7 ms·

Can anyone working in cartography comment on this? What is the state of active research in cartography? Map making used to be one of the most prized technical

Can anyone working in cartography comment on this? What is the state of active research in cartography? Map making used to be one of the most prized technical skills and I was wondering where the field is moving. First thing that came to mind was new d3 transform...

Waterluvian 10y ago
I don't work as a cartographer but I studied geography for a Bachelor's and Master's. That being said, my knowledge has atrophied a lot. I was exposed to hundreds of map projections in school, so I'm still kind of trying to figure out what the novelty is here. It just looks like a Japan-centrerd equal-area projection. The tetrahedron is kind of neat. Projections are generally planar, cylindrical, or conical. By using a mosaic of planes, I imagine the formula for this projection would be unbelievably complex, making it much less general purpose. When you want to represent a spheroid on a 2d surface, you have to distort one or more of shape, area, direction, distance. The objective when selecting a projection is to understand the intended purpose and audience. If I had to make a guess, I'd say the relevance here is the exploration of what projections we most commonly use and if they're still the best choices for today's purposes and audiences. I would love to see what Tissot's indicatrix looks like for this projection.
- gcb0 10y ago
  you nailed it. Japan has a huge love for Japan centered maps. if you look at any early nes or Sega games, all the maps are your familiar Cartesian but with Japan in the center, making the jump from ny to London look much longer than la to Tokyo.
  - yongjik 10y ago
    > Japan has a huge love for Japan centered maps. You make it sound as if it's unusual. Of course they prefer Japan in the middle, they're in Japan! Just like Brits (I presume) have a huge love for Britain-centered maps. I'll bet not a single world map in British textbooks has the world cut in the Atlantic.
    - Waterluvian 10y ago
      Japan is known for an exceptional level of nationalism through history. But yeah, you see this a lot.
    - contingencies 10y ago
      Same here in China: not bad for we expat Australians!
    - wst_ 10y ago
      It isn't that. Or at least not only that. They are pragmatic nation. They are considering who and why would want to use the map, and then tailor it to meet the needs of an audience. I am not sure about other countries but European maps are always oriented for north. Which works fine when you read them in book, but not necessarily when you are looking at big guide map on the street. Yes, sure, this map may show red "you are here" mark, but it is still difficult to find out your way as a map itself is always oriented to north, while you may not be. Japanese street maps are oriented on you (most of them). When you're standing in front of it - "up" on map always means "go forward". Which might be confusing at the beginning, if you came from Europe. But the more you look at them, the more easy it is to find your way.
      - pitt1980 10y ago
        Hollywood Studios in Disney World had a park map where the entrance to the park wasn't at the bottom, I found it really disorienting, I think due to the effect you mentioned, I naturally want to walk 'up' to 'go forward' through the map http://www.wdwmagic.com/attractions/disneys-hollywood-studios/gallery/04apr2016-april-2016-disney's-hollywood-studios-guide-map-with-backlot-area-removed/27424.htm http://www.wdwmagic.com/attractions/disneys-hollywood-studio...
sarcher 10y ago
From my perspective, new projections are very interesting but largely irrelevant until a specific use case is defined. Just minimizing a specific type of projection error isn't typically enough, since a large amount of map comprehension is guided by the users pre-existing conception of what a map of a specific area should look like. And there are a lot of projections/coordinate systems/datums to consider, here is a source to browse through: http://spatialreference.org/ http://spatialreference.org/ I'm usually working in an EPSG code, and of the 4,000+ EPSG codes I probably use 10 regularly and perhaps have used 100 in my lifetime (if anything, that's an overestimate). They are designed this way, most of them simply represent a small area of the earths surface well. Since I primarily work in the Northeast United States, I don't need to use a whole lot of them. I'm playing loose with terms here, the EPSG codes are really 'spatial references' made up of various 'projections' and 'coordinate systems' and 'datums' and etc. I usually find it easier to just talk about spatial references because they are so commonly used as a shortcut approach as opposed to individually selecting the attributes that make up the spatial references. Cartography includes an incredible wide body of work, defined largely by the data sets being used, the audience/purpose of the display, and the medium. For example, if we narrow our focus to just models of the actual world this includes surveys, navigation programs, hiking maps, remote sensing rasters, etc. - all of which have very little overlap concerning data, audience, and medium. I build geography products, but I'm not a researcher - hopefully someone with more research experience can chime in, but perhaps your questions is too broad given how deep various geography niches can run. I know I certainly haven't even started to answer your question.
- SOLAR_FIELDS 10y ago
  I think Google has done a lot to get us away from different projections with EPSG 3857 and 4326. It's both a blessing (because you can expect most geographic data to be available in that format) and a curse (because 3857 is kind of a shitty projection and the standardization sort of disincentivizes people to learn about the intricacies of coordinate representation).
  - sarcher 10y ago
    Totally agree about prevalence of 4326/3857 as a net positive. Given I still find projection/coordinate representation a little tricky, I think the chances that widespread education on the subject would ever exist to be pretty slim. Might as well just have a good default. Displaying state/country level data using EPSG:4326 isn't great, thankfully it doesn't happen often.
mxfh 10y ago
There is some research on getting cartography from static 2d (print) to interactive 2d (web/devices), like always having a locally optimized projection at any given time. http://cartography.oregonstate.edu/ScaleAdaptiveWebMapProjections.html http://cartography.oregonstate.edu/ScaleAdaptiveWebMapProjec...

