Saitama University
Independent System Engineer
We describe here a basic design of a multi-purpose database
system for Humanities research that we are developing. The
system named “LiterarySpace” provides a new integrated
electronic workbench for carrying out various kinds of
processing acts on textual study: documentation, transcription,
interpretation, representation, publication and so on
(Shillingsburg 2006). In other words, it has a polar range of
possibilities of functions from a private tool to a publishing
medium: image data of any materials such as inscribed stones,
wooden tablets, papyrus rolls, paper codex etc. can be stored
for private and/or for public use, and recorded text data
such as the results of electronically scholarly editing and/or
outcomes of hermeneutic inquiry with/without relation to the
graphics, and then represent all or part of the data archives to
yourself in a stand-alone computer and/or to anyone via the
internet.
The essential architecture of this relational database is simple:
It is constructed from three database fi les: fi le-1, graphic data
fi le = GDF; fi le-2, text data fi le = TDF; fi le-3, control data fi le =
CDF. GDF has two data fi elds in each of its records: the one for
storing image data and the other for its unique serial numbers
the system automatically issues. We named this identifi cation
number IDGraphic. A record in TDF consists of three data
fi elds which keep three kinds of data: 1) text data, 2) its unique
identifi cation integers assigned by the system which we named
IDText, 3) IDGraphic as a relational key to GDF. In terms of
the relation between GDF and TDF, one distinguishing feature
is that one GDF record allows an overlapped link to plural
TDF records; namely the correspondence of records in GDF
to ones in TDF is one-to-many and not vice versa. A CDF
record contains the following two fi elds: 1) IDSet fi eld in
which IDGraphic or IDText is registered as a relational key,
and 2)Index fi eld which keeps numbers that have been input
in order to make the sequence of the records called out by
the IDSet. As for the identifi cation of the type of IDSet, with
just the two alternatives, IDGraphic or IDText, the name of
the CDF serves as a mark. If the fi lename given by the user
contains the word “GDF” or “gdf” , the fi le is interpreted as a
set of records from GDF. Otherwise, it is being from TDF.
The powerfulness of “LiterarySpace” is caused from this
simple structure which corresponds to the theoretical model
we made up based on our clear understanding about space and time in writing (Myojo & Uchiki 2006). Concerning space,
written existence demands two different spaces: physical
space which every written text occupies, and conceptual
space which the text generates in the minds of writers and
readers at the same time (Bolter 1991). On the other hand,
time is time; time is never physical but ever visionary and
such time in a written text can be controlled by the writer
and even the reader. The three database fi les of our system
coincide with these epistemological dimensions: GDF with
the physical space, TDF with the conceptual space, and CDF
with the variable time, or a little more precisely, fi ctional
spaces in which each conducted time goes (Stanzel 1989).
Within the one space of the one “LiterarySpace” there can
exist only one GDF, one TDF, and plural CDF(s). The space of
the single GDF is constituted of image data as non-charactercoded
documentation of extant materials. And the time of
its space is represented by the sequential order indicated
by integers assigned mechanically for each additional record.
Needless to say, the order corresponds to the succession in
which the records are input. The one-time-only character of
the inputting acts in reality is the reason why the integer, i.e.
IDGraphic is unique. So you might call the time generated by
the unique numbers the “real” one. The single TDF is made
up from text data as character-coded documentation of any
writing, that is to say, the products of conceptual work; and the
space has also the single “real” time represented in the order
designated by the IDText integers, which naturally corresponds
to the succession in which the data are put. The philosophical
substance of CDF(s) is, however, different; its space is a metaspace
which consists of data records chosen from GDF or
TDF. As the space is, so to speak, fi ctional, it is possible to be
set in imaginary time. The meta-time can be produced by the
sequence of the Index-numbers issued not by machine but by
the user. Because of the fi ctionality, one can make more than
one sequence if necessary. So, unlike other types of fi eld, the
Index-fi eld can be increased. To be sure, there also exists “real”
time in accordance with the succession of the records input
into the CDF and if no sequence is created, the records would
be represented in the input order.
The fi rst part of our presentation will illustrate this system
design more in detail, explaining signifi cant features including
e.g. the reason of the above mentioned one-to-many
correspondence of GDF to TDF in connection with the
theoretical modelling (McCarty 2005).
In the second part we will demonstrate a prototype of the
system which is built using the database application “FileMaker
Pro” (After the specifi cation of the software has been
completed, its program will be rewritten in Common Lisp).
