Responses to remixing on a social media sharing website

From AcaWiki
Jump to: navigation, search

Citation: Benjamin Mako Hill, Andrés Monroy-Hernández, Kristina R. Olson (2010/05/16) Responses to remixing on a social media sharing website. Fourth International AAAI Conference on Weblogs and Social Media (RSS)


Tagged: Computer Science (RSS) sharing (RSS), remix (RSS), scratch (RSS)


Attempt to survey and explain the reactions of users to remixing of animated stories, interactive art, and video games on Scratch. Licenses and norms encourage remixing on the site, but user reactions vary.

In first study, authors used Scratch metadata to discover creation-remix pairs and (11,861 projects out of 136,929 from the first 13 months of Scratch) and two independent coders coded comments left on the first 3,555 projects by the author of original projects on each remixed project as: no comment, positive, hinting plagiarism, plagiarism, negative, and none of the above. Coding was found to be reliable (agreement between coders), so the remaining 8,306 projects were split between the coders. Of originators who saw remixes and commented, 7% accused the remixer of plagiarism, 15% hinted at plagiarism concerns, 21% left positive comments, 7% negative comments and 6% none of the above -- positive comments as likely has plagiarism complaints.

The paper's second study tested hypotheses for the wide variety of reactions to remixing using the first study's data: 1) Originators of larger or more complicated contributions will be more likely to accuse remixers of their projects of plagiarism 2) Originators will be less likely to accuse remixers of their projects of plagiarism when the remixed project is itself a remix and 3) Originators will be less likely to accuse remixers of their projects of plagiarism if they have shared at least one remix themselves. Other factors controlled for: "The odds of a female accusing a remixer of plagiarism is less than 0.6 times the odds of males doing so." "[Y]ounger students are indeed more lilely to accuse a remixer of plagiarism". Support found only for hypothesis (1).

In study 3 a random selection of 40 originator-remixer project pairs from each of the 6 categories of comments were coded for similarity. Plagiarism comments were associated with more similar pairs than other types of comments.

Theoretical and practical relevance:

Although remixing has been posited as being increasingly salient, little empirical research on remixing practice has been done.

"To the degree our results generalize ... our findings suggest that the technical and normative permission to create remixes may be insufficient to supporting positive reactions to remixing in a social media remixing community ... charges of plagiarism may be of reduced concern in communities where individual contributions tend to be small ... remixing systems may be able to decrease charges of plagiarism by remixed users by promoting differentiation between remixed projects and their antecedents"

This paper provides ground for future research to build on, eg to "look at the effect that reactions to remixing have on the rate or nature of contributions."