Dear Dick,

I like to use a variation of your "elevator ride talk."  In my
version, the student takes the lift with the Head of
Department (HoD), and the lift gets stuck for some period of
time.

To encourage PhD students to develop ways of telling others
what they are doing, why, and how, I use the elevator ride
scenario in two ways: in one version, I say for how long the
lift is stuck for, two minutes, five minutes, eight minutes,
twelve minutes; in the second version, I don't say, I just let
the student start talking, without knowing for how long he or
she has to entertain the HoD.

I think it is useful to play this game with a student every
six to nine months, and I encourage the student to go away and
write up what they just told me (posing as the HoD).  This
way, it is easier for both of us to see how the student's own
ways of thinking about and talking about his or her work
changes and develops over time. If it doesn't change much, it
is, I think, a sign that not much is happening, so may be the
student is stuck.  If it changes a lot, it is a sign that the
student has not yet settled into a research line, or is
wandering off one.

Most of my students think all this is a kind of supervisor
inflicted torture, and some complain, asking what the point
is--they wouldn't talk to the HoD anyway, even if they did get
stuck in the lift with him or her.

Nonetheless, like you, I do think that it is very important
for researchers to have practised ways of telling others about
what they do, why, and how.  For me, good communication is
basic to doing good research.  And I think it is important
that each student develops his or her own way of telling
others about what he or she does.  I tend to discourage the use
of existing categories and classifications, since it is often
not clear if these are simply adopted for convenience, and not
for good reason.  Afterwards, it is possible to map on to more
accepted terms and categories, if necessary.

I started playing this game after I examined a PhD thesis on a
technique for reasoning about and planning the robot
manipulation of objects involving continuous contacts--sliding
pieces into place, for example. This was submitted by a
student in an Artificial Intelligence (AI) Department, but I
was surprised that the student displayed an almost total
ignorance of the field and how his work was properly situated
in it, and shocked by his lack of concern for the need to do
so.

I argued (with the other examiner) that without a proper
explanation of how this work was properly situated in the
field of AI, and thus how it related to things other than
other robot manipulation planning techniques, it was not
possible to identify and establish that a proper contribution
had been made by this work.  On that occasion, I was the
junior internal examiner, and I failed to convince the more
senior external examiner, but I did decide to try to help
my own students to avoid ending up in the same situation.

What would you, and others think, if peopled used this list as
a virtual elevator, from time to time? Two minutes is about
180 words.

Best regards,

Tim

================

>Dear Karel,
>
>Thanks for your response.  It is very helpful--an interesting set of
>distinctions.
>
>I wonder if you agree that it can be important for a doctoral student to
>understand how their research is positioned in these broader terms?  It
>seems so to me.  We speak of the "elevator ride talk" for graduate
>students--imagine that you get into an elevator with a CEO or some other
>person and he or she asks what you are working on.  One has only the
>duration of the elevator ride to convey the essence of one's work.  It is a
>good exercise, I think--and realistic, too.  I think your distinctions help
>in that direction, as well as others.
>
>I will think more about your distinctions and try to map some student work
>that I know about.
>
>Regards,
>
>Dick
>
>
>Richard Buchanan
>Carnegie Mellon University
>
>--On Thursday, December 16, 2004 7:53 PM +0100 Karel van der Waarde
><[log in to unmask]> wrote:
>
>>Dear all,
>>
>>A while ago, I collected about 70 PhD theses in the field of visual
>>communications (fairly loosely defined.) I found it useful to plot
>>these theses in two diagrams or maps:
>>
>>Map one: a square with the following corners:
>>- investigating an historical topic,
>>- investigating an educational topic,
>>- investigating a practical topic (topic in design practice),
>>- investigating a research topic (topic in design research).
>>
>>Map two: a triangle with the following corners:
>>- philosophical thesis: developing arguments
>>- empirical thesis: experiment based, developing interpretation of new
>>data - descriptive thesis: artefact based: developing new artefacts
>>
>>Disserations that appear close together on both maps have several
>>things in common, but it became also clear that these can have very
>>different theoretical bases.
>>
>>(An advantage of this kind of mapping is that the 'white areas'
>>become obvious ... If anyone is looking for a research topic in
>>visual communication?)
>>
>>Kind regards,
>>Karel.
>>[log in to unmask]