Table of contents

DONE

User-tailorable systems: pressing the issues with buttons

paper

%3 cluster_e5a7b7dd_f649_4acc_9b3c_756dc23457a8 User-tailorable systems: pressing the issues with buttons cluster_da519456_1a21_428e_8bf7_494c5aa79784 The Tailoring culture cluster_505ef1b8_87ac_48cf_8dd1_c7361e162751 Tailoring techniques _4ea80fe7_f238_448f_aa75_39697ccdc253 Realising the tailorability promise _eb3bb549_20a5_4f1c_b256_5ffbf83d1528 Cooperating with users _e140b0f1_e148_4edc_991f_7ab586e860ee Helping the users find good solutions _fce800c7_9a6e_45bb_b600_a813797bedb8 Tailoring-Oriented architectures _9ef2fd93_3754_49df_a7a9_1ff0d9fc9884 Modifying program code _af40775a_9d91_4eb4_a9fd_7d8a3e4692c4 Copying and Specialising buttons _d38ab1b1_1f58_41e8_ad78_553a3dcba7f9 Object Oriented Programming _af40775a_9d91_4eb4_a9fd_7d8a3e4692c4->_d38ab1b1_1f58_41e8_ad78_553a3dcba7f9 _6b3347e3_62ee_4a73_bcee_14cc8f3b860b Building blocks _842ab92d_4aa4_4158_96de_61fa44776e22 Situated creation _f845ebe4_9a7a_4461_a3bc_8dd0bf2ec31a The buttons user's view _6d56a003_7d10_455e_84ed_5fc598fccd49 Lisp _fe0481be_aa15_4e49_a740_8c043763e925 Principles _fe0481be_aa15_4e49_a740_8c043763e925->__0:cluster_e5a7b7dd_f649_4acc_9b3c_756dc23457a8 __1:cluster_e5a7b7dd_f649_4acc_9b3c_756dc23457a8->_6d56a003_7d10_455e_84ed_5fc598fccd49

Uses a mountain as a metaphore of the irregular complexity curve required to first use a computer system, then tinker with it, then program it.

With this in mind, explores ways in which end-users can tailor their systems to themselved by developing a system based around on-screen buttons.

Users are just screen objects in Xerox Lisp which look "pressable" and when pressed carry out an action. The Buttons architecture allows users with little or no programmin experience to modify various aspects of buttons for themselves (the labels, te graphical image, aspects of the actions).

The Tailoring culture

We stress the importance of building a community with a culture of changing the workstation environment.

Helping the users find good solutions

When users are asked to carry out what might seem a simple interface design task - designing a set of abbreviations for a given command set - they do a very poor job. Users typically have more problems with the abbreviations they produce themselves than they do
with a set which has been designed to conform to a simple abbreviation rule structure.

One approach to help users better understand the possibilities for tailoring would be to design a system so that the range of variations and their consequences were a salient part of the design

Cooperating with users

The approach we employed here was to have a member of our design team (the “handyman”) working closely with the target users. This arrangement provided a mechanism for the designers to take careful account of the users’ real requirements and for the users to gain a better understanding of how their working environment could be different by helping design it themselves.

Tailoring techniques

One role of the handyman was to seed the environments of users with buttons appropriate for their own personal day to day activities.

For example, one of our administrative staff had the task of sending theweekly EuroPARC calendar out by email to a number of different people, to the nearest printer for some other people, and distributing hard-copies to yet other people. A button was produced to carry out most of these tasks with a. single mouse click.

a remarkable amount of tailoring can be done simply by “begging, stealing or borrowing” appropriate buttons and placing them in strategic places on the screen.

Situated creation

It relies on capturing relevant aspects of the system state into a button for later re-use. The idea is that the user carries out some task using normal manual methods, and is then able to encapsulate relevant parts into a button without doing anything which looks like programming

Differences with [ Robotic Process Automation ]

Unlike programming by example, this approach allows the computer to regenerate the state in the most efficient way (if the user got there through a long-winded route, that route is not preserved in the button).

Copying and Specialising buttons

[users] may have a button of their own, or one provided by a colleague, which does almost what they now want, “except for…“

Instead of using OOP-like inheritance, buttons are copied and edited.

user who wants to create a variant of an existing button would typically copy the entire button and then change a few details as necessary.

By making individual buttons independent objects they are conceptually simpler for the user to understand. In addition, if a user wants to send a button to someone else by email, it does not require the recipient’s environment to already contain a complex hierarchy of classes on which the button relies.

Tailoring-Oriented architectures

Boxer, DiSessa ues a concept, which he calls shallow structuring,meaning that “...anything the novice is likely to need to use or modify must be near the surface of the environment”

Modifying program code

Since the relevant piece of Lisp is encapsulated within the button, someone who has some feel for programming and who wants to carry
out minor modifications to the button is faced with a relatively small piece of code and so can be quite happy working out which part of the Lisp expression to change.

Building blocks

we need to encourage a constrained approach to creating new buttons. One obvious example is that we want a consistent interface style for users to interact with buttons.

We provide a set of user-interface building blocks that can be used when the button action causes some interaction with the user. For example, they provide information to the user, ask for yes/no responses, ask for string input and so on.

The buttons user's view

early on, users talked about Buttons as being “not my personal buttons” or being “sewn to the screen” (i.e. not under personal
control).

Later, we started getting quotes such as “I don’t know what I’d do without my Buttons" or “Buttons are my friends, always there...“. Note the use of “my” in these quotes. Buttons became perceived to be very personal

Realising the tailorability promise

Multiple ways of tailoring are good, they add "terraces" to the complexity curve "mountain".

A major reason for our success in enabling non-programming users to tailor their own workstation environment is that we have produced an architecture which supports a large number of tailoring techniques.