diff --git a/doc/report/applications.tex b/doc/report/applications.tex index b22a954bb4e0c0c4c2e29a0e71366c44083f4d57..1af30c10545eb41d99ea221b3d0acc4b4b7ae918 100644 --- a/doc/report/applications.tex +++ b/doc/report/applications.tex @@ -196,52 +196,7 @@ implementations. dcowl}) and keeping concepts separate is not entirely unattractive in itself. - \item \textbf{$\mathcal{Q}_3$ ``Find integer sequences whose - generating function is a rational polynomial in $\sin(x)$ that has - a Maple implementation not not affected by the bug in - module~$x$.''} We see that this query is about finding specific - instances, integer sequences, with some property. This is a case - where information would be split between many sources. The name - of the sequences is part of the organizational data set, the - generating function is part of symbolic knowledge, the Maple - implementation could be part of concrete knowledge. - - \textbf{Organizational Aspect} Handling this query would probably - start by filtering for all integer sequences. It is not clear how - this should be achieved with ULO/RDF as it contains no unified - concept of a sequence. It might be possible to take advantage - of \texttt{aligned-with} or some similar concept to find all such - sequences~\cite{align}. If this succeeds, an ULO index can provide the first - step in servicing this query. Here we are in a similar situation - as with~$\mathcal{Q}_2$. It is not clear whether we should - represent the idea behind ``integer sequences'' as a native - component of ULO or as something building on top of what ULO - provides. - - As for the next steps, finding concrete algorithms and in - particular looking inside of them is not organizational data and - other indices will need to be queried. That said, it is an open - question whether ULO should contain more information (e.g.\ about - properties of the generating function) or whether such information - can be deduced from symbolic knowledge. - - \item \textbf{$\mathcal{Q}_4$ ``CAS implementation of Gröbner bases that - conform to a definition in AFP.''} Gröbner Bases are a field of - study in mathematics particular attractive for use in computer - algebra systems (CAS)~\cite{groebner}. This query is asking for - concrete implementations of Gröbner Bases that match the definition - in the Archive of Formal Proofs~(AFP)~\cite{afp}. - - We do have ULO/RDF exports for the AFP~\cite{uloisabelle}. Stated - like this, we can probably assume that $\mathcal{Q}_4$ is a query - for a very specific definition, identified by an ULO {URI}. No smart - queries necessary. What is missing is the set of implementations, - that is symbolic knowledge about actual implementations, and a way - of matching this symbolic knowledge with the definition in - {AFP}. While surely an interesting problem, it is not a task for - organizational knowledge. - - \item \textbf{$\mathcal{Q}_5$ ``All areas of math that {Nicolas G.\ + \item \textbf{$\mathcal{Q}_3$ ``All areas of math that {Nicolas G.\ de Bruijn} has worked in and his main contributions.''} This query is asking by works of a given author~$A$. It also ask for their main contributions, e.g.\ what paragraphs or code~$A$ has authored. @@ -256,7 +211,7 @@ implementations. first working version, the exports managed by \emph{ulo-storage} are enough to service this query. - As~$\mathcal{Q}_5$ is also asking for the main contributions + As~$\mathcal{Q}_3$ is also asking for the main contributions of~$A$, that is those works that~$A$ authored that are the most important. Importance is a quality measure, simply sorting the result by number of references might be a good start. Again, this @@ -292,7 +247,7 @@ implementations. GROUP BY ?work ORDER BY DESC(?refcount) \end{lstlisting} - We see that we can formulate the idea behind~$\mathcal{Q}_5$ with + We see that we can formulate the idea behind~$\mathcal{Q}_3$ with one not very complicated SPARQL query. Because here everything is handled by the database access should be quick. \end{itemize}