Show HN: Artistic Maps of Over 4,000 U.S. Cities(projectjefferson.com)
projectjefferson.com
Show HN: Artistic Maps of Over 4,000 U.S. Cities
https://www.projectjefferson.com/
13 comments
My first guess would be that this is using OSM data, but I didn't see any mention of that on the site and didn't see any OSM copyright notices on the images. Could one of the creators explain where the map data comes from?
Yup, totally OSM data. Going to add attribution to them in our Our Story section right now. Totally missed that, but happy to give credit where credit is due. Thanks!
http://wiki.openstreetmap.org/wiki/Legal_FAQ#3a._I_would_lik...
---
Our requested attribution is "© OpenStreetMap contributors". You must also make it clear that the data is available under the Open Database Licence. This can be achieved by providing a "License" or "Terms" link which links to www.openstreetmap.org/copyright or www.opendatacommons.org/licenses/odbl.
... This credit needs to appear in a place reasonable to the medium you are utilising. In other words, you should expect to credit OpenStreetMap in the same way and with the same prominence as would be expected by any other map supplier.
... For a printed map, the credit should appear beside the map if that is where other such credits appear, and/or in the "acknowledgements" section of the publication (often at the start of a book or magazine).
---
EDIT: looks like this only applies to OSM maps, not data. What's the difference between a "Produced Work" and a "Derivative Database"?
---
Our requested attribution is "© OpenStreetMap contributors". You must also make it clear that the data is available under the Open Database Licence. This can be achieved by providing a "License" or "Terms" link which links to www.openstreetmap.org/copyright or www.opendatacommons.org/licenses/odbl.
... This credit needs to appear in a place reasonable to the medium you are utilising. In other words, you should expect to credit OpenStreetMap in the same way and with the same prominence as would be expected by any other map supplier.
... For a printed map, the credit should appear beside the map if that is where other such credits appear, and/or in the "acknowledgements" section of the publication (often at the start of a book or magazine).
---
EDIT: looks like this only applies to OSM maps, not data. What's the difference between a "Produced Work" and a "Derivative Database"?
Produced Work is an output that uses but does not alter the OSM data. So you can use an OSM database with your own data overlaid on top. These lovely paper prints would be a produced work. One basically does not change the OSM data, you just produce things with it.
Derived database is where you adapt the OSM data with your own data, or perform some kind of manipulation of the data For example, vector simplification or calculating the road lengths and adding that into the data, or merging addresses from one dataset with the polygons from OSM.
Derived database is where you adapt the OSM data with your own data, or perform some kind of manipulation of the data For example, vector simplification or calculating the road lengths and adding that into the data, or merging addresses from one dataset with the polygons from OSM.
Thanks for the explanation.
Seems like there is wiggle room there, but we're not opposed to dropping the OpenStreepMap copywrite into the bottom corner of the map.
In theory, if your maps help advertise OSM, contributions will increase, helping OSM and ... your future maps :)
So if you are using OSM data, why not non-US cities as well?
Did you use TileMill for the styling or something else?
No, we actually hand-designed all of our styles in XML and generated all images from the pure data in python.
I'm in your target audience -- I used to have a wall size Thomas Brothers map of LA on my wall, and I have a large map of Southern CA geography there now. But:
Why is are most of the color schemes so garish? Did you run these by a designer? I'd think that many of the LA posters (Blackfeet, crow, Shoshone, Pawnee, even Bannock) are so garish as to be unusable in most rooms.
Why is the water the same color as the border?
And, you need deeper zoom (at least in selected areas), or I can't see what I'm getting. Within the preview, some indication of the scale of the shown features within a print of a given size would be useful.
Why is are most of the color schemes so garish? Did you run these by a designer? I'd think that many of the LA posters (Blackfeet, crow, Shoshone, Pawnee, even Bannock) are so garish as to be unusable in most rooms.
Why is the water the same color as the border?
And, you need deeper zoom (at least in selected areas), or I can't see what I'm getting. Within the preview, some indication of the scale of the shown features within a print of a given size would be useful.
