Let Me Share My Newest Diversion....
Mar. 12th, 2015 11:21 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
stormdogI've found a new thing that I see a high potential to get obsessive over. It has just the right combination of creativity and rationality; art and science. It's similar to the way I feel when I'm editing photos; the process is really satisfying, and I think I need to be on guard against spending too much time doing this and neglecting my other work. Here's a screenshot of what I'm working on, before I drag myself away and get ready for bed.
Each numbered plus sign is a point of correlation between the 1972 map and a current street map of the city. I've invested about two hours into this map and have nearly two hundred control points. It feels rather like performing cartographic surgery. It's also a lot like smoothing bubbles out from under a sticker. I have to keep plotting points that resolve distortion introduced by the last point. Once there are enough, though, it all lines up in a really beautiful way.
As I work on this, it's been really interesting to think about the choices made by the cartographers who worked on creating these maps. No map represents reality perfectly. There may be, for instance, an intersection that doesn't line up quite perfectly in real life. Maybe somebody messed up when platting a subdivision. Maybe the roads didn't match up when the area was annexed. Who knows? On the map, maybe the real life offset is there, but if it's small enough, maybe it isn't. If it's not, I have to decide how to compromise my referencing to minimize inaccuracies by spreading them across the area in question rather than bunching them up in one spot. Thus, I'm frequently making judgment calls, when my map isn't quite lining up with the current street map, about whether I'm looking at change over time, cartographic error, or just some necessary fudging in the map design. I make my choice and reference accordingly. That's the art part. And maybe even a little psychology, as I get into the head of the person who was putting the map together and think about why choices were made and what the implications are.
My nice, new computer lets me work on ArcGIS on the middle monitor, Google Earth (with all my other completed overlays available) on the left monitor, and the original full size image file of the map I'm working with on the right monitor, as well as all the ArcGIS tool panels. I can easily refer between all of this stuff spread out in front of me and it makes the work *so* much faster. I'm really happy with my setup.
I'm going to make some of these available as Google Earth KMZs for interested folks to download. I can't do that with the Sanborn maps, as the Wisconsin Historical Society asked that I not distribute high-res versions. But I want to have a big collection of overlays that maybe the University of Wisconsin-Parkside Archives and Area Research Center could distribute to interested researchers. I'd definitely like all the work I'm doing to be a contribution to common knowledge.
Speaking of such things, does anyone I know have any experience getting either KMZ files or georeferenced TIFs or JPGs to overlay correctly in Google Maps?
Each numbered plus sign is a point of correlation between the 1972 map and a current street map of the city. I've invested about two hours into this map and have nearly two hundred control points. It feels rather like performing cartographic surgery. It's also a lot like smoothing bubbles out from under a sticker. I have to keep plotting points that resolve distortion introduced by the last point. Once there are enough, though, it all lines up in a really beautiful way.
As I work on this, it's been really interesting to think about the choices made by the cartographers who worked on creating these maps. No map represents reality perfectly. There may be, for instance, an intersection that doesn't line up quite perfectly in real life. Maybe somebody messed up when platting a subdivision. Maybe the roads didn't match up when the area was annexed. Who knows? On the map, maybe the real life offset is there, but if it's small enough, maybe it isn't. If it's not, I have to decide how to compromise my referencing to minimize inaccuracies by spreading them across the area in question rather than bunching them up in one spot. Thus, I'm frequently making judgment calls, when my map isn't quite lining up with the current street map, about whether I'm looking at change over time, cartographic error, or just some necessary fudging in the map design. I make my choice and reference accordingly. That's the art part. And maybe even a little psychology, as I get into the head of the person who was putting the map together and think about why choices were made and what the implications are.
My nice, new computer lets me work on ArcGIS on the middle monitor, Google Earth (with all my other completed overlays available) on the left monitor, and the original full size image file of the map I'm working with on the right monitor, as well as all the ArcGIS tool panels. I can easily refer between all of this stuff spread out in front of me and it makes the work *so* much faster. I'm really happy with my setup.
I'm going to make some of these available as Google Earth KMZs for interested folks to download. I can't do that with the Sanborn maps, as the Wisconsin Historical Society asked that I not distribute high-res versions. But I want to have a big collection of overlays that maybe the University of Wisconsin-Parkside Archives and Area Research Center could distribute to interested researchers. I'd definitely like all the work I'm doing to be a contribution to common knowledge.
Speaking of such things, does anyone I know have any experience getting either KMZ files or georeferenced TIFs or JPGs to overlay correctly in Google Maps?
