[state_of_wikimedia_research_2015] / 20150717-wikimania_research.tex

\documentclass[xcolor=dvipsnames]{beamer}

% set up the file to create notes in the output PDFs
\usepackage{pgfpages}
\input{notes.config}

\renewcommand{\rmdefault}{ugm}
\usepackage[garamond]{mathdesign}

\renewcommand{\sfdefault}{phv}
\usepackage{relsize}

\usepackage{ucs}
\usepackage[utf8x]{inputenc}
\usepackage[T1]{fontenc}
\usepackage{textcomp}

% add tikz and a bunch of tikz foo
\usepackage{tikz}
\usetikzlibrary{shapes,shapes.misc,backgrounds,fit,positioning}
\tikzstyle{every picture}+=[overlay,remember picture]

% add functions to circle parts of slides (e.g., in tables)
\newcommand\marktopleft[1]{\tikz \node (marker-#1-a) at (0,1.5ex) {};}
\newcommand\markbottomright[1]{%
  \tikz{\node (marker-#1-b) at (0,0) {};}
  \tikz[dashed,inner sep=3pt]{\node[violet!75,ultra thick,draw,rounded rectangle,fit=(marker-#1-a.center) (marker-#1-b.center)] {};}}

% DEPRECATED function to build a huge centered dropshadow
\newcommand\dropshadow[3]{%
  \node[black!30!white] at (#1+0.1,#2-0.1) {
    \scalebox{2}{\Huge \textbf{#3}}
  };
  \node at (#1,#2) {
    \scalebox{2}{\Huge \e{#3}}
  };
}

% create an empty quotetxt so we can reuse it
\newcommand{\quotetxt}{}

% more flexible non-tikz alternative with no dropshadow 
\newlength{\centertxtlen}
\makeatletter
\newcommand\centertext[2]{%
  \setlength{\centertxtlen}{#1}%
  \setlength{\centertxtlen}{0.48\centertxtlen}%
  {\centering
    \fontsize{#1}{2\centertxtlen}\selectfont
    \e{#2}

  }
}


% add function to stop numbering appendix slides
\newcommand{\backupbegin}{
   \newcounter{framenumberappendix}
   \setcounter{framenumberappendix}{\value{framenumber}}
}
\newcommand{\backupend}{
   \addtocounter{framenumberappendix}{-\value{framenumber}}
   \addtocounter{framenumber}{\value{framenumberappendix}} 
}

% packages i use in essentially every document
\usepackage{graphicx}
\usepackage{url}
% \usepackage{dcolumn}
% \usepackage{booktabs}

% replace footnotes with symbols instead of numbers
\renewcommand*{\thefootnote}{\fnsymbol{footnote}}
\usepackage{perpage}
\MakePerPage{footnote}

%% BEAMER THEME STUFF
\usetheme[pageofpages=/,% String used between the current page and the
                         % total page count.
          bullet=default,% Use circles instead of squares for bullets.
          titleline=false,% Show a line below the frame title.
          alternativetitlepage=true,% Use the fancy title page.
          titlepagelogo=figures/logo.pdf,% Logo for the first page.
          %watermark=watermark-polito,% Watermark used in every page.
          watermarkheight=100px,% Height of the watermark.
          watermarkheightmult=4,% The watermark image is 4 times bigger
                                % than watermarkheight.
          ]{Torino}

\usecolortheme{mako}
\useinnertheme{rectangles}
%\setbeamertemplate{blocks}[rounded][]
\setbeamercolor{block title}{bg=makopurple3, fg=White}

\setbeamertemplate{items}[default] 
\setbeamertemplate{blocks}[shadow=true] 

\usepackage{tcolorbox}
% These options will be applied to all `tcolorboxes`
\tcbset{%
    noparskip,
    colback=makopurple5, %background color of the box
    colframe=makopurple1, %color of frame and title background
    coltext=black, %color of body text
    coltitle=white, %color of title text 
    arc=0em,
    left=0.1em,
    right=0.1em,
    fonttitle=\bfseries,
    alerted/.style={coltitle=red, 
                     colframe=gray!40},
    example/.style={coltitle=black, 
                     colframe=green!20,             
                     colback=green!5},
    }


%\useoutertheme{infolines}
%\usepackage[breaklinks]{hyperref}

\hypersetup{colorlinks=true, linkcolor=Black, citecolor=Black, filecolor=makopurple1,
    urlcolor=Plum, unicode=true}

% create a boldface version of the header
\setbeamerfont{frametitle}{series=\bfseries}
\setbeamerfont{title}{series=\bfseries}

% tweak the beamer font to make it a bit lists a bit smaller
\setbeamerfont*{itemize/enumerate body}{size=\small}
\setbeamerfont*{itemize/enumerate subbody}{size=\footnotesize}
\setbeamerfont*{itemize/enumerate subsubbody}{size=\footnotesize}

