Symbiosis or serfdom?..."Are you/we/they being Served?"

1. Introduction (Burnard)

We often talk rather glibly about the "humanities computing community"
without much examination of the nature and needs of the communities
actually served. Perhaps if we understood the concerns and
requirements of our clientele a little better we would have a better
grasp on what distinguishes a a humanities computing unit from any
other kind of computing unit. In this session we will explore varied
perspectives on the relationship between the users and providers of
humanities computing services, as this has developed at our own
institution over the last quarter century.

We suggest that the key to understanding humanities computing is a
focus on the needs and requirements of our clientele. Because we are
located in a service department rather than an academic one, we have
grown up in a user-driven culture. At the same time, because
humanities computing was originally a minority enthusiasm, we have
also grown up with the need to stimulate demand, and to demonstrate
the benefits of leading edge technology to a sometimes skeptical
audience.

That audience has always extended beyond the immediate community of
academic teaching and research staff at our own University. Twenty
years ago, when perhaps only a handful of staff at UK universities
were aware of the potential uses of IT in teaching and research,
Oxford took a leading role in promoting and exploring its use
nationally and internationally. Twenty years on, as we try to balance
the demand for local services with the need to satisfy external
committments, we find that, increasingly, evangelism is less important
than evaluation.

In this open-ended debate, each of four components of the HCU will
present its own perspective on the same basic issues: which parts of
the diverse and dispersed user community does it seek to serve; how
has it responded to those user-needs in the past; to what extent has
it helped articulate and direct user needs; and what development of
them does it now anticipate. We plan to leave ample time for attendees
to contribute their own experience and views to what we anticipate
will be a lively and wide-ranging discussion.

1. The Oxford Text Archive (Morrison)

Satisfying the expectations and needs of our users has been a constant
challenge for the Oxford Text Archive for almost twenty-five years.
Discovering what these needs are and how they can be met has proved to
be an evolving task. Some of the factors which motivated these changes
will be examined during this presentation, specifically we want to
look at how the OTA has responded to user feedback in the past, and if
this feedback has shaped our current and future policies.

Up until the early nineties, the core users of the OTA formed a
relatively small and geographically divers group of humanities
scholars and computer-scientists who had a vested interest in the
creation and use of digital textual material. These early users tended
to be knowledgeable about the technology they were utilizing and,
despite the cumbersome nature of the early methods of transferring
texts to and from the OTA, most were satisfied with the services
provided. In the nineties, the means of accessing the OTA was modified
in order to exploit the new possibilities offered by the web, and our
user base increased significantly. How OTA policy was modified to meet
the needs of these new users will also be discussed here.

The OTA's involvement with the Arts and Humanities Data Service has
led to an on-going requirement to assess and be responsive to users'
needs -- which have proved to be both demanding and sometimes
contradictory. This has encouraged us to review the extent to which
services like the OTA can and should be user driven, and whether both
the users and the wider community might be better served in the
long-term by services adopting a more proactive approach.

As part of the AHDS, the OTA works closely with bodies such as the
UK's Arts and Humanities Research Board to ensure that publicly funded
research producing digital outputs is carried out in accordance
with the latest thinking on standards and best practice. To this end,
we offer a consultancy service and a programme of workshops and
publications which are designed to encourage scholars to evaluate
their approach to the techniques and tools which are used to produce
and exploit digital resources -- not simply for their own benefit, but
also for the benefit of future generations of scholars. In this way we
not only respond to the needs of our users for advice and guidance, we
also hope to influence the nature and scope of those needs, and try to
ensure that scholars are asking the "right" questions of their
material, of themselves, and of us.

2. The Centre for Humanities Computing (Lee)

The Centre for Humanities Computing (CHC) has been in existence for
over 12 years and its basic remit as the focal point of support for
the University of Oxford's humanities faculties has never actually
changed, though its user group has. As originally conceived, the CHC
was designed to fill the gap in IT provision for the humanities at
Oxford by providing a centralised centre of advice and for
hardware. This was needed because of the fragmented teaching structure
at Oxford, based on a collegiate system, which meant that Arts
lecturers had no central place to go to (such as a departmental IT
service). In this sense, our original user group consisted of the
small number of IT-literate Arts lecturers. In the early days their
demands were few but highly specialised, predictably to help them with
research projects.

Over the years this has changed. The user base expanded initially by
requests from graduates for access to specialised advice and
facilities, and then to undergraduates. As more and more of the
lecturers acquired personal computers in the nineties, a need
developed to make them aware of the potential use of IT in teaching
and research. We addressed this by running a full-day 'taster' course
billed as a 'Humanities Training Day' and open to staff and
postgraduates, which provided an overview of many areas of humanities
computing with guidance on follow-up courses. This turned out to be a
particularly effective strategy: in many cases, graduates attending
the course would then return to their faculties and enthuse their
supervisors. To date this course has been running for 10 years,
repeated 6-9 times each year, with average attandance numbering 25 on
each day).

