Reading Time: 4 minutes read

I attended, MediaX

Held at Stanford university to share emerging academic research. The notes from the MediaX talks can be found here. I suggest copying the content and moving them to your favorite editor.

The openning was given by Claude Steele, a scholar of Stereotype Threat. Claude’s openning emphasized the changes in the research that come about from looking at connections between concentrations of study. He gave an example of psychology since the implementation of the MRI brain scan. Prior to image scanning, psycology was a measurement of behavior (subjective and inconsistent). MRI imaging transformed the industry by creating new opportunities to measure (quantitatively and objectively) and establishing a new language for explaining a powerful science.

Interpersonal interaction coding.

The first talk by Mark Schar analyzed productivity, creativity, and group dynamics between “divergers” and “convergers”. Divergers were described as people would questioned one another to find new answers. Convergers were oppositely seen as people who agreed with the rest of a group.

The study looked at how people collaborated, by calculating the pace at which questions were asked and new ideas were presented. The study clearly showed a distinction between two seperate groups (composed on only one type). The pace of ideas were expressed by diveregers at a rapid pace in the beginning and a bit slower in the middle. *The convergers had a slow start *to discuss ideas, then near the middle exchanged ideas and slowed down again at the end. The important point is to have both groups together when constructing teams.

The goal of the study was to understand and code interaction-dynamics between individuals working together in groups. The result was understanding that regardless of the type of people, the most important thing was to insure that no “blocks” were created during discussion. As long as no blocks existed, the group continued to look for alternaltives in approaching the situation.

Greg Kress studied how to predict long term team performance based on personalities. Study showed that all 17 factors studyed are important, but one specific point was significantly correlated with innovation and creativity. People who were extraverted in expressing their feelings created a better team dynamic and design.

Johnathan Edelman looked at how to influence media and how media influences people. ()Im not exactly sure how this was translated to the evidence below) Edelman studied two groups interaction to see how their behavior would be in a radical design process. The result showed groups that had extraverted physically expressive emotionally involved individuals were identified as the ones who would contibute more to a group. These individuals were seen to gesture more than others.

Marketing strategies

Ramesh Johari, professor in management science and engineers looked at how markets can be engineered. He looked at the variables important on influencing exchange in a online market place.

When looking at a market we ask ‘who can we trade with’, ‘wou are our competitors’ and ‘how much should we charge’. These questions are difficult to answer because of the limited amount of information provided to us in determining the answer. The rise of online market platforms change our ability to make these decisions.

Two important points of platforms are: Fine-grained matching of market participants and fine-grained collection of information about matches. In otherwords, transparency and centralization.

Market designers, like Uber, centralize the marketplace around their control. You request a ride, but don’t have a choice between who to select. Once you make a request, Uber opaquely decides who will serve you.

Oppositely, Ebay or Odesk are very decentralized and allow the use all the access to make the best decision. Still, because there are so many options, the decision is influenced by factors of search, rating, history and filtering.

Both decentralization and centralization, opacity and transparency have its benefits. Decentralization is powerful when the platform does not know what the best match for the user will be. Centralization is best in the opposite situation when the platform knows what is best for the user.

Opaque markets can also be benefitial in crowded markets. Tradiationally economics states that having more choice allows for the market to make the best decision. Instead, in a “web 3.0” world where there is too much information, it is not always obvious to know what to choose. (i.e. taobao). Having too much information and not enough can both be negative.

The big questions are how do you know how to price these market items and how much information do you need to release. A project describing this issue is looking at the pricing of mobile apps and the variables that influence the purchase of these products. When the question regarding the best marketing strategy, visibility is the most important variable in competitive atmospheres.

Academic publication meta-coding

John Willinsky and Alex Garnet talked technically about providing an effective markup structure to existing PDF journals. The focus was kill the PDA and establish a constant method for automaking the parsing and rendering of scholarly materials. The benefit of a good markup is that your document becomes your metadata. You dont need a well coded abstract, but instead the document itself contains all the content availible for understanding the contents.

Reasons: Markup is expensive when someone needs to manually markup the document. Especially for small publishers, it can be very expensive to manually tag existing documents. As a result, having a system to systematically parsing exissing content is valuable.

PDF doesnt have well-structured text mining and indexing. It also goes against current patterns of data storage, in that it does not have a way to render in different formats on mobile platforms. This prevents dyanmic content from being loaded into the documents. (Imagine having a up to date graph in your content while reading a journal)

Links: Although I dont know how this will help, some of the programs/services used in developing this software was: ParsCit, BitTex, PDFx, OJS Plugins, CrossRef, and mPach.

(to be continued…That was only 1/3rd of the presentations.)