The current main database in the prototype is Franz Kafka’s
database constituted of information particularly from his
“Oktavhefte”. One of us, Myojo is originally a Kafka researcher
and the primal need of this system arose in the context of
her literary research project (Myojo 2002). Myojo’s study
following her own methodology as a combination between
“Editionswissenschaft” and “critique géntique” has always
demanded a cross-referenced complicated work handling
three editions simultaneously: the practical edition by Max
Brod, the critical edition “Kritische Kafka-Ausgabe”, and the
facsimile edition “Franz Kafka-Ausgabe”. So we should admit
that the fi rst attempt of our modelling was not theoretical
but practical (Myojo 2004). It is, however, because our project
started to grasp the actual needs of the one scholar in the
Humanities that this system has grown up as a real useful
entity. Using the Kafka database as an example, we would like
to demonstrate not only how the system operates but also
how effi ciently the performance assists the investigation for
work in the humanities.
The third part will be dedicated to showing the other feasibilities
of this system. One of the many defi ning characteristics of
this system is that it allows us to contain several databases
at the same time. For instance in addition to Kafka’s database
one can also construct Shakespeare’s as well as the work
of some Japanese authors’ etc. in the same system. Also, a
more signifi cant point is that one can input as much data as
necessary without worrying about grouping or sequences.
Simply put, one can make a record of Kafka’s information just
after inputting that of Shakespeare’s. This unrestrictedness is
enabled by the aforementioned epistemological data structure
because the grouping and the sequence can be controlled
afterwards by means of creating a meta-area, i.e. a CDF.
Presently in our prototype two more databases of Japanese
authors are under construction: Kenji Miyazawa (宮沢賢治)’s
and Soseki Natsume (夏目漱石)’s. The strong merits of this
capability will be presented as an exhibition of research results
in the fi eld of comparative literature (Myojo 2003).
One more important point we have to mention: as we noted
at the beginning of this text, the system serves as a private tool
and/or a publishing medium. The signifciance of this feature
would be recognized well if you imagined the case of dealing
with works protected by copyright. Of course the dual aspects
of privateness and publicness correspond to the original
distinguishing character of the computer itself. From not only
this but the all above points of view it might be suggested that
this system could become a powerful electronic substitute
for a physical notebook. In the last part of the presentation
we would like to discuss the multifaceted possibilities, i.e. the
universality of the system.
Acknowledgments
This work was funded in part by The Japan Society for the
Promotion of Science (JSPS) under Grant-in-Aid for Scientifi c
Research (C) [18529001]. We would also like to thank
Christian Wittern (Kyoto University), Tetsuya Uchiki (Saitama
University) and other members of The Japanese Association
for Scholarly Editing in the Digital Age (JASEDA) for their help
in establishing regular and fruitful discussions.Bibliography
Bolter, David J., Writing Space, Hillsdale (Lawrence Erlbaum
Associates) 1991.
McCarty, Willard, Humanities Computing, Houndmills (Palgrave
McMillan), 2005.
Myojo, Kiyoko, Kafka Revisited, Tokyo (Keio University Press),
2002 [in Japanese].
Myojo, Kiyoko, “Kafka und sein japanischer Zeitgenosse
Kenji”, Saitama University Review 39.2 (2003): 215-225.
Myojo, Kiyoko, “What is the Future of Computer-Assisted
Scholarly Editing?” IPSJ SIG Technical Reports CH-62 (2004):
37-44 [in Japanese].
Myojo, Kiyoko & Tetsuya Uchiki: “A Theoretical Study on
Digital-Archiving of Literary Resources”, IPSJ Symposium
Series 17 (2006): 153-160 [in Japanese].
Shillingsburg, Peter L., From Gutenberg to Google, Cambridge
(Cambridge University Press) 2006.
Stanzel, Franz K., Theorie des Erzählens, Göttingen
(Vandenhoec & Ruprecht), 1989 [1979].
If this content appears in violation of your intellectual property rights, or you see errors or omissions, please reach out to Scott B. Weingart to discuss removing or amending the materials.
Complete
Hosted at University of Oulu
Oulu, Finland
June 25, 2008 - June 29, 2008
135 works by 231 authors indexed
Conference website: http://www.ekl.oulu.fi/dh2008/
Series: ADHO (3)
Organizers: ADHO