User Observation, Assignment 1

HCI 450 – Lawler

DePaul University Summer II 2001

Jeffrey Absher xxx-xx-xxxx

Frank, the subject, is a novice computer user of the “Baby Boomer” generation. He provides content for his own website, but performs none of the technical work. His level of knowledge of computers includes shallow abilities such as simple web browsing, email, and word processing. During a recent training session on a method to allow him more direct control of his website, I had the chance to note the difficulties and obstacles faced by novice computer users and the problems brought about by rote procedural learning of tasks in a complex environment.

The technical skills that are new to the user:

Download an HTML file off of the Web to a local directory.

Open and Edit the HTML file with Windows Notepad and add a list element to and unordered list.

Save the edited HTML file back to the local directory overwriting the old file.

View the HTML file for errors in a browser.

Attach the file to an Internet email and send to the webmaster. (already knows this)

The subject does not have a good concept of a computer’s file system structure (tree model) and therefore tends to “lose” files. The first part of training is to teach the user to download a file into a directory that he can easily find again. Windows95 tends to save files in the last Current Working Directory of the machine (not the application’s last CWD). The “Save As” dialog and the concept of a Current Working Directory is not understood by the user, so a procedure of always going to the “top” of the machine and finding the “My Documents” folder followed by explicitly naming the file is settled upon. I also put forth the procedural rule of always using “Save As” and never using the save button or menu item in all applications to solve the problem of “losing” documents. The tendancy of Windows to provide a guess at an initial filename also contributes to this loss of documents.

At this point the user feels that he understands, but I feel that he does not, and force him to practice by saving and reloading files with me resetting the Current Working Directory at various points in the process. This is generally how the pace of the training goes, the user is impatient and does not want to practice or experiment, he merely wants a procedure that works, but does not want to document that procedure (“I can remember it”). Simultaneously with his desire to learn only the steps instead of the underlying concepts such as file system structure and Current Working Directory, he feels frustration with the limited power of procedural computer usage. “Why won’t it just do what I want it to?” Answer: “What you want is somewhat different this time than it was last time and the procedure does not cover your new desire.” The pace of training of each element was a negotiation between the user’s feeling that he had it mastered because he could observe an expert perform the task (even slowly) with the task appearing easy and straightforward and the trainer’s knowledge that he had to practice the task in a variety of situations to be prepared for different variables that can crop up in computer usage.

The user does not enjoy trial and error and does not want to learn what can go wrong in this process. Staged (planned) obstacles to the completion of the task during practice time fall into 2 categories when encountered by the user: inconsequential (details the user does not notice as he blindly follows the procedure), and abortive (problems that cause the user to give up too easily and ask for help rather than attempting to solve himself.)

The user also experiences some dexterity problems with mouse and keyboard, hand-eye problems and interface conceptual problems. The actions and results of a click, double-click, right-click, click-and-drag and right-click-and-drag are not intuitive to some users. Double-clicks present problems, and the user is instructed to relax, but keep his finger in contact with the mousebutton all the time. He is reminded that clicking harder does not increase the chance of a double-click. Knowing what the enter key does at different times as well as knowing where typing would show up pose some problems. Small screens exacerbate this situation of mis-drag-and-drops and errant clicks. The concept of selecting the target object first followed by the action to perform on the target is problematic.

The interface does not do a good job of maintaining all of Normans principals. Most people tend to ignore the keyboard as a Window interface controller because it takes training to use; instead they become dependant on the mouse. The mouse and keyboard are a layer of abstraction away from the screen which is itself a layer away from the functional operation of the computer. Translating actions and options through this layer is clunky at best. Having only 2 buttons on a mouse to cut, copy, move, paste, scroll, invoke, open, link, select or to present a context sensitive menu of even more choices is problematic. There are either too many choices in the iconic interface or not enough controls to deal with the iconic interface. So good mappings are not maintained in this interface. Simultaneously, visibility is not maintained because by looking, you cannot tell all of the actions that can be performed on an object. At least one additional action is required to determine the actions that can be performed on an object (right-click). Computers have a good conceptual model, and though generally consistent, the model is complex and difficult to learn. The model is getting better all the time, but simultaneous with the improvement in the model, the interface is altered at each revision so that old habits must be unlearned (Windows 3.0, 3.1, 95, 98). As well more features and capabilities added to the system so that the system’s capabilities are always a generation ahead of its conceptual model. Feedback is getting better, but it is still only visual and auditory, tactile feedback that crosses through the interface layer of abstraction between the mouse and the display may or may not help (mouse heavier when dragging objects or similar). The user cannot feel that he successfully double-clicked, he only knows that he clicked twice.

The user experiences a high level of frustration that the computer does not know what he wants it to do, and expresses this frustration with both body language and willful excessive response time to directions and questions from the instructor. The body language includes facial gestures such as frowning, pursing of lips, and furrowed brow. Other full-body gestures such as sighing, adopting a posture indicative of frustration and fatigue, defensive arm-crossing, and excessive touching of the forehead (headache-style with eyes closed) and face (fist on chin similar to Rodin’s Thinker) and simulated hair-pulling also occur. I interpret his willful excessive response time in acting and responding as either a manner to ‘get back’ at the trainer who is wasting his time with practice that he feels he does not need or as a communication to the trainer of how much the training ‘hurts.’

Technical shortcomings of the system, internal and external mental blocks to training, and lack of motivation present problems for training a user on a computer system. For novices, the mouse is not an effective tool to manage the Windows interface. It is tough for novice users to determine the object of the system’s attention such as what is selected and where they are going to type. The interface is a reasonably effective tool to manage the underlying functionality, but the underlying functionality must be understood, and that requires a pre-existing knowledge of the underlying conceptual model. Users that are seeking a narrow pathway toward their goal in computing (procedure) will not perform experimentation and will give up easily in the case that the procedure fails. Psychological issues and interpersonal issues play a large role in technology training. Motivation plays a role as well. Trainees may use body language to communicate their actual stance toward training even if they are unwilling to speak it.

I feel that modern computer use requires a large, complex, and abstract conceptual model coupled with good hand-eye coordination, typing ability, and the ability to change focus and concentration between multiple levels quickly. It also requires the ability to easily switch mental problem-solving modes between a procedure-oriented mode and an improvisational mode based on the conceptual model. I also believe that guided trial-and-error training is the most effective way to instill a deep understanding of this conceptual model. Impatient users or users that have already discarded the goal of understanding the model and have surrendered to procedure pose difficulties for trainers. These users also may have problems due to pride and their public image as well as fears of aging and obsolescence. The user may want all of the benefits of being a competent computer user, yet not be willing to listen to and trust a less-powerful, threatening, and younger trainer. The user may not have the motivation required to perform the needed trial-and-error tasks to become competent. Many businesspersons are conditioned to be results-oriented, yet that stance does not lend itself to learning to manipulate a computer properly. Coupled with a genuine motivation in the user, a trainer may succeed if the trainer has an appropriate balance of the following attributes: Authoritative, Patient, non-judgmental, non-threatening, reassuring and encouraging. An unmotivated user or a user with external and personal blocks to learning presents major problems that training alone will not overcome.