% indent the margins of the itemize lists a little bit
\setlength{\leftmargin}{0pt}
\setlength{\leftmargini}{0.7cm}
\setlength{\leftmarginii}{0.7cm}

% create a new \e{} command to make things purple and bold
\newcommand{\e}[1]{\textcolor{makopurple1}{\textbf{#1}}}

% remove the nagivation symbols
\setbeamertemplate{navigation symbols}{}

\title{Presentation Title}
% \subtitle{Presentation Subtitle}
\author[Benj. Mako Hill]{\textbf{Benjamin Mako Hill}\\ makohill@uw.edu}

\institute[UW/Harvard]{\textbf{University of Washington}\\
  Department of Communication\\
  \emph{Assistant Professor}\\
  \hspace{1pt}\\
  \textbf{Harvard University}\\
  Berkman Center for Internet and Society\\
  \emph{Resident Fellow}}

\date{December 2, 1980}

\newcommand{\credit}[1]{%
  \tikz[overlay]{\node at (current page.south east)
    [anchor=south east,yshift=1.1em,xshift=0.35em]
    {\smaller \smaller {[}#1{]}};}}

\begin{document}

% remove some of the space in the itemize to make it quite compact
\let\olditemize\itemize
\renewcommand\itemize{\olditemize\itemsep-1pt}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{Introduction}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

%% SLIDE: Title Slide
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{frame}[plain]

  \begin{tikzpicture}
    \node at (current page.center) [xshift=-3.5cm, yshift=0.5cm, opacity=0.4]
      {\includegraphics[height=\paperheight]{figures/wikimedia_projects.png}};


    \node at (current page.south east)
    [anchor=south east,text width=1.8\paperwidth,align=right,color=black]
    {
      {\spaceskip 0.3em%
        \fontsize{2.5em}{2.5em}
        \selectfont {\bf \color{makopurple4} The State of Wikimedia\\
          Research: 2014-2015} \par}

      \vspace{1em}

      {\spaceskip 0.3em%
        \fontsize{2.0em}{2.1em}
        \selectfont {\bf \color{black} Benjamin Mako Hill\\
          Tilman Bayer\\
          Aaron Shaw\\
          Wikimania 2015, Mexico City\\
          July 17, 2015} \par}

    };
  \end{tikzpicture}

% Removed to accommodate Tilman's setup  
%  \input{vc}
%
%  \tikz[overlay,shift=(current page.south west)]{\node [xshift=5.6em,yshift=0.5em]{\colorbox{makopurple1}{\color{white} \tt \smaller \smaller \smaller revision:\ \VCRevision\  (\VCDateTEX)}};}

  \note{I've been doing this for many years. I started in 2008 and
    have done this almost every single year since.

    This began as an excuse for me to make sure I was up to date on
    Wikimedia Research.}

\end{frame}

%% SLIDE: Anecdote from Wikimania 2008
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\renewcommand{\quotetxt}{``This talk will try to [provide] a quick
  tour – a literature review in the scholarly parlance – of the last
  year's academic landscape around Wikimedia and its projects geared
  at non-academic editors and readers. It will try to categorize,
  distill, and describe, from a birds eye view, the academic landscape
  as it is shaping up around
  our project.''\\
  \hfill – \e{From my Wikimania 2008 Submission}}

\begin{frame}

  {\smaller \quotetxt}

  \pause
  \includegraphics[width=\textwidth]{figures/google_scholar_result.png}

  \pause
  \tikz{\draw (current page.center) [xshift=-2.1cm, yshift=0.9cm, color=red]
    ellipse (1.5cm and 0.5cm);}

  \note<1>{Back in Wikimania 2008, I set out to run a session at
    Wikimania that would provide a comprehensive literature review of
    articles in Wikipedia published in the last year.

    \begin{quote}
      \quotetxt
    \end{quote}

    Then, about two weeks before Wikimania, I did the scholar search
    so I could build the literature.}

  \note<2->{I tried to import the whole list into Zotero and managed
    to get banned for abusing the Google Scholar because they thought
    that no human being could realistically consume the amount of
    material published on Wikipedia that year.

    So anyway, I had a 45 minute talk so it worked out to 3.45 seconds
    to per paper...

    And believe it or not, this year is even bigger.
 
