If this gets enough traction this could make historical maps so much more accessible. Instead of every historian writing their own magnum opus of a particular area and timespan, all the geographical information can be combined in a single system. Is there a system built in for keeping track of annotations/sources? How are debated topics/unresolved locations dealt with?
What if different schools of thought exist for where something was in the past, could multiple hypothesized locations exist for a single object? Direct link with Wikipedia would also be useful
Very nice project, hope it becomes as useful and detailed as the present day OSM
There is already a super accessible presentation of history maps: Paradox games like Europa Universalis, Hearts of Iron, etc. The games achieved what youâre talking about already, thereâs already an old, large and very active community of modders across their games that make historical accuracy (or gameplay-favorable inaccuracy) changes. Itâs possible the mean Europa player, over all the games, individually spent more hours looking and interpreting the historical map in those games than anyone has ever spent looking at any Open Street Map property ever.
If youâre thinking about contributing to OSM History maybe consider EU4 instead. Maybe you think these are different things. I donât. One satisfies your curiosity (ie entertains you) a lot better than the other.
Maybe though OSM History could be a place, like Wikipedia, where people could settle âterritorial disputes.â I donât know, that shit, wherever it appears, is a pretty toxic part of any community. Iâm not 100% sure how Europa deals with it - you can download whatever mod you want, after all, itâs your game. But my feeling is, most of this territorial dispute shit hardly affects the people talking about it at all, and the feelings of playing a video game and all that entails - positive feelings - are better than the negative ones you feel when youâre pursuing what you believe is justice in a Wiki talk page or a Twitter flame war.
What API does Paradox offer for getting information out of their historical games?
I think this is a pretty key question that also applies to any effort by any other commercial provider. The detail in the historical models in EU4, any of Ubisoft's games, etc., is fantastic. I'd love to see any of them make stripped-down versions of their geodata available through open data licensing. Maybe there's something we've missed in the EU4 modding that could be of use? We'll take another look!
It would be fun if it had events/battles listed with dates.
> Instead of every historian writing their own magnum opus of a particular area and timespan, all the geographical information can be combined in a single system.
Isn't that true of all historical data? It seemed so obvious to me that professional historians would share data that I asked several professional historians what the standard format was. Nobody had even considered it, as of ~10 years ago.
Obviously you can't structure all historical information, but you can do a lot of it: Location & time, other things present at that spot in space-time (people, events, weather, diseases, governments, etc. etc.), and sources. And then some unstructured fields, including one for analysis.
I'm sure that's misguided, as I'm not an historian, but it sure seems like much more value could be extracted from research if it was accumulated and shared, and not stuck in someone's personal storage, and that it would save a lot of duplication of effort.
You're right on. I've asked many for shapefiles of the world's historical boundaries & some do exist, but are often encumbered by copyrights / licensing that add friction to sharing. Our goal is to have as much CC0 data as possible, although we have added some CC-BY-SA-NC data. All data is copyright / license marked at the object level, with no label defaulting to CC0.
"What if different schools of thought exist for where something was in the past, could multiple hypothesized locations exist for a single object?"
How to represent such things on a map can become quite complex very fast. A placename from an old document may refer to multiple locations, but these are mutually exclusive. The probability of attribution may vary in each case. Different scholars may disagree about these probabilities. Etc, etc.
> Instead of every historian writing their own magnum opus of a particular area and timespan, all the geographical information can be combined in a single system.
The two options are not mutually exclusive: Dr Shmoe and Dr Kroe can write fascinating books on the movement of the border between Hannover and Braunschweig that mostly agree but disagree on a couple of fundamental points; both could be represented, and perhaps compared, via different views.
For an existence proof: Wikipedia is able to have decent articles about (some) subjects where strong consensus does not exist.
Being Dutch and a history-buff, I have one mayor gripe with historical maps, that sees land as 'moving lines' on a fixed map. When I look at a historical maps (also in books), I often see the Netherlands represented in it's current form, i.e. containing land that is not even 50 years old, being prestent hundreds or even thousands of years. I get that it is an intricate history, and maybe too much effort in some cases, but please at least remove flevoland. https://en.wikipedia.org/wiki/Flevoland
It's true, and it is not just a Dutch problem. Basically all major river systems change throughout the centuries: meandering arms evolve and others get cut off; river deltas grow and change shape and of course human activity shapes the landscape (canals, irrigation, drainage of swamps and floodplains etc.). The ancient city of Eridu for example was a coastal city around 5000BC and today lies hundreds of km inland.