To each their own, I suppose. Our goal was create designs that were bold and bordering on the abstract, rather than to create traditional maps. I'd say maps are one of the more ubiquitous pieces of art, but rarely are cause for any attention. We wanted to create art anchored in people's identities -- not just another standard map of Paris.
I think what I'm getting at is that you probably aren't in our target audience for these current designs. That's totally okay... Perhaps one of our future designs will suit your fancy.
I think what I'm getting at is that you probably aren't in our target audience for these current designs. That's totally okay... Perhaps one of our future designs will suit your fancy.
Hey there,
I'm one of the two creators of Project Jefferson. We'll do a full run down of the tech we used to do this some time this week, but I'm 100% willing to chat about anything to do with the project.
We started working on this just a couple weeks ago and are super excited to ship it tonight. If anything breaks or looks wonky, please drop me a line. :)
Cheers, Sean
I'm one of the two creators of Project Jefferson. We'll do a full run down of the tech we used to do this some time this week, but I'm 100% willing to chat about anything to do with the project.
We started working on this just a couple weeks ago and are super excited to ship it tonight. If anything breaks or looks wonky, please drop me a line. :)
Cheers, Sean
I can't help thinking it's a bit tasteless, and at best arbitrary, to name your maps of New York, a territory brutally taken from the Lenape by the Dutch, after (unrelated) Native American populations.
See: http://en.wikipedia.org/wiki/Kieft%27s_War
Edit: I see now it's the color schemes for every map, in which case I'd say it's perhaps a bit tasteless to name the colors of maps of colonial cities after exterminated peoples.
Otherwise, really lovely project.
See: http://en.wikipedia.org/wiki/Kieft%27s_War
Edit: I see now it's the color schemes for every map, in which case I'd say it's perhaps a bit tasteless to name the colors of maps of colonial cities after exterminated peoples.
Otherwise, really lovely project.
Thanks for your comment.
We were on the fence with the naming of the maps, but decided to roll with the potential misunderstanding of what they represent. We were really interested in playing into themes of American exploration for this project. We settled on Project Jefferson as the name of the project, in honor of Jefferson's commissioning of the Louis and Clark expedition. We named our map themes after the tribes found during their journey, in homage to the shared identity of the land.
We were on the fence with the naming of the maps, but decided to roll with the potential misunderstanding of what they represent. We were really interested in playing into themes of American exploration for this project. We settled on Project Jefferson as the name of the project, in honor of Jefferson's commissioning of the Louis and Clark expedition. We named our map themes after the tribes found during their journey, in homage to the shared identity of the land.
"We named our map themes after the tribes found during their journey, in homage to the shared identity of the land."
- I don't understand what you mean here. For the NYC "Sioux" map you mean that there were Sioux tribes in the NYC area? I am sure that is not true. It doesn't feel like "homage to the shared identity". For me it really strikes falsely
- I don't understand what you mean here. For the NYC "Sioux" map you mean that there were Sioux tribes in the NYC area? I am sure that is not true. It doesn't feel like "homage to the shared identity". For me it really strikes falsely
>> We named our map themes after the tribes found during their journey
> For the NYC "Sioux" map you mean that there were Sioux tribes in the NYC area?
I don't think Louis & Clark did much exploring in the NYC area...
> For the NYC "Sioux" map you mean that there were Sioux tribes in the NYC area?
I don't think Louis & Clark did much exploring in the NYC area...
Wouldn't it be copyright infringement if the printed maps don't attribute OSM?
http://www.openstreetmap.org/copyright
This seems to indicate that yes, OSM must be attributed in print.
This seems to indicate that yes, OSM must be attributed in print.
Attribution is required, however that does not stop them applying any other license they want. In other words, the maps they create are not CC-by-SA - they are "Produced Works" and can be All Rights Reserved if the publishers want. (although the map data would still be copyright OSM contributors)
Potentially, yes. We're going to go ahead and print the attribution on the maps just in case. Thanks!
I'm not understanding how the cities are chosen. For example, Madison, WI (home of a Big 10 university where probably more than a few folks here went to school) is non-existent, but Fitchburg, Middleton, and Sun Prairie, three suburbs of Madison are included. Madison is the second largest city in the state. Just for giggles, I looked up Michigan, and the second largest city (Grand Rapids) is represented even though it's a smaller city than Madison.
