----------------------------Original message----------------------------
 
               State University of New York at Stony Brook
                       Stony Brook, NY 11794-3331
 
                                            David Y Allen
                                            Library-Reference
                                            516 632-7110
                                            09-Feb-1996 04:43pm EST
FROM:  DYALLEN
TO:    Remote Addressee                     ( [log in to unmask] )
 
Subject: GIS and levels of service
 
        This is a very late response to Janet Collins' questions about GIS and
levels of service.  I think she raises some very important questions which have
largely gone unanswered, and after much thought I am still not sure how to
answer them, but here are my latest thoughts on the subject.
 
        I have spent many, many hours learning how to use ArcView 2 and First
Street, and have by no means mastered all features in these programs.  I have
found the programs frustratingly difficult to work with--in part because of
poor written documentation and grudging support from ESRI.  It would be helpful
if there were a ListServ just for ArcView users.
 
 
        I have learned the programs well enough to produce maps, although so
far the demand for them has not been heavy.  Even when students come asking for
a computer produced map, I can often meet their needs better and more quickly
by photocopying a sheet map or a page from an atlas.  I do make maps for
patrons using ArcView, and it takes me anywhere from 30 minutes to 1 hour to
create a map.  We do not charge for maps, but for reasons of time I have
discouraged professors from making assignments requiring maps and refused the
occasional request to "produce all possible demographic maps" of an area.
 
        I would not consider the option of pointing a student to the computer
and saying good luck.  I have written out five pages of fairly detailed
instructions on how to use ArcView with First Street, and one faculty member
has told me that he has found my instructions more helpful than anything
written by ESRI or Wessex in learning how to use the programs (faint praise
indeed).  Only a few graduate students and faculty have the need or motivation
to learn how to use the programs on their own.
 
 
        Displaying raster maps and printing simple outline maps is not much of
a problem.  The crunch comes with complicated demographic maps where you have
to choose variables and make decisions on how to display them.  The concerns of
people like Larry Keenan have considerable justification in my eyes.  The
making of good demographic maps requires the complex skills of a professional
mapmaker, and most librarians don't have the time to learn or apply them.  Our
forte is acquiring, storing, and making information available, including
cartographic data in digital form.  Teaching people how to make maps with this
data should probably be the task of the faculty in various departments.  The
closest analogy is using programs like SPSS to analyze demographic data.
Librarians receive the data and make it available, but we don't analyze it for
our patrons.
 
        I find it difficult to deduce from the above considerations any very
definite answer to the question about "what levels of service" we should be
providing.  I suppose it depends on the level of demand, as well as on staffing
considerations, and even on the abilities and inclinations of individual map
librarians.