374
David Nathan
objects can function like icons, and also that simply using images for navi-
gation does not, by itself, ensure a usable interface.
Figure 4. Navigating in the Paakantyi talking dictionary (Hercus and Nathan 2002)
4.3. Idiomatic interfaces
The Spoken Karaim CD-ROM (Csató and Nathan 2003 b)
12
was developed
as a multimedia documentation of the language, culture, and environment
of the Karaim community of Trakai, Lithuania. It is centered on several
narrative monologues spoken by the last full speakers of the language;
these are all transcribed, morphologically annotated, and linked to a rich
lexicon, grammar, and concordance; in addition, the linguistic material is
accompanied by songs, videos, photographs, and eight thematic articles on
Karaim history, religion, etc.
The CD also contains a system we call “Active morphology” that gener-
ates inflected nouns using an inbuilt full computational model of Karaim
morphophonology (Nathan 2000 b). Inflection is represented by the idiom
of dragging objects onto headwords. The system is presented to the user as
a set of small moveable blocks that, when dragged onto a dictionary head-
word, initiate the generation and display of the appropriate inflected form
(see Figures 5 and 6). By keeping terminology basic, the morphophono-
Chapter 15 – Thick interfaces: Mobilizing language documentation
375
logical rules unseen, and its operation as simple as possible, the system is
backgrounded for most users, and may, paradoxically, even remain undis-
covered. However, given a rational context for use, it is easy to learn how
to use.
The effectiveness of this system was confirmed during the recent
Karaim Summer School in 2004. We developed interactive, computer-
generated multimedia crossword games as language teaching aids (there
were three types: normal crosswords, talking crosswords, and picture
crosswords). The Karaim students were encouraged to use Spoken Karaim
to look up words they didn’t know. Their responses illustrated the three
factors of idiom, rational design, and open interpretation mentioned in the
introduction to this section. Motivated by competitive crossword games,
and faced with clues whose answers required inflected forms, students
found and used for the first time the idiomatic Active morphology controls,
and they explored the CD in new ways, including its interactive concor-
dance (which finds inflected roots within narratives). In performing these
activities, the students explored the CD’s language content and selected and
interpreted the results in order to complete the crosswords.
4.4. “Dangerous” interfaces
A community’s knowledge and modes of presentation can go even further
in defining a product’s interface and usage. Barbara Glowczewski’s Dream
Trackers: Yapa Art and Knowledge of the Australian Desert (2001a) is a
comprehensive CD presenting the land, language, and culture of the
Warlpiri (also known as Yapa) people of the Northern Territory, Australia.
Its interface is structured by Warlpiri forms of knowledge representation,
which highlight networks of associations. It opens with an interactive map
of “dreaming paths” in an extraordinarily complex, criss-crossing pattern.
Nodes on these paths are interlinked with various stories, text, and artwork
throughout the CD. Glowczewski (2001b: 142) wanted the links “to follow
rules and to have meanings that respected the connections that the Warlpiri
themselves establish according to their own cognitive logic.”
The visual transposition of this Aboriginal cognitive mapping into an inter-
active map gives the user an immediate experience of this inter-connectivity
… Multimedia is an ideal tool for rendering this Indigenous mapping. This
invitation to wander in the territory of the Dreaming story-telling, painting,
singing and dancing made the old persons extremely happy when they saw
376
David Nathan
Figure 5. Operation of Spoken Karaim’s “Active morphology” (extract from the
CD’s Help file)
Figure 6. Result of “Active morphology” operation
Chapter 15 – Thick interfaces: Mobilizing language documentation
377
tangible proof of their teaching about the inter-relatedness of the Dreaming.
The elders and all the women I worked with were excited by the new me-
dium because it did not threaten their encyclopaedic knowledge or their
power in the society. On the contrary, their legitimacy was affirmed by the
fact that they are recognized by name as story-tellers and painters.
(Glowczewski 2001b: 146)
However, some Warlpiri people were also deeply disturbed by the CD,
because they saw that, for the first time, something fundamental about their
knowledge was revealed to outsiders. Ultimately, this led to the community
restricting the distribution of the CD (Glowczewski 2001 b: 150). This re-
striction, paradoxically, was a measure of the success of the mobilization in
Warlpiri terms. Warlpiri people took responsibility for control over it in the
same way that they generally seek to be custodians over their culture and
knowledge.
5. Conclusion: challenges for multimedia
This chapter has presented motivations for and examples of using multime-
dia and “thick interfaces” to mobilize language documentation in support of
endangered languages. We could summarize factors in creating such a
product as a set of heuristics all of which are essential for a “good” mobili-
zation:
– it supports exploration;
– its functions go beyond the underlying data;
– it manifests community input and participation;
– it fully exploits the capability to present and control sound;
– users can interact with content in relevant (and innovative) ways;
– it has design integrity.
Obviously, developing custom, high-quality multimedia applications to sup-
port endangered languages can consume considerable resources and require
solid dedication (see Nathan 2004 for more information about planning
such a project). In fact, such projects have occasionally been described as a
waste of scarce resources (e.g. Simpson 2003). However, if it is true that
developing multimedia can offer a distinct contribution in endangered lan-
guage situations then to call it a “waste” is to place community needs as a
low priority.