\section{Introduction}

To tackle the vast array of mathematical publications, various ways of
\emph{computerizing} mathematical knowledge have been researched and
developed. As it is already difficult for human mathematicians to keep
even a subset of all mathematical knowledge in their mind, a hope is
that computerization will yield great improvement to mathematical (and
really any) research by making the results of all collected research
readily available~\cite{onebrain}.

One research topic in this field is the idea of a \emph{tetrapodal
search} that combines four distinct areas of mathematical knowledge
and data. These four kinds being (1)~the actual formulae as \emph{symbolic
knowledge}, (2)~examples and concrete objects as \emph{concrete knowledge},
(3)~names and comments as \emph{narrative knowledge} and finally
(4)~identifiers, references and their relationships, referred to as
\emph{organizational knowledge}~\cite{tetra}.

Tetrapodal search wants to provide a unified search engine that
indexes each of those four different subsets of mathematical
knowledge.  Because all four kinds of knowledge are inherently
different in their structure, tetrapodal search proposes that each
kind of mathematical knowledge should be made available in a storage
and index backend that fits exactly with the kind of data it is
storing. With all four areas available for querying, tetrapodal search
wants to then combine the four indexes into a single query interface.