background image
DigiCULT
.
Info
49
ly testing it on other texts and trying it
out on older texts as well. The first older
texts we are analysing are some eigh-
teenth-century novels. In this time peri-
od, authors capitalised not only names,
but many nouns as well. We will test
whether the different use of capitalisation
will significantly change the usability of
the parser. If so, we will try to adapt the
name parser for eighteenth-century
Dutch texts and give the user the choice
of different versions of the parser.
Autonom can also be used for other lan-
guages that are presented with the
Roman alphabet, namely for those lan-
guages that make use of the same capitali-
sation rules as Dutch. A lot of languages
have the same system and can therefore
make use of Autonom. (For modern
German, where all nouns are capitalised,
Autonom actually functions as a noun
parser.) We are thinking of developing the
interface to enable the researcher to tag a
text completely in any language (in the
Roman alphabet) and of any time period,
thus to tag each word in it with a dic-
tionary entry and with a tag denoting
part of speech.
XML, XSL, CSS), and on open-source
software (e.g.TomCat, Xalan, Dom4J etc);
and we use a lightweight development
protocol (eXtreme Programming or XP),
which we feel is flexible enough for use
in academic and research environments.
NEXT STEPS
T
he first tool we have added supports
`nearness analysis'.This tool presents
a table of all the tagged name types in the
text, and all other names that are located
between one and ten words away from
each instance of this name.The number
of names in the `neighbourhood' is given
as well as the mean distance.When, after
elaborating on the tag set and the attrib-
utes, the results of this nearness analysis are
more promising and valuable, we will try
out several ways to visualise them.
T
he name parser is developed for
modern Dutch texts. We are current-
frequencies of different categories of
names.This means that first we will have
to elaborate on the tag set. Up until now,
we have worked with proper names only.
For our research we will have to enlarge
the set of attributes for this tag: we want
to separate first names from surnames,
personal names from place names, and
distinguish different types of place names
from each other, and so on, but keep all of
them in the large group of `proper names'.
Each researcher will have to be able to cre-
ate his or her own tag set and attributes.
We are currently developing tools for this.
A
s far as possible, Autonom has been
developed in respect of good prac-
tices: we intend Autonom to be open-
source software under a GPL licence (as
we think all software developed within an
academic context should be); we only rely
on platform-independent programming
(Java), on open standards (i.e. HTML,
For more information on Autonom,
please visit
http://www.niwi.knaw.nl/en/neerlan-
distiek/onderzoek/toon.
TRY OUT THE NAME PARSER
If you would like to try the name parser as it is, please visit
http://autonom.niwi.knaw.nl and register as a user (this is necessary to keep the
texts safe from copyright infringement; registering and making use of all tools is free
of charge).We would be interested to hear about your experiences and we welcome
any suggestions you might have.
T
HE
O
BJECT OF
L
EARNING
: V
IRTUAL AND
P
HYSICAL
C
ULTURAL
H
ERITAGE
I
NTERACTIONS
A
FTER THE PUBLICATION OF
D
IGI
CULT T
HEMATIC
I
SSUE
4: L
EARNING
O
BJECTS FROM
C
ULTURAL
AND
S
CIENTIFIC
H
ERITAGE
R
ESOURCES
(
HTTP
://
WWW
.
DIGICULT
.
INFO
/
PAGES
/
PUBLICA
-
TIONS
.
PHP
), D
OMINIC
P
ROSSER
,
FROM THE
U
NIVERSITY OF
E
XETER
T
ELEMATICS
C
ENTRE
,
AND
S
USAN
E
DDISFORD
, T
IVERTON
M
USEUM AND
THE
T
ELEMATICS
C
ENTRE
,
SPOKE TO
D
IGI
CULT.I
NFO
ABOUT THE THEORY AND PRACTICE
OF LEARNING THROUGH VIRTUAL EXHIBITIONS
.
A
t the Telematics Centre, University of
Exeter, we have just started our 12
th
digital heritage education project. A full
description of all our projects can be found
on our Website at
http://telematics.ex.ac.uk. Although we
cannot describe them all here, exploring
the following projects will demonstrate
our work:
RealCornwall (http://www.realcorn
wall.net/): in partnership with Cornwall
County Council, the Telematics Centre
is creating an open-access Website to
stimulate and support involvement with
Cornish Culture. It is designed to sup-
port the adult literacy core curriculum.
(Supported by the New Opportunities
Fund).
Virtual Victorians (http://www.victori
ans.org.uk/): working in partner-
BACK TO PAGE 1