    And my talk is even shorter.}

\end{frame}

%% SLIDE: Citations Per Year
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{frame}

  \includegraphics[width=\textwidth]{figures/citations_by_year.pdf}

  \centering

  {\smaller \emph{Number of citation, per year, with the term
      “wikipedia” in the title.\\
      (Source: Google scholar results. Accessed: 2013-08-06)}}

  \note{Academics have written \e{a lot} of papers about
    Wikipedia. There are more than 500 papers published about
    Wikipedia each year and although we've reached and moved past a
    peak it seems, it's not slowing by much.}

\end{frame}

\begin{frame}

  \begin{itemize}
    \larger \larger
  \item \e{2968} Wikipedia-related publications in the Scopus database
    as of November 2013

  \item \e{160} recent publications reviewed or mentioned in the 12 issues
    of the Wikimedia Research Newsletter August 2013-July 2014.
  \end{itemize}

\end{frame}

%% SLIDE: My Scope Conditions
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{frame}

  \includegraphics[width=\textwidth]{figures/multiple_issues.png}

  \larger \larger
  In selecting papers for this session, the goal is always to choose
  examples of work that:

  \begin{itemize}
  \larger \larger
  \item Represent \e{important themes} from Wikipedia in the last year.
  \item Research that is likely to be of \e{interest} to Wikimedians.
  \item Research by people who are \e{not at Wikimania}.
  \item \ldots with a bias towards \e{peer-reviewed} publications 
  \end{itemize}

  \note{This is my disclaimer slide...

  Within these goals, the selections are \e{incomplete}, and \e{wrong}.}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{Paper Summaries}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{frame}
  \centertext{6em}{Wikipedia as a Source of Data}

  \note{Mako}
\end{frame}

\begin{frame}

  \frametitle{Wikipedia as a source of data}

  \larger \larger Ronen, S., Gonçalves, B., Hu, K. Z., Vespignani, A.,
  Pinker, S., \& Hidalgo, C. A. (2014). \e{Links that speak: The
    global language network and its association with global
    fame}. Proceedings of the National Academy of Sciences, 111(52),
  E5616—E5622. \href{http://doi.org/10.1073/pnas.1410931111}{doi:10.1073/pnas.1410931111}

\end{frame}

\begin{frame}
  \frametitle{Wikipedia as a source of data: Ronen et al.}

  \includegraphics[width=\textwidth]{figures/ronen_fig1.png}
\end{frame}

\subsection{Community and Organization}

\begin{frame}
  \centertext{6em}{Community and Organization}

  \note{Mako}
\end{frame}

\begin{frame}

  \frametitle{Community and organization}

  \larger \larger Warncke-Wang, M., Ranjan, V., Terveen, L., \& Hecht,
  B. (2015). \e{Misalignment Between Supply and Demand of Quality Content
  in Peer Production Communities}. In Ninth International AAAI
  Conference on Web and Social Media (ICWSM).

  % Retrieved from \href{http://www.aaai.org/ocs/index.php/ICWSM/ICWSM15/paper/view/10591}{http://www.aaai.org/ocs/index.php/ICWSM/ICWSM15/paper/view/10591}

\end{frame}

\begin{frame}
  \frametitle{Community and organization: Warncke-Wang et al.}

  \includegraphics[width=\textwidth]{figures/warncke-english_confusion.pdf}
\end{frame}

\begin{frame}
  \frametitle{Community and organization: Warncke-Wang et al.}

  \includegraphics[width=\textwidth]{figures/warncke-english_overunder.pdf}
\end{frame}

\subsection{Content Quality}

\begin{frame}
   \centertext{6em}{Content quality}

   \note{Tilman

     A decade after the landmark "Nature" study, there still aren't too
     many systematic evaluations of the accuracy of Wikipedia's content. 
     Health articles continue to receive scrutiny, though. With good 
     reason: Wikipedia is "the most frequently consulted online health 
     care resource globally" [NEJM article].}
\end{frame}

\begin{frame}
  
\frametitle{Quality of drug articles}

   \larger \larger
   Hwang et al., ``\e{Drug Safety in the Digital Age}.'' 
   N Engl J Med 2014; 370:2460-2462 June 26, 2014 
   \href{http://dx.doi.org/10.1056/NEJMp1401767}{doi: 10.1056/NEJMp1401767}.
   \bigskip
   
   Kräenbring et al., \e{Accuracy and completeness of drug 
   information in Wikipedia: a comparison with standard textbooks of 
   pharmacology}. PLoS One 9 (9): e106930. 
   \href{http://dx.doi.org/10.1371/journal.pone.0106930}
   {doi:10.1371/journal.pone.0106930}
   

   \note{Tilman
   
     We selected two papers that evaluated drug articles, with 
     different approaches. The first one is a short article in the 
     extremely prestigious NEJM.}
\end{frame}

\begin{frame}
  
\frametitle{Quality of drug articles: NEJM}

   \includegraphics[width=0.49\textwidth]{figures/Pradaxa_tweet_FDAMedWach.png}
   % from https://twitter.com/FDAMedWatch/status/281547908095041536
   % = first one in the list at http://www.nejm.org/doi/suppl/10.1056/NEJMp1401767/suppl_file/nejmp1401767_appendix.pdf
   \includegraphics[width=0.49\textwidth]{figures/Dabitragan_Contraindications_WP_FDA_warning}
  
