From: Andrea Asperti <andrea.asperti@unibo.it>
Date: Mon, 8 Nov 2004 12:27:44 +0000 (+0000)
Subject: First draft (introduction).
X-Git-Tag: v_0_6_4_1~21
X-Git-Url: http://matita.cs.unibo.it/gitweb/?a=commitdiff_plain;h=b50dc843a35a9831d3d9ac4829fee0738db2d52b;p=helm.git

First draft (introduction).
---

diff --git a/helm/papers/use_case/use_case.tex b/helm/papers/use_case/use_case.tex
new file mode 100644
index 000000000..5372f3369
--- /dev/null
+++ b/helm/papers/use_case/use_case.tex
@@ -0,0 +1,104 @@
+\documentclass[11pt,epsf,a4wide]{article}
+\usepackage{latexsym, amssymb, amsfonts}
+\usepackage{graphicx}
+
+\addtolength{\topmargin}{-1.7cm}
+\addtolength{\oddsidemargin}{-1.5cm}
+\addtolength{\evensidemargin}{-1.5cm}
+\addtolength{\textwidth}{3cm}
+\addtolength{\textheight}{3.4cm}
+
+\def\mowgli{{\sc MoWGLI}}
+
+\title{A complex use case for XML-technology:\\
+\small{The European Project IST-2001-33562 MoWGLI}}
+\date{}
+\author{}
+
+\begin{document}
+\maketitle
+
+\thispagestyle{empty}
+
+\begin{abstract}
+The following paper contains the results of an extensive validation
+of XML-technology conducted during the last three years in the framework
+of the European Project IST-2001-33562 \mowgli. 
+\end{abstract}
+
+\section{Introduction}
+The European Project IST-2001-33562 \mowgli, activated in march 2002 and
+lasting three years, aimed at the exploitation of semantic based 
+techniques for mathematical knowledge management\cite{mkm03,mkm04}, 
+with particular
+emphasis on {\em web publishing}, {\em transformation} and {\em 
+searching and retrieving} issues. \mowgli was based on an 
+extensive use of XML-technology, aiming both at providing a
+major XML test-bench and at becoming an example of best-practice in its
+use. 
+
+To grasp the dimension of the validation, we managed a 
+repository of $tot$ fully structured xml-documents\footnote{Each document is
+piece of formal mathematics (a definition or a theorem), exported
+from the library of a well-known tool for the automatic support to formal
+reasoning: the Coq proof assistant} of very different sizes,
+spanning from very few elements to $tot$ gigabytes (for a total of
+$tot$ gigabytes). We wrote 13951 lines of XSLT (39 stylesheets) 
+plus $13398$ additional lines ($21$ stylesheets) which are 
+{\em automatically generated} 
+starting from a very high-level
+xml description of mathematical notation\footnote{We also reused 
+a stylesheet transforming MathML content to
+MathML presentation (4007 lines) developed at \dots.}. Each document
+in the repository has an associated RDF file mostly modeling, in addition
+to traditional Dublin-Core metadata, dependendy relations among the
+objects in the repository. This kind of metadata are automatically 
+computed from the structured description of the object, and extensively
+used to improve both searching and browsing functionality (e.g. providing
+graphical descriptions of the dependency relations among the object in
+the repository).
+
+A lot of different tools have been tested during the development of
+the project covering most aspects of XML-technology.
+Our validation effort provided valuable feedback to many developers
+of these tools,  both in the form of bug reports (abbiamo idea di quanti
+bug reports abbiamo redatto e della loro classificazionI? quanti cioe'
+gravi?) and suggestions for improvement (especially in performance).
+
+We also actively contributed to software development. In particular,
+let us mention the implementation of Gdome2 \cite{Gdome2}, a level2 
+compliant DOM api written in C++ for the gnome programming environment, 
+that was mostly performed and tuned within the framework of the 
+\mowgli project, and GtkMathView (\cite{}), a rendering widget for 
+MathML.
+
+Summing up, \mowgli has been a very intense project, explicitly aimed 
+to {\em stress} XML-technology up to its very limit (and possibly a 
+little beyond, as in the case of stylesheets). The following paper
+is just a report of our experience. 
+
+Cenni sulla struttura del paper? Vediamo alla fine.
+
+
+\section{The repository}
+Un po di dati sulla libreria dimensione minima, media max dei files,
+idem per profondita' e larghezza. 
+Cenni sul DTD? Spiegare se/perche' non abbiamo mai sentito bisogno
+di una schema?
+
+
+
+\begin{thebibliography}{}
+\bibitem{mkm03}A.Asperti, B.Buchberger, J.Davenport (eds). 
+Proceedings of the Second International Conference on Mathematical
+Knowledge Management, MKM 2003. Bertinoro, Italy. LNCS, 2594.
+\bibitem{mkm04}A.Asperti, G.Bancerek, A.Trybulec (eds). 
+Proceeding of the Third International Conference on
+Mathematical Knowledge Management, MKM 2004. Bialowieza, Poland. LNCS 3119.
+\bibitem{Gdome2} P. Casarini, L.Padovani. The Gnome DOM Engine. 
+Markup Languages: Theory \& Practice, Vol. 3, Issue 2, pp. 173--190, 
+ISSN 1099-6621, MIT Press, April 2002. 
+\end{thebibliography}
+
+
+\end{document}
\ No newline at end of file