background image
objective is to build a virtual community
of sponsors, developers, contributors, inte-
grators and users, each playing an impor-
tant role in the evolution of the project.
One striking aspect of this collaboration is
the amount of contributions needed from
non-technical people.
urrently, PLEADE uses its own
Website and the Sourceforge plat-
form for open-source projects as the main
tool for building this virtual community.
At, one will find
general information about the project, its
actors, various events relating to the
project, and up-to-date documentation.
Most of the information is currently in
French, but translation into English and
other languages should occur early in
2004 (contributions welcome).
pleade/, one will find links for down-
loading the latest distributions of the
software, and a mailing list (and its
archives) for the PLEADE community.
The Sourceforge services are available
in various languages including English,
and the mailing list accepts posts in any
21h/pl-demo/, there is a standard
installation of PLEADE, including
the example documents distributed
with the platform. No specific con-
figuration or adaptation has been
made for this installation.
f anyone requires to know more about
PLEADE or wishes to contribute, the
first step is to subscribe to the mailing list
and then send a message. All contributions
and questions are welcome.
between one EAD document and another,
or between one EAD document and the
digital surrogates of the archival material,
and so on.
ollowing the tradition of open-source
software and modern software devel-
opment methodologies, PLEADE is a thin
layer built on top of some other technolo-
gies.The most important ones are, from
the bottom up:
a Web server, such as Apache, for
handling HTTP requests from a Web
browser, and sending back the informa-
tion as HTML pages;
a Java servlet engine, such as Apache
Tomcat, and beneath it a Java virtual
and SDX (, a search
engine and publishing platform for
XML documents, based on the Cocoon
2 infrastructure.
LEADE can thus be seen as a layer on
top of SDX, a layer that lets anyone
publish EAD documents on the Web.This
layer takes control of tasks such as index-
ing, organising and displaying EAD docu-
ments.This layer is quite generic and can
be easily configured to suit different needs
or layouts. And since SDX supports the use
of thesauri for searching, along OAI-PMH
harvesting or repositories, these features
could be easily added to PLEADE.
s PLEADE is based on Java technolo-
gy, it can be installed on any platform
supporting a Java virtual machine, includ-
ing Linux and most of UNIX,Windows
and Mac OS X. It is thus easy to install it
on a personal computer for testing or pre-
sentations, and also on high-end servers for
public access.
eveloping successful open-source
software involves a lot more than
technologies. For PLEADE, the main
ed with various options such as choosing
one or more archival levels (e.g. fonds, file,
item), one or more hierarchical levels (e.g.
any third-level component) or even an
EAD element (c0, c1, etc.).This means that
the user will see meaningful search results,
at the file level for instance, but these
results will always be contextualised (the
titles of its parents are presented also).
isplaying EAD documents
PLEADE usually involves a three-
pane window.The top pane presents the
bibliographic information about the find-
ing aid; the right-hand pane contains the
contents, split into searchable units in order
to avoid large pages on slow network con-
nections; and the left-hand pane contains
interactive tables of contents for browsing
the finding aid by the component titles
and, if available, by indexation terms found
in the documents. Browsing and reading
the EAD documents is possible with a
standard Web browser, like any other fea-
ture of PLEADE, as the documents are
dynamically converted from XML to
he above description of the main fea-
tures of PLEADE should not conceal
the fact that everything is highly config-
urable, from the interface `look and feel' to
the way the documents are indexed. For
instance, when an administrator adds a
document (or a batch of documents shar-
ing similar characteristics), he or she may
choose the subsets it belongs to, how the
document will be fragmented for searching
and displaying, how to build the title of
the searchable units from the various EAD
elements, whether a table of contents
should be used and, if so, what its label will
be, whether the indexing terms should be
used for browsing and/or global searching,
what presentation formats to use from
among a list of available formats (two are
currently available), the set of phrases for
labelling the EAD element contents for
display, how to process hypertext links