   \tikz{\node [yshift=1.5cm,xshift=-0.4cm] at (current page.center) {\includegraphics[width=1.5cm]{figures/long-arrow-right.png}};}
   \begin{itemize}
     \larger \larger
     \item The US Food and Drug Administration (\e{FDA}) frequently 
     issues safety warnings about prescription drugs. How long does it 
     take until these are reflected on English Wikipedia?
     \item 41\% updated within two weeks (58\% for high-prevalent 
     diseases), but 36\% still unchanged after more than a year.
   \end{itemize}
   
   \note{Tilman
   
   Articles about drugs used to treat high-prevalent diseases (affecting 
   > 1 m Americans / year) were updated faster.\\ 
   But the result still caused concern.\\
   Authors find "there may be a benefit to enabling the FDA to update or 
   automatically feed new safety communications to Wikipedia pages, as 
   it does with WebMD". The paper raised awareness among WikiProject 
   Medicine editors, but there's no systematic updating mechanism yet.}


\end{frame}

\begin{frame}

\frametitle{Quality of drug articles: PLoS One}

  \begin{itemize}
   \larger \larger \larger
   \item Selected 100 drugs from German undergrad curriculum in pharmacology
   \item Extracted information from two standard textbooks
   \item "Accuracy of drug information in [German] Wikipedia was 99.7\%±0.2\% when compared to the textbook data." Similar results for English Wikipedia
   \end{itemize}

\end{frame}


\begin{frame}

\frametitle{Quality of drug articles: PLoS One}

  \begin{itemize}
   \larger \larger \larger
   \item Completeness (as compared to the textbooks):
   \begin{itemize} \larger \larger
     \item 83.8\% (of 224 statements) for German WP
     \item 87.2\% for English WP
   \end{itemize}   
   \item Completeness of contraindications information was 100\% in the En WP sample.
   \item English WP cited academic publications more often than German WP.
   \item Quality "significantly improved" in drug articles assessed 
        in a 2010 study.
   \end{itemize}

    \note{Tilman
    
       The majority of the missing information (62.5\%) on German WP 
       was judged non-relevant for undergrad students.
    
       The result on completeness of contraindications information is 
       somewhat in contrast with the NEJM study. Then again, the 
       textbooks were probably not perfectly up-to-date either.}
\end{frame}



% \begin{frame}
%   \frametitle{Wikipedia Viewership and Flu Prediction: Results}

%   \centering
%   \includegraphics[width=\textwidth]{figures/flu.png}

% \note{\begin{itemize}
%   \larger \larger

%   \item Wikipedia better than Google at predicting peak flu weeks.
%   \item Wikipedia better at predicting relative influenza rates.
%   \end{itemize}}

% \end{frame}


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{Conclusion}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

%% SLIDE: Other Resources
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{frame}{More Resources}

  \begin{itemize}
    \larger \larger
  \item \e{Wikimedia Research Newsletter} [[:meta:Research:Newsletter]] / @WikiResearch
  \item \e{WikiSym/OpenSym} (This August in San Francisco!)
  \item \e{WikiPapers Repository} [http://wikipapers.referata.com]
  \item \e{Much More}
  \end{itemize}

  {\centering 
    \includegraphics[width=0.25\textwidth]{figures/Wikimedia_Research_Newsletter_Logo.png}

  }

  \note{Those are my six exemplary studies from the past year. 

    There has been just tons and tons of work in this area. Trying to
    talk about this in 20 minutes strikes me as increasingly crazy
    every year I try to do it.

    The most important source, now going for a couple years, is the
    Wikimedia Research Newsletter which is published monthly in the (English)
    Signpost and syndicated on the Wikimedia Research.

    But there are other resources as well. And I encourage you to get
    involved.}

\end{frame}

\subsection{Meta-Analyses}

\begin{frame}

  \frametitle{Meta-Analyses}

  \begin{itemize}

    \larger \larger

  \item Okoli et al.,
    \href{https://spectrum.library.concordia.ca/978618/}{``The sum of
      all human knowledge'': a systematic review of scholarly research
      on the content of Wikipedia}.

  \item Bar-Ilan and Aharony,
    \href{http://dl.acm.org/citation.cfm?doid=2615569.2615643}{Twelve
      years of Wikipedia research}.
    
  \item Taraborelli. \href{https://meta.wikimedia.org/wiki/Research:Newsletter/2013/August\#Keynote\_on\_applicable\_Wikipedia\_research}{Keynote
      on Wikipedia Research}. OpenSym 2013. Hong Kong.

  \item Benkler, Shaw, and Hill,
    \href{http://mako.cc/academic/benkler\_shaw\_hill-peer\_production\_ci.pdf}{Peer
      Production: A Modality of Collective Intelligence}.

  \end{itemize}

\end{frame}

\end{document}