Tags

, , , , , , , , , ,

With the rise of web sites like Kickstarter or Indiegogo for crowdsourcing, we learned that thanks to the Internet people from different parts of the globe can gather to create, fund or be a part of a project.

Short summaries of the papers:

1. Crowdsourcing for performing arts history:

New York Public Library (NYPL) has a huge collection of programs of theatre, music and dance events in the United States since the end of the Civil War. The problem with such a gathering of knowledge is that it can be cumbersome to retrieve informations about multiple events with a common factor.
With the goals of preserving and spreading that knowledge, the NYPL came up with project Ensemble to collect, transcribe and make accessible every programme in the library.
Needing fast and cheap ways to upload the maximum of data in order to begin, they scanned reels of microfilm dating from 1860 to 1930 with content that was not physically in the collection.

One of the problems with the content itself is that it asks of the contributor more than any other crowd-sourcing based project since one has to sometimes interpret the title of the event to identify some important information as opposed to most projects that fall in the same category which ask only to transcribe without understanding or take photos. This makes it exclusive and implies low participation.

So the next objective of Ensemble is to gamify the project by giving easier tasks for less experienced people and harder ones for enthusiasts. The other clever mechanism used is to create user councils to judge of the quality of a transcription leading to its entry into the database or its revision.

Once this project is complete, it would be completely open to everyone to make an application to get all musicians that played Tchaikovsky’s Swan Lake in Maryland between 1910 and 1930 for example.

2. Participatory archaeology:

The Micropasts project is a web platform for gathering all kind of people from the neophyte to the enthusiast of archaeology.
It allows various ways to be a part of a project such as co-producing a project, collaborative development of a new project or crowdfunding initiatives.

The ultimate goal of Micropasts is to have a unified platform for creating innovative, new and voluntary projects using the crowd and an open model for everyone to use in the future.
They use several tools for the technical aspects:

An other goal was to study and get informations on what motivates the contributors to help Micropasts, what is the nature of the relationships created in the community and what kind of resources have been used by the contributors themselves.

3. Swiss Voice App:

The Voice Äpp is based on the crowdsourced Dialäkt Äpp. This application makes swiss-german-speaking users to pronounce a set of 16 chosen words and add the original location of their dialect.

The Voice Äpp uses ASR (Automatic Speech Recognition) to tell apart the different dialects of its users. The problem with ASR is that it is supposed to get every word said independently from the accent, intonations or phonetics variants of the speaker.
This is where Dialäkt Äpp has its role to play. It adds the power of the large database classified by a schema created by phoneticians in order to recognise who is from where.

The real advantage of this kind of crowdsourced project is that it is as useful to the user as it is to the researcher. An application such as Voice Äpp is easily usable later on by the contributor. Any contributor in fact and not just the passionate linguist who wants to study speech pace between Swiss german regions.

Comparison:

These different projects have taken the concept of crowdsourcing and used it in a different way.
For the Ensemble project from NYPL, the task was difficult meaning that most people would need to give content interpretation in order to make a good transcription. There is the control mechanism that gathers different users and they judge the quality of the work done.
The Micropasts project uses the power of the crowd to do work but the contributors would not really see the result of it but it would create a flexible and completely reusable model for future projects.
The Voice Äpp on the other hand gives directly to the contributor the result of its work in Dialäkt Äpp.

References:

  • Pett Daniel Edward John, Bonacchi Chiara, Bevan Andy (2014). Mixing contributions, collaborations and co-creation: Participatory archaeology through crowd-sourcing. http://dharchive.org/paper/DH2014/Paper-852.xml
  • Reside Doug (2014). Crowdsourcing performing arts history with NYPL’s ensemble. http://dharchive.org/paper/DH2014/Paper-131.xml
  • Kolly Marie-José, Leeman Adrian, Dellwo Volker, Goldman Jean-Philippe, Hive Ingrid, Almajai Ibrahim (2014). Swiss voice app: a smartphone application for crowdsourcing swiss german dialect data. http://dharchive.org/paper/DH2014/Paper-819.xml
Advertisements