Waterluvian 10y ago

I don't work as a cartographer but I studied geography for a Bachelor's and Master's. That being said, my knowledge has atrophied a lot. I was exposed to hundreds of map projections in school, so I'm still kind of trying to figure out what the novelty is here. It just looks like a Japan-centrerd equal-area projection. The tetrahedron is kind of neat. Projections are generally planar, cylindrical, or conical. By using a mosaic of planes, I imagine the formula for this projection would be unbelievably complex, making it much less general purpose. When you want to represent a spheroid on a 2d surface, you have to distort one or more of shape, area, direction, distance. The objective when selecting a projection is to understand the intended purpose and audience. If I had to make a guess, I'd say the relevance here is the exploration of what projections we most commonly use and if they're still the best choices for today's purposes and audiences. I would love to see what Tissot's indicatrix looks like for this projection.

gcb0 10y ago
you nailed it. Japan has a huge love for Japan centered maps. if you look at any early nes or Sega games, all the maps are your familiar Cartesian but with Japan in the center, making the jump from ny to London look much longer than la to Tokyo.
- yongjik 10y ago
  > Japan has a huge love for Japan centered maps. You make it sound as if it's unusual. Of course they prefer Japan in the middle, they're in Japan! Just like Brits (I presume) have a huge love for Britain-centered maps. I'll bet not a single world map in British textbooks has the world cut in the Atlantic.
  - Waterluvian 10y ago
    Japan is known for an exceptional level of nationalism through history. But yeah, you see this a lot.
  - contingencies 10y ago
    Same here in China: not bad for we expat Australians!
  - wst_ 10y ago
    It isn't that. Or at least not only that. They are pragmatic nation. They are considering who and why would want to use the map, and then tailor it to meet the needs of an audience. I am not sure about other countries but European maps are always oriented for north. Which works fine when you read them in book, but not necessarily when you are looking at big guide map on the street. Yes, sure, this map may show red "you are here" mark, but it is still difficult to find out your way as a map itself is always oriented to north, while you may not be. Japanese street maps are oriented on you (most of them). When you're standing in front of it - "up" on map always means "go forward". Which might be confusing at the beginning, if you came from Europe. But the more you look at them, the more easy it is to find your way.
    - pitt1980 10y ago
      Hollywood Studios in Disney World had a park map where the entrance to the park wasn't at the bottom, I found it really disorienting, I think due to the effect you mentioned, I naturally want to walk 'up' to 'go forward' through the map http://www.wdwmagic.com/attractions/disneys-hollywood-studios/gallery/04apr2016-april-2016-disney's-hollywood-studios-guide-map-with-backlot-area-removed/27424.htm http://www.wdwmagic.com/attractions/disneys-hollywood-studio...

sarcher 10y ago

From my perspective, new projections are very interesting but largely irrelevant until a specific use case is defined. Just minimizing a specific type of projection error isn't typically enough, since a large amount of map comprehension is guided by the users pre-existing conception of what a map of a specific area should look like. And there are a lot of projections/coordinate systems/datums to consider, here is a source to browse through: http://spatialreference.org/ http://spatialreference.org/ I'm usually working in an EPSG code, and of the 4,000+ EPSG codes I probably use 10 regularly and perhaps have used 100 in my lifetime (if anything, that's an overestimate). They are designed this way, most of them simply represent a small area of the earths surface well. Since I primarily work in the Northeast United States, I don't need to use a whole lot of them. I'm playing loose with terms here, the EPSG codes are really 'spatial references' made up of various 'projections' and 'coordinate systems' and 'datums' and etc. I usually find it easier to just talk about spatial references because they are so commonly used as a shortcut approach as opposed to individually selecting the attributes that make up the spatial references. Cartography includes an incredible wide body of work, defined largely by the data sets being used, the audience/purpose of the display, and the medium. For example, if we narrow our focus to just models of the actual world this includes surveys, navigation programs, hiking maps, remote sensing rasters, etc. - all of which have very little overlap concerning data, audience, and medium. I build geography products, but I'm not a researcher - hopefully someone with more research experience can chime in, but perhaps your questions is too broad given how deep various geography niches can run. I know I certainly haven't even started to answer your question.

SOLAR_FIELDS 10y ago
I think Google has done a lot to get us away from different projections with EPSG 3857 and 4326. It's both a blessing (because you can expect most geographic data to be available in that format) and a curse (because 3857 is kind of a shitty projection and the standardization sort of disincentivizes people to learn about the intricacies of coordinate representation).
- sarcher 10y ago
  Totally agree about prevalence of 4326/3857 as a net positive. Given I still find projection/coordinate representation a little tricky, I think the chances that widespread education on the subject would ever exist to be pretty slim. Might as well just have a good default. Displaying state/country level data using EPSG:4326 isn't great, thankfully it doesn't happen often.

mxfh 10y ago

There is some research on getting cartography from static 2d (print) to interactive 2d (web/devices), like always having a locally optimized projection at any given time. http://cartography.oregonstate.edu/ScaleAdaptiveWebMapProjections.html http://cartography.oregonstate.edu/ScaleAdaptiveWebMapProjec...