IS 3055: Advanced Topics in Information Science

    Augmenting Human Activity through Document Tools

    IS 3055
    Fall 1994(95-1)
    Thursday, 12:00-3:00
    Room 533

    Michael B. Spring
    Department of Information Science
    University of Pittsburgh
    spring+@pitt.edu

    Introduction

    This seminar will address the history and current research on building systems to augment human endeavors, particularly those related to documents. It focuses on human intellectual activity that is heavily dependent upon the manipulation of information in weakly structured form, i.e. documents. (Documents may be contrasted with information resources that are highly structured such as records and databases and information that is highly unstructured-text and bit streams.) The partciular focus of this seminar will be on those processes which may be automated so as to augment human intellectual activity. The focus of augmentation, as it is used here, is to use automated systems to free the human to engage in the core intellectual activity with minimal interference. For example, making a comment on a paper involves:

    • starting a word processor,
    • writing the comment,
    • naming the comment file,
    • storing the comment file, and
    • linking it to the original file.

    One approach to augmentation would be to automate steps 1, 3, 4, and 5 to the point of transparency.

    Goals

    The goals of the seminar are:

    • to provide an introduction to some of the literature on the topic.
    • to ground the literature review with work on a project that seeks to achieve some of the goals discussed in the seminar.
    • to provide an opportunity for the participants to engage in a discussion of the literature and theory in the area.
    • to establish a framework for a research seminar in this area.

    Organization

    The seminar will be broken down into four parts:

    • Over the first three weeks, the participants will be introduced to a software project written in C and X. The project provides a simple framework for document viewing and linking. As one of the requirements of the seminar, particpants will develop a component of the framework that aids individuals or groups in their interactions based upon the document structure and system information.
    • Weeks four through eight will be devoted to a discussion of the papers listed in this preliminary syllabus. The seminar leader will be responsible for framing this discussion. The particpants will be expected to develop positions on the papers and to bring other readings to the attention of the group.
    • Weeks nine through eleven will be devoted to particpant-led discussions based upon papers provided by the particpant-leader(s) for the week. It is expected that each particpant will have selected a narrow focus within the area for investigation and presentation.
    • Weeks twelve through fifteen will be used to demonstrate and discuss software modules developed by the particpants. These sessions will focus on walkthroughs of the projects and suggestions for improvement and testing.

    Outcomes

    There are three outcomes expected in this seminar:

    • Each participant is expected to develop an overview of the literature in the area to the point where they will be able to identify several additional papers in the area, read and digest those papers, and guide a class discussion of the papers.
    • Each particpant is expected to write a review of the literature that begins with the papers discussed in the class and continues on to other relevant papers in some subarea of the participants choice. The focus of the review should be to raise an issue or make a point about what should be possible or might be done by way of further research in this area.
    • Each participant is expected to develop a software module for the document viewer described below so as to enhance the capabilities of individuals working with the document set.

    Based on the seminar, participants will be invited to participate in a research seminar in the next term that would begin with the literature review and project module developed during the seminar. The research seminar will build on this seminar in two ways. First, a mini experiment will be set up to test whether or not the project module achieves the stated goals. Based on that experiment, the particpants will develop a research proposal and/or a draft article for publication; both will be based upon the literature review conducted during this first seminar.

    "fv": a Description

    The seminar will involve system design based upon an existing software called "fv"-faculty viewer. While "fv" originated as a tool for assisting faculty in viewing and commenting on a document collection, it is a general tool for use by a group in document manipulation. It began as a simple viewer that allowed the user to rapidly traverse a directory structure taking actions on the document being viewed. A set of buttons is dynamically allocated based on the contents of some starting directory. One button identifies the parent directory. The other buttons identify all files in the directory, parsing and identifying those that are subdirectories. When a button is pushed, if it refers to a subdirectory, the subdirectory becomes the current directory and the buttons are redefined for that directory. If it is an ordinary file, it is opened for viewing. (At the current time, all ordinary files are considered to be ASCII text files with the possibility of embedded hypertext links. It is anticipated that one enhancement to the system will be the development of facilities for handling other document types.)

    A file that is being viewed may be printed, paged, mailed, etc. The displayed file is parsed at display time to identify embedded links to other documents. If the view space is the recipient of a button click, the location of the click is checked. If it is on a link, the link is traversed. If it is not on a link, one is inserted, and a file is named (algorithmically) and opened to allow input. When closed, the file is saved in a subdirectory in the current directory that identifies it as a subdirectory related to the file which had been the subject of the comment. The file name reflects the author and the date/time of comment.

