This posting has three main sections contain within it: the first relates to the tasks that members of the public can do in a crowdsourced project. This section will have two subsections, transcribing and correction. The second section will talk about how to attract contributors to the project, and the last section will talk about how to build an inviting interface needed to engage contributors.
Contributing to a Crowdsourcing Project:
Part I: Tasks
Transcribing is one activity that members of the public can take part in during a crowdsourcing project. Transcribing is just as it sounds, this project usually involves many volumes or collections of letters, documents, and statements that are not machine readable – meaning, these documents cannot be read via an OCR or optimal character recognition software. Therefore, a majority of the documents have to be read by a person and then transcribed into a software or interface that is searchable by users of the project site. A prime example of this type of project is The Papers of the War Department. This is a massive collection of late eighteenth century and early nineteenth century War Department documents.
The main job of the volunteer is read the documents (often there are a few in a bundle), and type out what the documents say on a mirroring text enable interface – kind of like having to retype into a Word document a handwritten letter you received. The purpose of this is to allow researchers to use a finding search or topic search to skim through the collection with ease – not having to read document-by-document. Here is a general guide should you feel inclined to help with this project.
The next possible role volunteer contributors can part take in is called correction of data. Simply, this means that users will go through and add metadata on objects. In both project cases, the users are not adding any items, they are working on the collections opened by the project staff. In this case, the user will add corrections to what a machine may have missed or overlooked. Corrections to the metadata are very common, such as adding a locations or year of publications. In this example, we will be looking at the NYC’s Public Library and their effort to turn old atlases into interactive maps. This project is called Building Inspector, and users can do a range of activities, but to the two will focus on are “check footprints” and “fix footprints”.
In this example, both check footprints and fix footprints are computer generated outlines of buildings using old NYC atlases. The computer tempts to read where the buildings are outlined via the old maps; however, sometimes the software blends two or more buildings together. Therefore, under fix footprints, one is able to see how the software initially starts with a rough outline of building, but the user is then to make sure that the outline is correct – not cutting off parts of the building or adding in unwanted parts. Check footprints goes through approved outlines (those that have been approved from the fix footprints category) and allows the users to approve or disapprove of the outline. This approval process is done via a general vote. After a building has its outline checked by 10 different people, then that outline is approved if a majority say the outline is correct. This is a perfect example as to how crowdsourcing projects offer the opportunity for self-police. Just like in Wikipedia vandalism, Building Inspector allows the other volunteer users to correct the mistakes of others.
Part II: Attraction
The question is how to attract contributors? In this case, crowdsourcing projects have many participants because they offer a unique or interesting collections for users to engage with. In the case of the Papers of the War Department, Dr. Sharon Leon mentioned that, counter to popular belief, a project does not have to have thousands of participants for the crowdsourcing to be worth it. In fact, there needs to only be a few/handful of dedicated enthusiasts willing to put in the effort.
Moreover, Mia Ridge’s article “Crowdsourcing our Cultural Heritage” states that there also needs to be a way for contributors to track their progress. This tracking of the progress allows the users to see, in real time, their contribution to the project, and how their work fits in with the other work being done. In essence, the goal is to show contributors that their work is important to the health of the overall project.
Part III: Interface and Community Building
Furthermore, the design of the interface is very important. While my discussion on The Papers project and its read-and-type method of an interface might sound simple, it is, perhaps, not that straight forward for most. Yes, reading and then typing projects offer the project designers an easy way to create the interactive interface. Simply scan the documents and then provide a text box for the users to re-type what they have just read. Moreover, The Papers project is very well designed in the fact that it provides step by step direction with pictures. I believe pictures help, this allows the project to reach a larger audience.
This type of interface awareness is important because Dr. Marie-Louise Ayres of Australia’s Trove project in her article “Signing for their Supper” notes that many users on her project are older individuals looking for family or local history. Therefore, this is another example as to why interfaces need to be easy to understand and use. If the interface is written in HTML or one would need to code-in the information, then that leaves out or limits your possible volunteers. Overall, if volunteer editors are contribution a great deal of time to your project, it is best to have a way for them to create a local account on your project’s website. This account will keep track of the work in progress and the work done. Moreover, the ability to have your name attached to the work done is another way for contributors to receive their deserved credit.