Yet still the users demand more. Feedback to the CHC comes from a
variety of sources, the users themselves, and faculty IT
committees. Over the past three years the demand has been threefold:
specialised training for lecturers with direct reference to teaching;
specialised training for graduates to enhance their research skills;
and a structured course for humanities computing specialists. From a
handful of courses run on an idiosyncratic level throughout the early
nineties, the CHC now runs a prolonged and structured series of
courses throughout the entire academic year. In the first term these
mainly consist of training sessions for new humanities gradautes (many
of these are compulsory) followed in the second and third terms by a
series of modules (all voluntary) covering several areas in much
greater depth (e.g. XML, multimedia, databases, etc.).

In addition, throughout the year there are milestone one-day events
entitled 'Essential IT Skills for Humanities Lecturers', 'Humanities
Training Days', and 'Bibliographies and Bibliographic Concepts'. These
culminate in an annual national one-day colloquium held in the Oxford
Union Debating Chamber each May. In total the Centre trains or teaches
around 1,000 staff and students each year.

The CHC also maintains a user area which adapts to changing user
demands (i.e. by maintaining a stock of humanities computing titles)
and exapnding into new areas such as Digital Video. As more and more
computers become available to students aross campus (either through
the increase in personal computers or faculty computing rooms) the
need to make the CHC a centre for more specialised facilities has
increased (e.g. CD-ROM writing, scanning and OCR, software
depositories, etc.).

In presenting this history, we will try to assess the extent to
which changes have been reactive and the extent to which they been
proactive. We will argue that user feed back is one reliable way of
detecting emergent trends and hence avoiding being wrong-footed by
future developments.

3. Humbul: the Humanities Hub (Stephens)

In the 15 years or so since its inception Humbul has undergone many
changes, both in its technology and its user community. The early
Humbul gave its users a personalised service, mostly because there
were only a handful of users and they all seemed to be on first name
terms. As it grew in popularity this level of personalisation became
hard to maintain. With the move to the web in the mid 90s the feeling
that the individual user had any part to play in the information they
were presented with was almost completely lacking. The static nature
of the web, coupled with the lack of staff, time, or money meant that
Humbul became a fixed collection of links which, if anything,
reflected the interests of the few people who could spare the time to
keep it up to date.

Now, as a JISC funded service, we are again bringing user requirements
back into the picture. We are trying to greatly extend the number of
people involved in collecting and cataloguing links for the
collection. We have developed a web based cataloguing system so that
cataloguers can be distributed across the world and can work at their
own convenience.

The resulting records, stored in a database, are free from any fixed
way of viewing them. The user can already choose from a number of
different ways of looking at the data as records are catalogued, for
instance, under different resource types, intended audience, or
period. We are currently in the process of greatly extending this.

By tracking what our users do, and allowing them to set certain
options for themselves, we hope to be able to profile individuals and
set up a 'MyHumbul' view. The user can set parameters on the type of
records they are interested in, for instance: primary sources in
Philosophy or Religion & Theology records covering such and such a
period, aimed at postgrads. They can choose to be alerted by email
when new resources appear which fall within their criteria. They can
mark records as 'of interest' when browsing or searching the Humbul
collection and add them to their profile. They can submit a query to
the search service and be alerted by email if a resource which fits
their query becomes available.

The user also has various options for re-using the data
elsewhere. They can dynamically include Humbul content in their own
web site, they can also choose to have the data delivered to them in
XML, or as plain text, or maybe as a bibliographic record suitable for
importing into Endnote.

This is all a long way from the sort of personal individual service
with which Humbul began and implementing the technologies to allow
profiling and individualization in the present enviroment brings many
headaches. In this presentation I will give an idea of some of the
ways we are bringing Humbul back to its roots.

4. The Humanities Computing Development Team (Porter)

The HCDT was set up to meet a specific need expressed by Humanities
users at the University of Oxford, when they were surveyed in 1998
about their use of technology in teaching and research. According to
the survey, Humanities staff would like to make more use of technology
but did not consider that they had the necessary expertise or time:
the HCDT was set up to meet both these needs, and thus to fill a gap
in the services provided by the other sections of the HCU, by
facilitating collaborative development of small, medium and
large-scale web sites with a specific teaching or research focus.

As a dedicated technical development unit, the HCDT has three different
levels of users with whom there are varying degrees of interaction: the
content creator; the system maintainer, and the end-user (which can in
itself contain several sub-categories of use). There is regular contact
with the content creator so we can ensure that their needs are being
met; but it is more difficult to define the needs of the system maintainer
and end-user. We have developed a process to try to accomplish this
through a combination of informed prediction of needs, surveying expert
users, and focussed end-user feedback and discussion sessions.

We applied this three-stage process to the development of the Sphakia
Survey Internet Edition that was developed for, and is used by, a wide
range of users including everyday people with an interest in the
region; researchers in classical archaeology and history; and students in
these subject areas. This position paper will report upon the final
evaluation of the Edition by the end-users and use the findings to the
evaluate the success of this three-stage process in the design of
user-oriented, teaching and research web sites.