I've recently been thinking about this with regards to borders. There are plenty of borders around the world that rely on rivers as a boundary, but river's aren't static, and meanders shift all the time.
Of particular interest to me was the England/Scotland border, as I was visiting the area, and noticing that it doesn't always follow the path of the river Tweed.
When a river is a border, that is by nature always changing, how do countries decide which bit of land/islands are sovereign?
It seems like this is adding some vectors on a gridded base map. But OSM has every object in a vector form, also coastlines etc. So I don't think there is a technical reason why coastlines would have to be fixed for a historical map. On the timescale of ages and millennia, rivers in delta planes are also usually not very static.
Well... coastlines are handled in a deep, dark, arcane part of OSM planet and data processing, but you're right. Rivers, on the other hand should be straightforward to change. We'll have some cleaned up examples in the near future.
Not only deltas, any river meandering over any plain will change its course pretty "quickly" (which creates problems if such a river is defined as the border between two countries BTW)
I live in the Fens in England and noticed the same thing. You can set it back 1000 years and not only is the Fens still drained, there are artificial drainage channels that weren't dug out for another 600 years. It seems an assumption has been made that rivers and coastlines don't change over time but it's so far from true! For me it's one of the most interesting, and scary, things when looking at old maps. Does the data model support changing natural features like rivers and coastlines?
Yes any feature as far as I'm aware can have a start date and end date so can show the changes over time. As can be seen in London (https://www.openhistoricalmap.org/#map=16/51.5108/-0.1086&la...) where the shape of the river changed when the Embankment was built in 1869. As far as I'm aware there is no reason why changing coastlines couldn't be mapped, though it is a lot of work to draw in a long length of coastline.
I just added a start-date to a canal, so apparently yes.
My favourite historical map resource on the internet is the National Library of Scotland's georeferenced side-by-side viewer: https://maps.nls.uk/geo/explore/side-by-side/#zoom=5&lat=56....
View a variety of historical maps on the left, with the same area "now" on the right, with the pointer synced between the two sides so that you can trace roads, building outlines, etc.
Neat! On zooming in on the UK, I landed mostly by chance on Gatwick Airport, only to discover that it's built on the site of a former racecourse. A quick web search brought me to the Wikipedia page with a potted history of said racecourse, and how Gatwick has named various features after it. Fascinating.
Also, don't miss the swipe view...
I have seen many issues with historical maps on Wikipedia, one of the worst offenders is https://commons.wikimedia.org/wiki/File:Territories_of_Dynas... which is basically propaganda. We desperately need a new one with citations.
I would suggest that, if various portions the vectors could cite evidence, it would be an excellent basis for historic vector maps on Wikipedia. Currently, this whole area is a shambles.
Further, I recently learned some historical maps I had created with great effort for my Tang Dynasty text translation at https://en.wikisource.org/wiki/Translation:Manshu were deleted by some Wikipedia policy-thumber who reckoned including low resolution satellite backgrounds to contextual maps was grounds for erasing them. You'd think being an open source project for academic caliber translations you'd be able to use, you know, open sources under academic fair use. My tolerance for idiots is truly at an all time low.
>I have seen many issues with historical maps on Wikipedia, one of the worst offenders is https://commons.wikimedia.org/wiki/File:Territories_of_Dynas... which is basically propaganda. We desperately need a new one with citations.
can you explain a bit more about that? Not too familiar with the background of Chinese map.
The notion that there is a fixed line between cultures is itself a farce, but, historically the majority of such maps drawn for empires are written using the implied borders from latter-day interpretations of geography that was mentioned by an eventual victor as annexed. However, the reality on the ground is usually much different. In some cases, thousands of years different.
This is a difficult thing to describe for any pre-modern polity, where vassalage and weak frontiers make it difficult to describe if a central power actually controlled a territory. The Ottoman Empire de-facto exercised little control its North African vassals, the Roman Empire had increasingly weak control over its frontiers due to migrations and incursions, etc.
There's a lot of great work in Wikimedia commons, but I think there are 2 primary / very common issues: * Format - many of these maps are .svg or illustrator-created images with no projection information, which makes data extraction very difficult. This is why we're trying to share with GIS-formatted data, so people can do with it what they'd like. * Sourcing - very often, a map is depicted with a source of "own work". That, unfortunately, really isn't good enough & we're trying to source all objects in OHM. What's scary is that putting something in Wikimedia almost ensures wide propagation as "fact", even if it lacks a source. Here's an example... the base map is sourced, but where did this red line come from? It appears to be hand-drawn on a map... who drew it? How'd they get that outline? https://commons.wikimedia.org/wiki/File:Comancheria.jpg
Relatively accurate projection detection should be feasible for some, particularly vector, maps. Perhaps a library for this could be made. A simple approach would be convolution of nearby topographic features by assumed potential projection and then a least-difference comparison with observed positions. I am sure the CIA foreign map processing people have had this stuff for decades already.
