Who? What? Where? When? and Why? While these make up a catchy song from Spy Kids, they’re also questions that can get lost when looking at digital images, especially when metadata is missing. PastPin wants to help answer these questions, by tagging the location and time of vintage images on Flickr Commons, with the hope that one day they will be searchable through the Where? and When? of the images themselves. By doing this, PastPin wants to create new ways to do research using public domain images online.
Created by Geopast — a genealogy service — PastPin uses 6,806,043 images from 115 cultural institutions hosted on Flickr. When a user brings up the PastPin website, they’ll be prompted with images that PastPin believes come from your geographic area. When you click on an image, you can then search a map for its specific location and enter in a date, which is then saved. The image then becomes searchable by PastPin users through the entered information. The hope is that all of these images will be identified, so that all users can search through location or date.
Some images are easier to geolocate and date than others. PastPin pulls in metadata and written descriptions from Flickr, so images that are published by an institution — such as the University Laboratory High School, like several images I encountered — may already have this information readily available, making it easy to type that into the map and save it. Other images become more difficult to locate or date because they lack that information, and take more outside knowledge to suss out. PastPin also lacks adequate guidelines for locations, in particular. As many of the images that come from the University of Illinois are from digitized books, are they looking for the location of where the book was printed? Or of the library it resides in? It’s unclear.
PastPin faces what would seem like a Herculean feat. As I’m writing this, only 1.79% of the nearly seven million images have been located so far, and 2.13% have been dated. Today, there have been 18 updates, including two that I made, so the work moves slowly.
Still, PastPin is an awesome example of the power of crowd-sourced projects, and the potential of new thinking to change the way that we do research. The Internet creates so many new opportunities for kinds of research, and the ability to search through public domain images in new ways is just one of them.
Do you know of other websites that are trying to crowd source data? How about websites that are trying to push research into new directions? Let us know in the comments!