What is crowdsourcing?
Let us start with a simple definition.
“Crowdsourcing is the process of getting work or funding, usually online, from a crowd of people. The word is a combination of the words ‘crowd’ and ‘outsourcing’. The idea is to take work and outsource it to a crowd of workers.” [1]

the-great-crowdsourcing-tt-2011

Crowdsourcing Illustration from [1]

Moreover, one can use financial support to projects by using crowdfunding.

A good illustration is Kickstarter, an online platform where creators can post projects’ ideas with their respective goal and budget. The crowd can choose to support their project by pledging money.
Therefore, crowdsourcing and crowdfunding are new ways of connecting people into a worldwide scale project!
Let us now summarize three articles (accessible here) and later on compare their experiences of crowdsourcing.

Crowdsourcing Performing Arts History with NYPL’s Ensemble

The New York Public Library (NYPL) for the Performing Arts would like to create a digital database “Ensemble” containing approximately 125’000 dance programs, 400’000 music programs and one million theatre programs (from the late ninetieth century till now).
“Ensemble” will improve our knowledge of the arts in New York from the late ninetieth century.
Until now, researchers had to come all the way to NYPL in order to examine the programs, and their examinations had to be very careful.
Therefore, NYPL decided to create a digital database: “Ensemble”, using crowdsourcing.

NYPL concluded that the beta version “Ensemble” performed poorly (participation too low compared to another crowdsourcing project from NYPL). This may be explained by the difficulty of the task: indeed, the person in charge of the digitalization process of a program should have to determine the playwright or the producer in addition to the transcription!
Another issue is that “Ensemble” requires several agreements from users in order to accept a transcription.
Furthermore, the tests showed that the first users transcribing a program felt insulted because the degree of confidence of their transcriptions was low. Indeed, this degree of confidence represents the number of people who agreed with the transcription.

Swiss Voice App: A Smartphone Application for Crowdsourcing Swiss German Dialect Data

The University of Zurich launched a smartphone app “Dialäkt App” offering the user several variants of 16 words. The user selects the variant corresponding to his dialect. The user has the possibility to record his own word’s pronunciation.
The University of Zurich is developing “Voice App”, a more advanced version of “Dialäkt App” which is working with automatic speech recognition (ASR).
“Voice app” crowdsourcing experience could be used to compare dialects and establish population statistics.
The user records words through his phone and “Voice App” identifies the pronunciation variant.
ASR is based on the data collected by “Dialäkt App”: crowdsourcing.

The accuracy of the ASR depends first of all, on the number of different word’s pronounciations “Dialäkt App” has.
Secondly, ASR might encounter some issues regarding the distance the user has from the microphone and the noise of the environment he is in.

Mixing Contributions, Collaborations and Co-Creation: Participatory Archaeology Through Crowd-Sourcing

MicroPasts is a project developed between the University College London and the British Museum.
Its aim is to gather together traditional academics and established groups around archaeology on a communal web platform.
The main subjects treated on this platform can be classified as follow:
“Co-production of open licensed research data; Collaborative development of completely new research projects and Crowd-funding.” [5]
MicroPasts employes several “crowd-oriented” platforms in order to finance its projects (Crowfunding) and develop them (CrowdCrafting).

MicroPasts aims at understanding what motivates community engagement of the contributors. Is money a main motivator? Or the cultural experience that drives the crowd into contributors?

Conclusion

One can see that the first two articles use crowdsourcing in order to transcribe programs for the first article and digitalize voice for the second one.
Both projects encountered issues related to crowdsourcing. Indeed, in order to be accurate, one’s work has to wait for other users approval and contributors might be frustrated when their transcription’s trust degree is low!
Moreover, crowdsourcing used for voice recognition encounters technical problems due to the crowd’s microphone quality.
Those issues leave us to the degree of trust we should put on the contributors of crowdsourcing.
Finally, the third article questions the motivations behind the contributors. Indeed, is it money related? Or more cultural experiences linked?

References

[1] What is Crowdsourcing?
http://dailycrowdsource.com/training/crowdsourcing/what-is-crowdsourcing

[2] Crowdsourcing illustration.
http://fr.slideshare.net/manikkinra/crowdsourcing-24677173

[3] Crowdsourcing Performing Arts History with NYPL’s Ensemble
Reside, Doug.

[4] Swiss Voice App: A Smartphone Application for Crowdsourcing Swiss German Dialect Data
Kolly, Marie-José; Leemann, Adrian; Dellwo, Volker; Goldman, Jean-Philippe; Hove, Ingrid; Almajai, Ibrahim.

[5] Mixing Contributions, Collaborations and Co-Creation: Participatory Archaeology Through Crowd-Sourcing
Pett, Daniel Edward John; Bonacchin, Chiara; Bevan, Andy.

Advertisements