    The software has been developed to allow extension in a number of ways. The participants will select some function and code it in the framework. For example, the seminar leader is interested in developing a "caucus" button that would allow an individual reading a document to discuss it with other members of the group responsible for the document. In a preliminary sketch of the functionality, it seems possible that a significant amount of the effort required to set up and manage such a meeting can be conducted with no input from the user beyond the data already embedded in the system. Other kinds of "buttons" might include analysis of the exiting document structure, the development of active documents to collect information, structured authoring interfaces, and the construction of linear documents from the heirarchy. Participants will be expected to choose one of these or to define some other augmentation feature to implement.

    Preliminary Reading List

    Week Four: History

    Rayward, W.B., Visions of Xanadu: Paul Otlet(1868-1944) and Hypertext, JASIS, Vol 45, No. 4, May, 1994, pp235-250.

    Bush, Vannevar, As We May Think(Reprint), A History of Personal Workstations, 1988 ACM Press, Addison-Wesley Publishing Company New York, New, York

    Engelbart, Douglas C., Special Considerations of the Individual as a User, Generator, and Retriever of Information, American Documentation, April 1961 Berkeley, California

    Miller, George A., Psychology and Information, American Documentation, July 1968

    Kay, Alan, The Dynabook-Past, Present, and Future, ACM Press, Addison-Wesley Publishing Company, 1988

    Week Five: A Framework

    Engelbart, Douglas C., Vistas in Information Handling, Washington, DC Spartan Books, 1963

    Week Six: Models for Documents

    Belkin, Nicholas J. and Croft, W. Bruce, Information Filtering and Information Retrieval: Two Sides of the Same Coin?, Communications of the ACM, December 1992/Vol.35, No. 12

    Rosenberg, Jonathan, Kraut, Robert E., Gomez, Louis, Buzzard, C. Alan, Multimedia Communications for Users, IEEE Communications Magazine, May 1992

    Halasz, Frank and Schwartz, Mayer, The Dexter Hypertext Model, Communications of the ACM, February 1994/Vol 37, No. 2

    Hardman, Lynda, Bulterman, Dick C.A. and Rossum, Guido Van, The Amsterdam Hypermedia Model, Communications of the ACM, February 1994/Vol. 37, No. 2

    Week Seven: Computer Supported Collaborative Work Overview

    Grudin, Jonathan, Eight Challenges for Developers, Communications of the ACM, January 1994, Vol. 37, No. 1

    Grudin, Jonathan, CSCW Introduction, Communications of the ACM, December 1991/Vol. 34, No. 12

    Reinhard, Walter, and Jean Scheitzer, Gerd Volksen, and Michael Weber, CSCW Tools: Concepts and Architectures, IEEE Computer, Vol 27, No. 5, May 1994, pp 19-27.

    Bently, Richard, Tom Rodden, Pete Sawyer, and Ian Sommerville, Architectural Support for Cooperative Multiuser Interfaces, IEEE Computer, Vol 27, No. 5, May 1994, pp 28-36.

    Nunamaker, J.F., and Alan Dennis, Joseph Valacich, Douglas Vogel, and Joey George, Electronic Meeting Systems to Support Group Work, Communications of the ACM, July, 1991, Vol 34, No. 7, pp 40-61.

    Week Eight: Ubiquitous Computing and Agents

    Ferguson, Innes A., Touring Machines: Autonomous Agents with Attitudes, IEEE Communications Magazine, May 1992

    Lee, Kuo-Chu, William Mansfield, and Amit Sheth, A Framework for Controlling Cooperative Agents, IEEE Computer, Vol. 26, No. 7, July, 1993, pp8-17.

    Weiser, Mark, Some Computer Science Issues in Ubiquitous Computing, Communications of the ACM, July 1993/Vol. 36,No.7

    Wellner, Pierre, Digitaldesk, Communications of the ACM, July 1993/Vol. 36, No.7

    Suggestions for

    FV enhancements

    • Develop a editor interface that produces HTML compliant files. This will entail changing the current anchor structure to the HTML anchor structure.
      • The system should make it painless to develop an HTML structure. This means the system should allow the user to insert anchors and URL's to files of all types, both local and remote. The key to the editing process should be simplicity in construction based either on an existing flat file or development from scratch.
      • FV need only handle those files that are HTML or plain text. For all other types it would be sufficient to pop up a box indicating the path to the named file. It would be nice to verify the existence of those files-remote or local.
    • Develop an authoring control system that would allow the author of a document to assign it to various groups for comment. This system would check at startup time for an individuals groups and set the group id as appropriate. This would restrict the files which an individual could comment on. Several things would have to be examined here:
      • Should an individual only see files for which they are authorized comment, or should the option be available to them to see all or restricted groups. Maybe a semi-hazing of the files to indicate those they can comment on.
      • If a user wishes to comment on a file, and their current gid is not appropriate but they belong to a gid which could be used, shouldthey be shifted to that.
      • What are the relative costs and benefits of maintaining a separate authorization file.