ME.Digital Humanities Institute: Weeks 5-6: Mapping

Over the next two weeks, we will explore the examples of mapping tools along a hands-on introduction to the open source mapping software QGIS. Also, please remember to complete the survey, about directions for the final two weeks of the institute.

I suggest starting off by reviewing this excellent article, "We mapped it so you don't have to," from University of Georgia Digital Humanities Coordinator Emily McGinn and Librarian Meagan Duever, as it gives a very good overview of GIS tools. 

To make the lesson concrete, browse through some of the mapping examples below, and then again when you need to take a break from the tutorial. The tutorial was originally designed as a clasroom-based tutorial, and I have found some gaps in explanations that can make it frustrating to work with (i.e. in the classroom, the instructor would say, "Oh, just 'do this').  See my section "Tutorial Notes" below for some of those issues that I have found, and just keep in mind that this resource is there as you work through the tutorial

You will need to download QGIS, with instructions on how to do so here.

QGIS is one of several widely using mapping application tools. It is a desktop-based tool, meaning you are working offline, and the software does not integrate into an online environment. However, you can create PDF documents of the mapping visualizations you create and employ those in varied ways to integrate into a website.  We are primarily working with QGIS because it is free and open source, and the "lessons learned" apply across all tools in terms of basic vocabulary and terminology.  This lesson is intended to be an introduction to the concepts of mapping, rather than all-out effort to master one particular piece of software. Probably most of us won't get through the entire tutorial. If you feel like you've had enough, feel free to skim through the second half to get a sense of where the tutorial is going.

Another widely used mapping tool is ArcGIS (see examples below from College of the Atlantic). Many towns, companies, and other organizations are now using ArcGIS, which is pricey but easier to work with (in my humble opinion), especially in its latest and greatest cloud-based version. However, ArcGIS is beyond the means of most small organizations. The UMaine System  has a consortium-based subscription that divvies up the cost, so keep that in mind if you want to continue working with mapping.  (I'm currently scheming to get my insitution in on the subscription). See the "Resources" links below for similar tools.

Here is the link to the QGIS curriculum (and remember, you need to install the software first).

The lesson begins with vocabulary, and I'm going to review some here, because the names of different things gets confusing.  I think previewing some of the vocab will help as you go through the lesson. Some key vocabulary includes the following:

.CSV file -- The comma-separated value file is the foundation for most mapping work and other data visualizations. CSV is essentially an excel spreadsheet that is converted in machine-readable form, with piece of information separated by commas. To humans, it looks unreadable (even though we can pick things out), but this is how the computer processes the data/information.  Generally speaking, if working with a large data set (like that NY Times Public Library set), you might be working in Excel or a similar program, and then saving it as ("Save As") a CSV file when you want a machine to read it, i.e. on the command line, or in mapping software, or many other ways. For example, in a storytelling application like StoryMaps, you can keep your entire data set on an excel spreadsheet, convert it to a CSV file, and then continuously work with these tools as you work at updating and maintaining a StoryMap (as long as you have your own website-- not Flickr or WordPress-- to store images).

Layers -- The lesson uses a great metaphor of the old overhead projector with the acetate plastic sheets. Each data set creates a layer, then you lay the layers on top of one another to see and understand your data. Each layer must contain objects from the same category. For example, if you were mapping 19th century transportation networks in York County, you would have one layer for the trolley lines, another layer for railroads, another for ferry routes and another for major roadways. When you put those all of these layers together (layer them on top of each other) they represent the late 19th century transportation system in York County.

Vector layer --  GIS uses points, lines, and polygons, also known as vector data. A vector layer displays data either as points, lines or polygons. Each layer can only visualize one type of feature: points, lines or polygons. Points may be towns or telephone poles; lines could represent rivers, roads, or railroads; and polygons could encompass a farmer’s lot, town lines, or state borders. Historical data can be attached to this geographical information to trace or learn more about how humans interacted with geography.

Shape file -- A vector layer is packaged up in a "shape" file, .shp, and the Shape file interacts several other files it gets bundled up with. We don't need to know the specifics of all those other files, just that they all work together so they have to all stick around together in one folder.

Feature -- If you are "thinking in spreadsheets", the features are going to be the column on the left side of your spreadsheet, i.e. these are main category of items for which you are gathering and organizing data. A feature, in GIS terms, is tied to the map as a point, line or polygon.  For example, if you were tracking the medieval travels of Moroccan traveler Ibn Battuta across Asia, one set of features might be the places he visited, i.e. "points."  Another separate file might include the routes he took, i.e. lines.  And a third set might be the list of countries he visited presented on a modern day map. The features, in a sense, serve to delineated the "main characters" of your data.

Attribute -- Again, if you are "thinking in spreadsheets," the attributes will be the data and information that you are trying to gather for every feature. With Moroccan travel Ibn Battuta, if cities were my features, my attributes might include 13th century population estimates; snippets of Battuta text describing encounters, and medieval images of these places.

My colleague at CUNY Grad Center Olivia provided another example of the distinction between features and attributes. If a researcher is mapping school districts, then the feature would be the polygons on the map that represent the districts, and the attributes would be the information that describes certain aspects of the school district, such as the racial demographic of the students or the number of teachers employed.

Raster— usually a .tiff image, a raster image is static geo-coded image (e.g. a map) on which other vector files might be overlaid. Example: I might take an 1865 historical map of York County, geocode it, and then use it as the base map for overlaying 1870 census data to show population density.  (see links below for geocoding resources; it is not hard to get the idea, but in the case of my York County map, the Library of Congress had both a regular .jpg image and a raster image for this particular map). 

In research projects, rasters are used to analyze continuous data that does not change, such as elevation, and thus rasters tend to be used by people studying physical geography. If you wanted to dig into more details about raster data, here's an explanation from ArcGIS: http://desktop.arcgis.com/en/arcmap/10.3/manage-data/raster-and-images/what-is-raster-data.htm.