Dear Charles,

It is good to ask the hard questions - and the obvious ones.

At 7:13 PM -0500 13/11/97, Charles Rhyne wrote:

>Dear Jennifer,
>
>Many thanks for your most informative answer to what may have seemed like a
>provocative message (partly intended as such though principally aimed at
>sharing ideas).  It is impressive to see how carefully you have thought through
>the various steps involved in the project.  I was especially pleased to see
>your reference to the interest expressed by "creative/studio arts programs,
>computer science programs, cultural history programs and communications
>programs, to name a few," since my experience is that different disciplines
>have quite different needs, often ones you or I might not foresee.

These are often the traditionally underserved as well, but their interest in using visual materials is increasing. This increasing demand can actually be used to benefit 'traditional' users, as increased users can be used to leverage increased resources.

>
>The only difference I can see between the approach you are taking (some of
>which I was unaware of) and the one I should suggest is reflected in:  "We've
>opted for some use, across the board rather than a project tailored to the
>detailed desires of a few."  In order to see what it would be like to teach a
>course (in any discipline) using all or nearly all digital images, one would
>have to run a number of experiments in which a number of projects WERE tailored
>to the detailed desires of a few.  One might think of a high school courses in
>American history, a college course in the culture of the Italian Renaissance,
>and a graduate course in art conservation.  By providing a concentration of
>images in one or more of these areas, in consultation with those who would be
>teaching these courses, one would discover how such images might be used in a
>few specialized situations.  Even though specialized, these lessons might have
>more relevance for other types of courses in other areas than the lessons
>learned from the availability of a few images in all areas.

There are projects like this now on individual campuses, and they are being reported. Your work, and MESL, the work at the University of Michigan, all come to mind instantly. However, teaching with 'digital images' sometimes strikes me like deciding to teach with 'articles'. There are different sources that are good for communicating different things. Abandoning the textbook means loosing the overview. Teaching only with 'primary sources' means that basic facts are sometimes missed... I guess I'm more interested in learning what the basic elements of the learning experience are, and then being able to offer tools and content that can be put together to support it.

>
>It's my impression that, except for a few teachers who enjoy experimenting with
>computer images, most teachers are too busy to take on such a new, potentially
>troublesome technology.  It is the willingness of the teachers to actually use
>these images that seems to me your most difficult challenge (if I have a
>central point, thhis is it). For this, what may be needed is a rather dramatic
>breakthrough in which it can be shown (in a few select cases) how much better
>the teaching and learning experience can be with these images.  But this
>requires not making the teacher's life more complicated (with multiple image
>formats) but making it simpler with a single digital image format at high
>quality, readily available in the classroom and study workstations.
>

I fear that the highly funded case studies will remain just that, individual instances where someone 'got the money' and then based on their enthusiasm, created a course that only works for them. Personally, I'm in favour of a much more incremental approach, where digital manipulation unit, for example, might be built into a course on conservation, or drawing, or printmaking - all places where being able to see the incremental layers of image creation is essential.

We can enable this through delivery systems that work, and a library that works as a whole. This predictablility in content is at the core of our specifications for the AMICO Library. we can also facilitate the sharing of experiences between faculty and students, so that people can learn from each other, as they are doing something.

>Thus I think "significant bodies of material being created in museums for other
>purposes - new acquisitions, exhibition and research projects, reinstallations,
>publications" may provide more useful mini-units than a scattering of images
>over large fields.  In looking at "a content analysis of the slide libraries of
>major institutions, or at least the slide lists of professors," I hope you
>would consider the personal slide collctions of many teachers formed
>specifically to enable them to teach in depth on subjects of their special
>interest, not provided for by institutional slide collections.  

In selecting content to contribute AMICO museums identified two primary concerns. The first was for a representative sample - are the things that peole know are in your collection in your contribution (because if they are not, then the user will think that the computer is broken). This is paired with the desire to provide depth in areas of strength - the Italian drawings at the Art Institute of Chicago for example [there are 65 in the testbed data set]. The initial selection has some of both of these characteristics. Designing strategies for growing the Library are certainly part of our agenda.

>
>Since you know of my interest in the potential of the internet for not only
>faculty but also student and public research, I especially applaud those
>museums who are planning to provide access to portions of their  curatorial
>files, "unprecidented" as you say.
> 
>Regarding my definition of high quality, as you might guess my definition is
>not a technical one but a use definition.  I want viewers to be able to see
>whatever is needed to answer whatever question they are asking.  Thus the
>quality required depends on the use.  In most cases, I should like the quality
>of the image of a painting to answer the questions a viewer might have in front
>of the painting in a typical art gallery.  In an art conservation course, one
>would want the quality to answer the types of questions on would have in
>examining the painting in a conservation lab.  On the one hand, one might think
>that this was too demanding a standard, because it is always better to be able
>to see the original, direct.   However,. the light levels allowed in most
>galleries are (quite properly) too low to see everything even the normally
>inquisiitve public might want to see.  High quality digital images allow one to
>see aspects of the painting not visible in the gallery, just as using a
>flashlight canhelp in a dimly lit gallery (I'll send you a recent article). 
>This I think a really central issue and one I hope to examine in more depth
>myself.

I've seen the article, - if it is the VR piece with your students that you are referring to. One of the research projects I scoped for the GII but didn't complete before I left was a series of case studies designed to analyse users, uses and required quality levels, I hope to come back to this, as we still have a lot to learn about, in the language of systems, users and their functional requirements.

>
>At the moment I'm dashing off to a seminar with Robert Morris, who is on campus
>this week as one of our "distinguished visitors" ( sounds formidable), so am
>mailing this without proof-reading.  Hope it sounds largely supportive with
>hopefully a few worthwhile ideas.

I'm back to writing AMICO Data specficiations - and then will be travelling for a couple of weeks. Thanks for your thoughts.

jt