That's a somewhat random example from me searching for cities folks here might know, but it's certainly odd.
That's a somewhat random example from me searching for cities folks here might know, but it's certainly odd.
Really good question. We started with a list of the 5,000 highest population cities in the U.S and tried to generate maps of as many of them as possible. We are using Open Street Maps data as our source, which is generally pretty amazing but didn't provide the data in the format that we needed it for every city. There were a good couple hundred cities that we wanted to generate but didn't. We hand generated binding boxes for a good number of very large cities, but didn't do it for all of the missing ones.
If you'd like, we can make one for Madison!
If you'd like, we can make one for Madison!
Here is a similar remix of OSM, that allows you to zoom in and pan, export as .jpg, or embed - for free:
http://maps.stamen.com/#watercolor/12/42.3318/-83.0503
Good find! We ran across this while building Project Jefferson. It's a pretty cool project.
I typed in 'Eugene, Oregon' into the search thing and it didn't find it. The map does exist though.
I actually find having lots of nicely done details is what I like in a map. The super-detailed USGS quads, for instance, are very nice.
I'm also not wild about the native american names, I think they might be a distraction for a lot of people. Maybe you could go with US regions or something.... "northwest", "southwest", something like that? Or states?
I actually find having lots of nicely done details is what I like in a map. The super-detailed USGS quads, for instance, are very nice.
I'm also not wild about the native american names, I think they might be a distraction for a lot of people. Maybe you could go with US regions or something.... "northwest", "southwest", something like that? Or states?
Clever project that shows how access to data, normally behind massive paywalls, can be used for really creative projects that otherwise wouldn't exist.
Thanks!
Your search feature doesn't handle "St. Louis" type searches.
Searching for "St." anything returns nil, but searching for "Paul" returns "St. Paul" and others.
Otherwise it's an interesting exercise.
Searching for "St." anything returns nil, but searching for "Paul" returns "St. Paul" and others.
Otherwise it's an interesting exercise.
I was surprised to see the city I live in (because it's small - population 7100) but I guess 4000 cities is a lot, googling shows people estimating 20k cities total in the US.
Interesting concept, but it needs curation from people who live in the places it purports to map. Perhaps it's just my familiarity with the state, but there is a lot of dubious stuff in the results for New York and it gives a "this is completely automated and maybe not as useful as it could be" vibe.
Some background: by law, every part of New York state except New York city is under at least three levels of government: state, county, and town/city. Many places have four: a village or city within a town (there's even the case of hamlets, which are like villages but do not have their own governments). Frequently, a town will have a village with the same name as the town within its borders, but other villages and hamlets as well.
It appears (with one exception that I found, noted below) that villages are being treated as cities for map generation, but this is problematic in New York for two reasons:
1. Long Island, being right next to New York City, is incredibly densely populated. All of Nassau County could arguably be considered a single urban core, never mind its constituent towns (and the villages that make up those towns). Hempstead winds up being overrepresented because it looks like every single one of its villages makes it into the list. Hempstead itself shows up, but only the village of the same name that appears within its town.
2. Upstate, there are a lot of towns that have villages within them that have the same name. Unless specifically qualified with "the village of," I've found that people are usually referring to the town. I've found that's the case even when referring to a village that doesn't have a corresponding town of the same name, but happens to be the main center of that town.
For example, I grew up near a village called Washingtonville. I would say I'm from Washingtonville even though my house wasn't technically within the bounds of the village: it was in its town, Blooming Grove. If I'm looking for a map of Washingtonville, I would expect to see not just the comparatively tiny area incorporated as "the Village of Washingtonville", but its surrounding area as well. To make matters more confusing, there is a hamlet of Blooming Grove within the town of Blooming Grove as well, but nobody cares about it: I couldn't name 3 people would would ever want a map of the hamlet of Blooming Grove.
Another example is Wallkill, which is even more confusing. There's a town of Wallkill in Orange County: within it, there's the city of Middletown. If you're from there, you're going to say you're from Middletown unless speaking to someone from the area who could tell the two apart. There's also a hamlet of Wallkill in another county, part of a different town. If someone says they're from Wallkill, they're probably talking about the hamlet, not the town. You'd say you're from Wallkill (and not the town in which it resides, Shawangunk) because... well, I don't know. That's the way it's always been.
