Problem, solved.

Introduction
The growth of mobile systems (Personal Data Assistant, mobile handsets, etc.) sets new
methodological and technical problems if we intend to address the question of
studying, empirically, their use and their usability in wild field. Indeed, as written
by Zouinar and coll. (Zouinar, 2004), users’ mobility implies not only a change of environment
but also displacements which make more complex interaction with the system and context
data collection, than basic posted activities. Among the rare studies devoted to the
empirical interaction study with mobile systems (Beck, 2003), two approaches, based on
different epistemological approaches can be distinguished.
The first one consists in studying the interaction in laboratory. Underlies by the search for a
control of the variables which can affect the interaction, this approach conduces to put the user
artificially in mobility situation. For example, during the test is has been asked to
the user to walk on a running band (Pirhonene, 2002), (Beck, 2003). More generally, with this
experimental approach, the assumption is that it is possible to recreate in laboratory a “natural”
situation of user mobility. From this point of view, mobility is primarily considered
under its physical and awareness demonstrations (Beck, 2003). Such a reductionism is
faced with the traditional problem of the ecological validity of the data collected.
A displacement proposes to the actor more or less important context changes, not only
physically (lights, sounds, objects, etc.) but also socio- culturally (places, people,
situations, etc). Even if they do not have an automatic incidence on the activity, these
variations are potentially relevant and influence directly its course. Moreover, the
human-machine interaction generates itself its own context of achievement and
dynamics (historical), focused on the realization of a given goal (search for information, writing
messages, etc.) Mobility underlies complex problems of articulation and combination between various
types of context and activity.
Following works initiated and developed by Hutchins, Hollan, Kish, etc. (Hutchins, 1995), (Hollan,
2000), (Kirsh, 1999) a complementary way to the experimental approach, stresses the
design underlies by observing in natural situation, i.e. out of laboratory (Lyons,
2001), (Oviatt, 2000), (Bationo, 2003) etc. With this approach, we preserve, as much as possible,
the natural character of mobility and consequently its context. Doing so, we gain in ecological
relevance what we lose in control of the situation. However, one of the main challenges of the
naturalist approach is to use technologies to study empirically the use and the usability of mobile
technologies in wild field. It is in order to bring some elements that we describe briefly the
laboratory LUTIN, and then, that we
describe how we seek to use this methodology in wild field.