Using many geo-coordinate features - as you see for instance in Google Maps Sattelite Imagery - for Hârn is actually quite simple. All the relevant information can be stored in a format called GeoTIFF. You load a bundle of them into the relevant program (see below) and you are set. Further features, such as showing the maps in a browser, may require more effort.
The archive you can download from this page contains a couple of GeoTIFF files, which I can freely distribute. You will also see a CONTENTS.md file in the archive, which shows you how many more are available. I cannot publish them without the proper license. Many of them are based on lythia.com images and since they can freely be downloaded, you can send me a private message and I will forward the GeoTIFF files you need. That assumes that you actually could create the files yourself: you have the base material and in the following I will provide you with the procedure. This procedure you can apply to any other image you want to, including official maps you bought from Columbia Games or Kelestia.com.
I have added the database for the points of interest I have geo-located in this manner. You can find them in the workspaces/kethira subdirectory. You can import them into qgis or geoserver in vector data. There has been some discussion in the Harnforum about projection and it is far from obvious which one is canonical. I therefore added a basic Hârn image with geo coordinates for your convenience to more easily understand where everything is located in “my” projection. Again, import as raster data into the tool of your choice.
Installing qgis is not difficult. For instance, on Ubuntu, all it
takes is sudo apt-get install qgis. Also see the QGIS english
download page.
If all you want to do is get started, load the images into qgis.
You are set.
Without preparing at least kethira.tif, you will most likely only see an empty page!
In the following we use GEOSERVER_HOME to denote the place were you
have installed geoserver. If you only use qgis, ignore the
paragraphs for geoserver; this will be the directory you unpack the
archive to.
Unpack the archive directly in GEOSERVER_HOME/data_dir such that
workspaces from the archive maps to workspaces in data_dir and
www from the archive to www in data_dir. In other words, such
that the enclosing cryptic directory is not used. This will create
(or update) a directory called
GEOSERVER_HOME/data_dir/workspaces/images.
To create your own GeoTIFFs, you need make sure you also have the extension Raster -> Geo referencing available. For the base images use the format png with an alpha channel.
Start qgis and open Raster -> Geo referencing. This will ask you
for an image file. Use the one you just created. For each image you
will have either a corresponding points file in
GEOSERVER_HOME/data_dir/kethira/points/, you have created your own.
Load that. Alteratively, you can pick anchor points from another map
you already loaded, but for that you must, well, load a pre-existing
map. (They point files provided were all used by myself when creating
the GeoTIFFs from CONTENTS.md.)
Two more things need to be done, before you can create the data image
file by pressing the Go icon (second menu icon from the left), which
will be a so-called GeoTIFF. First, select the destination raster.
This should be the corresponding entry in
GEOSERVER_HOME/data_dir/workspaces/images/, so geoserver can immediately pick
it up. Second, select EPSG:4326 as Destination CRS. This should
always be the case and should even be the default in qgis.
For some of the main files, more detailed explanations are given below.
As said before, I suggest to install geoserver. That makes the “daily work” much easier, as you can view the maps through the browser of your choice. Depending on the OS you use, installation looks differently. Install as described here: Geoserver Installation
Once you have prepared the necessary files, you need to start geoserver in case you haven’t done so yet. (Starting is part of the installation instructions, so it is likely that you already did so.)
You then should create data stores according to the sub-directories (in images) as Image Mosaic and publish them as layers of the same names. Direct your browser to
http://localhost:8080/geoserver/www/kethira/maps.html
and the Kethiran base map will appear.
I extracted a 1418x471 image from kethira200e.pdf, page 6. Cut off the pole caps. (kelestia.com)
I extracted a 8967x6976 image from VenariveMap_100.pdf. (kelestia.com)
I extracted and scaled an image with dimensions 12000x8000 from Harn_Regional_Interactive_Map_201.pdf. (kelestia.com)
Alternatively I extracted an image with dimensions 16752x11514 from
the Harn map inside the Atlas-Update-pdfs.zip. However use
harn2.points instead of harn.points, it is prepared for that
image; it covers a slightly different area than the other map.
(columbiagames.com)
For any other image, the extracting process is similar. If you provide me with proof that you “own” a version of the commercial images, I will send the converted GeoTIFF to you (until I get too many requests). For community images I will send them on request (assuming I have them). If I get many requests, I will ask you to contact the copyright owners such that I can include the images here and make additional, separate requests obsolete. For community images this is generally not a problem. In fact, my own I have already included. For the commercial images, I expect the process to be different. Maybe they will provide their own installation in the future.
The generic community maps I converted are all arbitrarily placed, where it suited my campaign. The naming is such that the first number is the edition (e.g. pottage3) and the second number the page. They serve as backdrops for the interior maps, so even if you place them elsewhare the relationship between local and interior map is conserved. See also Community Maps.
These maps are mostly obtained from lythia.com. I hope to host the converted maps in the future. Until then you can get them from me privately through email, if you prove you are aware of who alleges Copyright.
Basically this works as the general import for pre-existing maps. You have to create the point files yourself. To get the scale right, make sure your map is of a standard resolution (local, interior, …). You can then reuse any of the existing points files, where you keep the relation between
horizontal pixel difference : longitude difference
and
vertical pixel difference : latitude difference
and then shift the whole set according to where you want it. A point of advice: qgis doesn’t really show enough significant digits) maybe there is a setting I didn’t find) so I did mine with an ASCII text editor and imported the point file.
If have a bit of experience with scripts (perl, bash, python, …), the included img2geotif.pl may be of help. It reduces the effort you need to spend to find pixel differences and also does not require QGIS as it creates GeoTIFF files directly. It is the reason why I don’t provide point files for all images listed in CONTENT.md. You will need to install a the perl module Image::Magick.
The harn-webmap project makes use of the following software packages. Without them, this project would not have been possible.