Helma's and Daniel's comments reinforce the sobering truth that we at
Alpheios were forced to recognize only reluctantly: there is no type or
degree of computer support that will be optimal for all students at all
points in their study. As a result, we have added much more variety than we
originally intended and expect that forn any given user at any given moment
MOST of this support should probably be disabled or unused.
The challenge then of course becomes how best to match the available
resources to the tastes and abilities of the student confronted with a given
text at a given point in his studies. Over time, as the application becomes
familiar with the individual student's grasp of vocabulary and grammar, it
should
be able to make some adjustments automatically, but a large number of relevant
variables will remain inaccessible to it. Under the circumstances, we are
choosing
to implement far more tools and approaches than any student is likely to need
or want in the hope that he or she (or a mentor) will be able to explore the
resources and find the mix that is most effective and/or congenial for them-
whether aligning texts with translations, examining original text
with just the aid of morphology tools or just the aid of lexicographic tools,
filling in details in tree diagrams, responding to explicit quizzes or
relying on
the system's ability to detect one's changing proficiency, etc. etc. This is
not an entirely satisfactory state of affairs, and it is hoped that we may have
a chance to explore the implications of individual user characteristics and
attempt to adapt to them automatically more fully in future releases. Taking
into account individual variables such as memory buffer size and the optimal
interval for review might well contribute to reducing the challenge that these
languages seem to present to many students. Although the computer
initially perhaps often aggravates the unfortunate consequences of an approach
that disregards individual differences, it has at least the potential, over
time, to
begin to take such differences into account. So we are certainly keenly
aware of the preliminary nature of our tools and would welcome any opinions
on where they seem to succeed or fail most conspicuously.
Harry Diakoff for the Alpheios Project
|