1942 stated that unleashed mass-killings from bullets, fire, and gas extermination to fullifll the demand that by the end of December 1942, all Jews, gypsied in Greater Germany be killed, leading to an unprecedented intensity of rates of mass-killings almost impossible to map on paper or by a graph, challenging as the spatial dynamics of the three-month long burst of killings is poorly documented–intentionally–and indicate a terrifying challenge to the world of the data vis that challenges the imagination to even attempt to “map” in the over 40,000 camps of imprisonment and mass-killing that were built between 1933 and 1945, dedicated to imprisonment, forced labor, or mass killing sites dedicated to exterminating Jews, Sinti, Roma, Communists, and so-called “enemies of the state.”

extermination

aau7292-f1

There is a sense of the utter inadequacy of an aerial view–or indeed even Google maps–to map the horrors of sites commensurate to their moral and ethical existence, as if they lay resolutely and stubbornly outside the known world and could not be assimilated to the categories by which we map it.

Landscap

Aerial Reconnoissance Flight over Auschwitz-Birkenau, April 4, 1944 

It’s perhaps not a surprise that every culture seems to have its own notion of Hell but of where the location of hell and its entrance is.  If one can pinpoint and map it in an image of the known world, perhaps one can escape its presence in one’s own mind.  The poet Czeslaw Miłosz wondered, in a very late poem of 2003, “Have we really lost our faith in that other space?/ Have they vanished forever, Heaven and Hell?/ . . . And where will the damned find suitable quarters?” and bemoaned almost tearfully the unimaginable proportions of the “enormity of the loss,” but there is considerable existential comfort in being able to map Hell with security, and indeed to map the intersection between hell and the world that seems normal, as if the presence of Hell demands of expressibility that elicit stubborn difficulties in placing recurring reappearances of Hell on the relative poverty of conventions we use in a global map of human settlement. The problem of mapping hell was perhaps long a part of humanity, as much as the evils of genocide stupefy in their excess, and raise questions of how to map not only people and places but souls in the world.  Mapping hell is, indeed, something of a poetic feat.

Mapping was long about finding a place for the soul in the world, however, as much as ordering spaces or offering way-finding.  You know the lay of the land, and the parts you want to avoid.  As if consciously and quite intentionally one-upping Christopher Marlowe, on seeing the efflux of modern industry afflicting  London, Percy Bysshe Shelley imagined “Hell is a city much like London— A populous and a smoky city,” to comment on the transformation of England; his belief that “It is not necessary to consider Hell and the Devil as supernatural machinery” is uncannily similar to the many maps that pinpoint Hell’s multiple worldly entrances for those eager to read them.  It has long been far more satisfying if one can try to pinpoint the entrance points in informed fashion, using some sort of knowledge or evidence to buttress the choice.  The location of “Hell” or the underworld was, of course, pre-Christian, even if it is now colored by Christian sources; Hell is a pre-Christian mental geography that was mediated by Christianity and its own specific notions of suffering  and remorse, but also is a place that we all know exists, and are eager to find–although not to go there ourselves.  Is it any surprise that the dominance of point-based mapping, with its comprehensive tally of location, raises the fundamental moral question of mapping a common relation to hell?

Perhaps it is a coincidence that the proliferation of hells began with the dominance of new national maps, and new military maps, crafted to enable us to think outside of a national frontiers, created a point-based mapping system like the Universal Transversal Mercator, that raised moral questions of where hell was, and that hell exists in the lives of most modern refugees, who live not only outside the edges of borders, but, as the unhoused, outside of geolocation systems.

But perhaps our current maps, dominated by geodata, force the question of the lack of location of a hell, at the same time as we are seeing a proliferation of global hells, all absent from the point-based maps that we treat as surrogates for reality.  According scripture, Hell is located deep down in the earth, without either geographic specificity and far more figuratively evocative than precise.  Hell is  reality and state of mind for the Gospels and Apocalypse; it is not a precise location:  it is a place where in “outer darkness: there shall be weeping and gnashing of teeth” (Matthew 25:30), whose inhabitants are “in agony in this fire” (Luke 16:24), surrounded by “the devil and his angels” (Matthew 25:41).  The topography of the new camps made little sense in rational mapping systems, but as a vampiric relation to the enslavement of people, before the scale of atrocities could be imagined.

nazi-vampire-1941

The image of perpetual burning, self-consumed bodies has been embellished with extensive pictorial detail as a place of eternal punishment, and a site of the destruction of both body and soul and of unending separation from both; it was based on the Old Testament idea of  “Sheol” as an abode of the dead (Psalm 49:13-14)–or of those with no abode or place to be, but this place with no life was always seen as closely connected to our own.  Hell was deeply spiritual for Dante and in his age–the appeal that we had an informant who had in fact been their to survey its complex topography and descending rings of punishments bore the satisfying sense that we knew where we are in the moral compass of life.  The appeal of Dante’s map of hell is evident in the considerable care and detail which Sandro Botticelli and others used to delineate the space through which where Virgil led Dante and navigated among the inhabitants of hell’s circles–an image popular in the late fifteenth century–that could be examined with some recognition and even more amazement as a site of the afterlife.

Botticelli's Ms Map of Dante's Hell

When Dante’s Florentine editor Girolamo Benivieni’s prepared a printed edition including engraved maps, the portal to Hell was strikingly placed in explicitly modern geographic terms within the terraqueous sublunary world:

Benivieni 1506 Dante's Hell

The deep comfort of this clearly mapped ontology of the afterlife is to some extent preserved today.  Online, we can also navigate this image, thanks to digitization of manuscript images, on one’s very own, and explore the mind-blowing map that Sandro Botticelli drew as if confronting the page from inches away in all its gloriously imagined Dantesque details.  The mapping of Hell has taken off in ways that oddly reflects a pretty secular age; sites of anguish and suffering are, it turns out, still pretty compelling to map in a geographical lens.

Compelling woodcut maps described the topography of the realm of the Dantesque afterlife with exquisite geographic care:

1527_33.wc1.150dpi

Hell was long an individualized affair, and rightly so, the culmination to a balance of sins physical and of mind.  But the mapping of a public geography of hell–entrances to the underworld, now navigated not only but Virgil and Vulcan, or even Percy Jackson, but able to be pinpointed on a map.  There seems to be somewhat of a flourishing of the addition of “Hell”-sites on the web today, in fact, something of a response to the absence of this all-too-concrete state of mind from the reaches of Google Earth–not that some folks haven’t tried.  Perhaps the absence of hell’s location on Google Maps–or how Hell frustrates that portal promising ubiquitous coverage to any user–may have helped generate something like a proliferation  of on-line pseudo-erudition about Hell’s possible locations, and the curiosity that it could be in fact right around the corner in some pretty familiar sites that we can arrive at by our devices.

The appeal of mapping hell–and at looking at the sites where others map hell–is a branch of the Googlish compulsion to provide a total mapping of humanity, as much as a religious ontology, and is reflected in the proliferation of models of Hell that circulate online and provide some sort of satisfaction that we known where we are.

Continue reading

1 Comment

Filed under Apocalypse, Biblical Geography, Global Displacement, Google Earth, Hell