Which brings me to the exception I mentioned earlier. Check out the listing for Monroe, NY: https://www.projectjefferson.com/?city=1804311
Three entities in New York share this name: a town in Orange county, a village within that town, and a county 300 miles northwest of the town that encompasses the city of Rochester and its suburbs. (There is no city of Monroe in New York.) The site shows a map for the county. Presumably, the list being worked off is actually referring to the town, which has a population of ~39,000 and not its village, which only has a population of ~6,000.
Part in parcel with Monroe and the confusing town/village distinction, Kiryas Joel is listed (https://www.projectjefferson.com/?city=175464). Kiryas Joel is a village within the town of Monroe comprising of about half its population. Even though it's massive, if you're from Kiryas Joel, you say you're from Kiryas Joel. But if you're from anywhere else within the town of Monroe, you say you're from Monroe. This is for cultural reasons: Kiryas Joel is a very insulated village consisting of mainly people of the Hassidic Jewish faith. If you're not Hassidic, you wouldn't self-identify with it even though it's the largest village within the town.
This was long, but hopefully it demonstrates some of the perils of taking an automated-only approach to this type of mapping.
Some background: by law, every part of New York state except New York city is under at least three levels of government: state, county, and town/city. Many places have four: a village or city within a town (there's even the case of hamlets, which are like villages but do not have their own governments). Frequently, a town will have a village with the same name as the town within its borders, but other villages and hamlets as well.
It appears (with one exception that I found, noted below) that villages are being treated as cities for map generation, but this is problematic in New York for two reasons:
1. Long Island, being right next to New York City, is incredibly densely populated. All of Nassau County could arguably be considered a single urban core, never mind its constituent towns (and the villages that make up those towns). Hempstead winds up being overrepresented because it looks like every single one of its villages makes it into the list. Hempstead itself shows up, but only the village of the same name that appears within its town.
2. Upstate, there are a lot of towns that have villages within them that have the same name. Unless specifically qualified with "the village of," I've found that people are usually referring to the town. I've found that's the case even when referring to a village that doesn't have a corresponding town of the same name, but happens to be the main center of that town.
For example, I grew up near a village called Washingtonville. I would say I'm from Washingtonville even though my house wasn't technically within the bounds of the village: it was in its town, Blooming Grove. If I'm looking for a map of Washingtonville, I would expect to see not just the comparatively tiny area incorporated as "the Village of Washingtonville", but its surrounding area as well. To make matters more confusing, there is a hamlet of Blooming Grove within the town of Blooming Grove as well, but nobody cares about it: I couldn't name 3 people would would ever want a map of the hamlet of Blooming Grove.
Another example is Wallkill, which is even more confusing. There's a town of Wallkill in Orange County: within it, there's the city of Middletown. If you're from there, you're going to say you're from Middletown unless speaking to someone from the area who could tell the two apart. There's also a hamlet of Wallkill in another county, part of a different town. If someone says they're from Wallkill, they're probably talking about the hamlet, not the town. You'd say you're from Wallkill (and not the town in which it resides, Shawangunk) because... well, I don't know. That's the way it's always been.
Which brings me to the exception I mentioned earlier. Check out the listing for Monroe, NY: https://www.projectjefferson.com/?city=1804311
Three entities in New York share this name: a town in Orange county, a village within that town, and a county 300 miles northwest of the town that encompasses the city of Rochester and its suburbs. (There is no city of Monroe in New York.) The site shows a map for the county. Presumably, the list being worked off is actually referring to the town, which has a population of ~39,000 and not its village, which only has a population of ~6,000.
Part in parcel with Monroe and the confusing town/village distinction, Kiryas Joel is listed (https://www.projectjefferson.com/?city=175464). Kiryas Joel is a village within the town of Monroe comprising of about half its population. Even though it's massive, if you're from Kiryas Joel, you say you're from Kiryas Joel. But if you're from anywhere else within the town of Monroe, you say you're from Monroe. This is for cultural reasons: Kiryas Joel is a very insulated village consisting of mainly people of the Hassidic Jewish faith. If you're not Hassidic, you wouldn't self-identify with it even though it's the largest village within the town.