I have a relative who has been spending years collecting data from archives and building historical maps of Ostrava, a city in Czech republic [1]. He built nine (9) time based layers (1860-1940). The amount of time and work this takes is just incredible.
He has been considering providing data to OHM, but obviously has not been convinced by this prospect yet, one of the reasons has been loss of control over his data, as there are academic institutions ready to snap this a put their name on it. It was already attempted.
I can imagine there are more researchers like this and the results of their work, all the data, is fragmented somewhere out there... and possibly lost.
This is also a great resource for a project like this: https://www.oldmapsonline.org/
These are mainly links to georeferenced raster images and databases, not vectors / OSM data. But, still -- I've already spent hours browsing old maps of my home city. For anyone interested in contributing to this project, oldmapsonline might be a great reference.
For the UK, the National Library of Scotlandâs historical maps are a fantastic resource:
Irelandâs Ordnance Survey also has some good historical resources:
https://www.osi.ie/products/professional-mapping/historical-...
This is cool. For my hometown Berlin I love to browse this map https://1928.tagesspiegel.de/ it shows Berlin in 1928 and most areas are still the same.
Yeah, it's very intersting indeed. Roads and street layouts usually don't change that much over the centuries -- they get integrated and amended, but seldom disappear. I was pretty amazed to recognize the street layout of Leipzig in the 1740ies: https://biblio.unibe.ch/web-apps/maps/zoomify.php?pic=Ryh_53...
There are a few nice timelapses on youtube of borders of Europe changing: https://www.youtube.com/watch?v=k-2zaOhYlAM
Fascinating stuff. There were some attempts with e.g. Yahoo's Geoplanet to take historical context into account. I know some people who are still working on that data set (which Yahoo released under a creative commons license): https://spelunker.whosonfirst.org/concordances/woe/
I live near a building in Berlin that has a sign "this building used to be in a different country". Also, I live on the only street still intersected by the wall. It's Bergstrasse which is intersected by the wall memorial. I remember listening to the news announcements in the Netherlands when the wall fell when I was a teenager.
Another interesting thing is that borders are kind of imprecise. We've only had GPS for a few decades and most maps predate the existence of that. Land surveillance to modern standards pretty much developed only in the last 200 years (e.g. Ordnance Survey in the UK is one of the older examples of an institute specializing in that). Post colonial dividing up of territory in e.g. Africa was done with rulers without too much regard for history, culture, or ethnical divisions. The border between the US and Canada is similar in some places. Also internal borders for some federal states are funny like that.
Besides, stuff moves around on this planet by centimeters per year in some areas. The meridian does not actually run through the little monument they have in Greenwhich for that (by over 100m). It hasn't for a long time. Rivers, which are commonly used for borders, move as well. There are some minor border disputes all over Europe (and a few major ones) as well as occasional pragmatic agreements between e.g. Belgium and the Netherlands on how to administer certain regions. Baarle Nassau of cause being the most surreal example where we are talking exclaves, enclaves and enclaves containing enclaves, etc. Some houses cross borders and have entrances/exits in different countries. It's completely nuts.
We LOVE those youtube timelines. Unfortunately, you can't zoom in & you cannot hit pause or easily scrub back and forth.
We're also big fans of WhosOnFirst!
Oh wow. I am literally slowly working on something like this as well. But my knowledge of mapping technology is very limited.
My focus was actually much more on trying to use partly public data, and partly contributed data in order to 'render' information on top of such a map. My vision was basically to have all historically known people and there locations being visible on the map. Wikidata already has a nice start to this, but you would need a contributor model to flesh this idea out quite a bit more.
Another step after that would be the movement of armies. This could be really interesting if you could see armies moving around at different times and getting closer to each other, or pass each other by.
City population is another interesting thing. There is no easy way today to get a good estimate of what some city X population might have been at any particular period. That information is buried in all all over the place.
There are many more ideas, but having his historical map is actually awesome for me, because that was the part that I know least about. So it should be an amazing piece of the puzzle.
I will defiantly try to contribute to effort.
Get a daily email with the the top stories from Hacker News. No spam, unsubscribe at any time.
