Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Tagging the YouTube War
Transcript of Tagging the YouTube War
- Maj. Matt McLaughlin (Jan. 2006)
Tagging the YouTube War
"I usually did basic searches on YouTube and Google to find them. Now, in those searches it is true that many other videos came up that were not shot by soldiers. They could have been TV reports on the Iraq war or videos by soldiers who were not in Iraq, or even videos uploaded by people who found videos elsewhere.
I don't usually rely much on tags for finding content because they are not reliable, people make mistakes in categorizing them, etc."
--Personal interview with Mark Glaser on 04/28/2008
YouTube.com as it appeared on April 28th, 2005
Dick Cheney as he appeared on May 30th, 2005
"While many other sites and video repositories online have video from Iraq soldiers, YouTube makes the process simple for uploading video and finding video by searching. You can also filter your search results by most recently uploaded videos, most watched videos or highest rated videos by viewers. And there's also a comments section for each video, so people who watch them can give feedback."
--Mark Glaser January 2006
"Because the U.S. military allows soldiers to take cameras into combat, and gives them Internet access at their bases, the soldiers can easily shoot, edit and upload videos from the war zone. And the technology itself has also become cheaper and easier to use, from the cameras to the editing software. So if you search for iraq soldier on YouTube, you'll get 800-plus results. Many of the videos are edited together like music videos, while others include pranks played by soldiers, such as this one where they push over a port-o-potty with a soldier inside. "
--Mark Glaser August 2006
How is it that so many independent contributors to YouTube have been able to seize on enough common language in describing their submissions to ensure that their videos would successfully be retrieved among a collection (i.e. a group of retrieved results) and identified by millions of users as being topically relevant (or at least putatively relevant) to the war in Iraq?
How did soldier-contributed Iraq War videos grow from the first few posts to a manifest “collection” and what does this process have to teach us about the ability of people to share information with one another in large networked information spaces?
Did contributors settle on a set of terms to apply to videos they uploaded, and if so, was the similarity a function of an already shared vocabulary, or did early posts influence the terms used to describe videos that followed?
Research Question Cont.
Our roles as mediators are increasingly occurring along the points of information exchange, use and interaction and less so at the point of origin or creation.
Keyword tagging of information objects varies widely in application and use, but generally refers to user generated, and sometimes collaboratively derived keywords, applied to some information object for the purpose of personal organization and/or resource sharing.
Motivations, strategies and structures of tagging systems (Huberman & Golder, 2006; JC Paolillo & Penumarthy, 2007; John Paolillo, 2008; Halpin, et al 2007; Marlow, et al, 2006).
Tagging as means for knowledge discovery (Aschoff, Schmalhofer, & van Elst, 2004; Carmagnola, Cena, Cortassa, Gena, & Torre, 2007; Mika, 2007),
Tagging and information retrieval (Yanbe, et al, 2007a, 2007b; Bao et al, 2007; Choy & Lui, 2006; Heymann, et al, 2008)
Tagging generally as a method of classification (Aschoff et al, 2004; Carmagnola et al, 2007; Sen et al, 2006).
I propose that tracing the history of collections such as the Iraq War videos on YouTube, will be instructive in understanding the methods, motivations and means by which individuals share content online and the degree to which these factors help to foster or hamper the discovery of information.
The phenomenon of video sharing from within and about the war in Iraq by large numbers of contributors offering many perspectives of the conflict has arguably become one of the more socially and historically important events yet to be chronicled with so-called Web 2.0 technologies such as blogs, social networking sites, and media sharing sites,
What does the research tell us about tagging?
Given a sufficient number of users and tags, the proportions of tag terms tend to stablize over time. Past a certain point, further tagging only reinforces the existing collective catergorization scheme (Huberman & Golder, 2006; Halpin et al, 2007).
Tagging distributions stablize into power law distributions with limited numbers of stable tags and many more infrequently used terms (Halpin et al, 2007).
The stability of the few consistently used tag terms is thought to happen because of either imitation or shared knowledge (Huberman & Golder, 2006).
Tagging Research Continued...
Typically the first tags used in a series of tags are the ones that have the highest frequency for a given resource (Golder & Humberman).
Implementations of tagging systems vary markedly in system design, user incentives and interface features. These differences impact, to a large degree, how tagging works on any given system (Marlow, et al, 2006).
While tagging has shown some promise as way of augmenting or improving infromation retrieval systems, the value of tagging to IR is still up for debate (Yanbe, et al, 2007a, 2007b; Bao et al, 2007; Choy & Lui, 2006; Heymann, et al, 2008).
Tagging Research Continued...
Most tagging research has looked at data from delicious or flickr.
The tension that lies between the potential usefulness of tags to expand a classification system in ways that benefit a community of users versus the potential that tags may only make sense to an individual tagger or small subset of taggers is central to most tagging research and discussions about tagging.
"A shared pool of tagged resources enhances the metadata for all users...These systems may offer a way to overcome the Vocabular Problem - first articulated by George Furnas (1987) et al..." (Marlow et al, 2006)
"tagging...has the potential to exacerbate the problems associated with the fuzziness of linguistic and cognitive boundaries. As all taggers' contributions collectively produce a larger classification system, that system consists of idiosyncratically personal categories as well as those that are widely agreed upon" (Golder & Huberman, 2006)
"[Web] domains are often highly correlated with particular tags and vice versa...It may be more efficient to train librarians to label domains than to ask users to tag pages" (Heymann, 2008)
Consider the Following Quotes
What the Research Isn't Saying
The "fuzziness of linguistic and cognitive boundaries" are typically not dealt with in depth in tagging research. The theoretical foundations of so-called semantic "problems" [polysemy, synonymy and basic level variation], the notion of vagueness, and the understanding of concepts (if discussed at all) seem secondary to the consquences they pose to query formulation and other search functions.
Little is known about the degree to which a given tagging implementation or system design impacts tagging habits and structures, thus what we "know" about tagging may be strongly determined by the system where it is used.
Researchers working with large data sets from Web sites can only really conjecture about the purpose or meaning behind what is meant by a tag.
What I've Been Doing
Using the YouTube API to gather videos and data from searches for terms related to the Iraq War.
Studying the tag terms, other metadata terms and usage information over time for videos retreived by these searches.
Is the structure of YouTube tagging stable? If so, how?
What might usage data tell us about the implications of user-derived metadata on discovery?
What insights might we glean from understanding these data that may reinforce and/or challenge our understanding of how people classify, make meaning, identify concepts and deal with vagueness?
Questions I Hope To Answer
Theoretical Frames for My Work
Eleanor Rosch's Protype theory and the notions of basic objects in natural categories.
Theories of catagories and cognition: Labov, Rosch, Lakoff
Scaling and Human Language; Cancho et al.