This was long, but hopefully it demonstrates some of the perils of taking an automated-only approach to this type of mapping.
Mark, you are incredible. I think you've hit the nail on the head in terms of certain locales needing more TLC. I think we can get away with the fully-automated approach for most of the maps, but there are certain regions where this approach more or less falls apart.
The one thing we say -- and mean quite seriously -- is that we are willing to generate maps of wherever you most desire. We've already begun the process of rendering custom areas for customers, all requested through the contact form of the site. I think we'll slowly refine maps of certain areas as time goes on. We wanted to cast the net wide and fast, but I think we can hone in from here on out.
Thanks for your thoughts Mark!
The one thing we say -- and mean quite seriously -- is that we are willing to generate maps of wherever you most desire. We've already begun the process of rendering custom areas for customers, all requested through the contact form of the site. I think we'll slowly refine maps of certain areas as time goes on. We wanted to cast the net wide and fast, but I think we can hone in from here on out.
Thanks for your thoughts Mark!
> I think we can get away with the fully-automated approach for most of the maps
Even though it might look nice to outsiders, I firmly believe you can never get representations of cultural and human data really right if you're doing them automatically.
Such data will always be full of quirks, and representing it will always need some tweaking if you want to get it perfect. Say, moving a label a bit further left so it doesn't hide a monument, displaying a street wider than its actual designation because it's an important way, not showing at all another path because it's disconnected from the rest of the network and really not that important (or displaying the more minor path that connects this one to the rest even though its designation means it should be hidden).
I really think a lot of people who make maps do not acknowledge the fact that human data is not something that is completely normalized, not normalizable, and sometimes might not be able to be processed with a set of rules, however fine.
Even though it might look nice to outsiders, I firmly believe you can never get representations of cultural and human data really right if you're doing them automatically.
Such data will always be full of quirks, and representing it will always need some tweaking if you want to get it perfect. Say, moving a label a bit further left so it doesn't hide a monument, displaying a street wider than its actual designation because it's an important way, not showing at all another path because it's disconnected from the rest of the network and really not that important (or displaying the more minor path that connects this one to the rest even though its designation means it should be hidden).
I really think a lot of people who make maps do not acknowledge the fact that human data is not something that is completely normalized, not normalizable, and sometimes might not be able to be processed with a set of rules, however fine.
No problem: I think maybe part of the problem is treating state-level subdivisions the same nationally. For New York, perhaps a decent compromise would be to map by town outside of New York City. Other states don't necessarily have the same concept of what a town is that New York has, but have some other type of subdivision that makes more sense for them. I bet it'd be a lot easier if the US had a system like NUTS: http://en.wikipedia.org/wiki/Nomenclature_of_Territorial_Uni...
Some cities require a larger footprint. For example, Atlanta should include the entire I-285 Perimeter.
Also, cities that are longer east-west than north-south should be in landscape, instead of portrait, orientation. Berkeley, CA and Cambridge, MA are examples. (And the San Francisco print should probably be a square.)
It is unfortunate that the name of your project honors a slave owner.
When using tribal or other demographic names, I'm always reminded of the argument once put against the Jeep Cherokee: there's no such thing as a Ford Caucasian or a Honda Negro. And for fairly obvious reasons. Or, if instead the names were European demonyms, you'd still have people asking why that colour scheme goes with that name. If it was 'French' instead of 'Shoshone', people would ask "why is this garish scheme associated with them?". If it was 'Czech' instead of 'Blackfeet', people would be asking why the nighttime colours are associated with that demographic.
As far as I can tell - and I could be quite wrong here - the demonyms used have nothing to do with the associated schemes, and are just 'friendly names'. I'd find a different naming scheme, where the names semantically matched the colours. For example, Bannock is proposed as a tribute to Warhol, so call it something like 'Warhol' (though maybe that exact name might also run you into copyright issues... Perhaps 'Soupcan'?)
Also, the reason for the colour scheme should be the first paragraph in the associated blurbs, not the third. The other two paragraphs are boilerplate and inherently less interesting when you're looking through the series.