whatsnext.tex

\vspace*{-1mm}
\section{What's Next?}\label{sec:whatsnext}

In the Big Data era, the need to process and analyze a high volume of
data is a fundamental problem. Industry
analysts point out that besides volume, there are also challenges
in variety, velocity, and veracity~\cite{jagadish2014bigdata}.
Streaming languages naturally handle volume and velocity of the data,
since they are designed to process data in real-time in a streaming way.
Thus, in the following, we focus on veracity and variety, since there
are more open research challenges in these directions despite much
recent progress in streaming languages. In addition, we elaborate on
the challenge of adoption, which is an important problem of
programming languages in general and of streaming languages in
particular.

\vspace*{-1mm}
\input{veracity}
\input{variety}
\input{adoption}

\begin{table}
  \centerline{\small\begin{tabular}{@{}lc@{ }c@{ }cc@{ }c@{ }cc@{ }c@{ }c@{}}
    \toprule
    \normalsize\emph{Languages} & \multicolumn{3}{c}{\normalsize\emph{Veracity}} & \multicolumn{3}{c}{\normalsize\emph{Variety}} & \multicolumn{3}{c}{\normalsize\emph{Adoption}}\\
    \midrule
    Relational       &       & $\mathbf{C_2}$ &       &                &       &       &       & $\mathbf{C_8}$ & $\mathbf{C_9}$\\
    Synchronous      &       &                &       & $\mathbf{C_4}$ &       &       & C$_7$ &                & $\mathbf{C_9}$\\
    Big Data         & C$_1$ & $\mathbf{C_2}$ &       & $\mathbf{C_4}$ & C$_5$ & C$_6$ & C$_7$ &                &               \\
    CEP              &       & $\mathbf{C_2}$ &       & $\mathbf{C_4}$ &       &       &       & $\mathbf{C_8}$ &               \\
    XML              &       &                &       & $\mathbf{C_4}$ &       & C$_6$ &       &                &               \\
    Stream Reasoning &       &                & C$_3$ & $\mathbf{C_4}$ & C$_5$ & C$_6$ &       & $\mathbf{C_8}$ & $\mathbf{C_9}$\\
    End-user         &       &                &       & $\mathbf{C_4}$         &       &       &                &               \\
    \bottomrule
  \end{tabular}}
  \caption{\label{tab:challenges}Which of the language families from
    Section~\ref{sec:languages} address which of the measures of
    streaming language challenges in Section~\ref{sec:whatsnext}.}
\end{table}

\subsection{Challenges Summary}

Table~\ref{tab:challenges} summarizes the challenges.  Compared to the
coverage of principles in Table~\ref{tab:principles}, the coverage of
challenges is more sparse and spread out over research prototypes.
That is why we tabulated it for language families instead of
individual languages.  There is much space for future work. The
measures highlighted in bold are most covered across all the languages
families.  The ability to handle a wide variety of data formats
appears to be a universal concern.
Ultimately, we aim at streaming languages that are
both principled and close the gap on all challenges.