<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body bgcolor="#ffffff" text="#000000">
Dear All,<br>
<br>
A quick update after the additional conversation Tom and I had with
Archaeology MA students and the meeting I had with Mike J regarding
the integration of intervention 3 with the back-end.<br>
<br>
We met 2 more students, one from Osteoarchaeology and one from
Palaeolithic Archaeology. The main things we learned are:<br>
<ul>
<li>Master projects seem not to directly follow the kind of
process we expected and discussed. In particular none of the two
we met are going to use the labs on the Avenue Campus for their
projects: one student (palaeo-) will do experiments with
primates to study communication mechanisms, the other student
has two options, either classify bones on a site in Hungary, or
work with human bones in Portsmouth (it is unlikely these will
be taken to Southampton).</li>
<li>Students tend to keep notes from their UG and master projects
for years. These are often not used, and just sit in a drawer.
At least in one case the notes were added as an appendix to the
submitted project report, however, this does not seem to be
common practice: normally the notes are kept by the student and
only the report is submitted. So we can use these notes as
testing material.</li>
</ul>
<br>
Based on these two points, here are our (=Tom and me) thoughts: we
should try and get available notes as soon as possible (Tom should
be on the case), first of all to analyse them and then to enter them
into our system. In this way we can completely decouple the "input"
and "output" aspects of the study. Entering the pre-existing notes
in batch can also help us validate/debug the basic functionality of
the system. <br>
<br>
So on one hand we will have an experiment where we compare finding
information using our system to the use of the original notes and
the original report based on the same notes. This will basically be
an information retrieval task (there are studies around this topic,
so I will try to look for specific references we can use). One
question that still needs to be clarified, though, is what exactly
to expect from this comparison (beyond obvious advantages of
electronic systems). This strongly depends on the design of the
"output side" of the system, on which I am still working. In general
I hope that ideas from provenance can be integrated here (Luc, I am
looking at the documentation on the opm website to get ideas..).<br>
<br>
We will still need to show that it is also possible to use the
system to enter notes directly (rather than adding pre-existing
ones), so we still need to run a "field" (=not usability lab) study.
To add to the motivation for participants to use the system, it
would be good to either have multiple users working around the same
topic (so by using the system they learn about each other's work) or
have the system pre-populated with information that is relevant to
the participants. The topics of the MA projects seem to be quite
scattered, so we are still trying to figure out who else could take
part in this -- perhaps we could do something related to Portus? We
are planning to talk with Pina (a PhD student involved in Portus,
Tom to arrange this meeting), with a PhD student (the only one) in
osteoarchaeology (Angeliki should set up that meeting) and with the
supervisor of Osteo-MA students (Graeme, Angeliki & Tom
suggested that you arrange this meeting). I also just sent an email
to Leif asking for more information about the digital data related
to Portus.<br>
<br>
From the meeting with Mike J about the integration of intervention 3
and the ontology, it seems that the intervention and the ontology
are quite aligned, thanks also to recent discussion between Mike and
Luc. Open issues (on which I need to work) include:<br>
<ul>
<li>finalizing the input side of the interface, including deleting
notes and handling user accounts</li>
<li>design the output side of the interface<br>
</li>
<li>list the back-end calls that the interface needs (e.g. from
create and edit note to search operations) -- it seems that most
(if not all) of these are already supported <br>
</li>
</ul>
Tom & Mike, is there anything I forgot?<br>
<br>
As usual, feedback is welcome. <br>
<br>
In general, my apologies for slow responses on this, but I am very
busy with preparing the lessons for a new course I teach in the
second half of this semester (on Computer Vision).<br>
<br>
Thanks,<br>
Enrico<br>
<pre class="moz-signature" cols="72">--
Dr Enrico Costanza
Lecturer, Intelligence, Agents, Multimedia Group
School of Electronics and Computer Science
University of Southampton, UK, SO17 1BJ
<a class="moz-txt-link-freetext" href="http://users.ecs.soton.ac.uk/ec">http://users.ecs.soton.ac.uk/ec</a>
<a class="moz-txt-link-freetext" href="http://d-touch.org">http://d-touch.org</a>
</pre>
</body>
</html>