See what you need: Helping end-users to build abstractions

被引:12
作者
Blackwell, AF [1 ]
机构
[1] Univ Cambridge, Comp Lab, Cambridge CB2 3QG, England
基金
英国工程与自然科学研究理事会;
关键词
VPL-II; A.10; programming-by-demonstration languages; VPL-III; A.1; data abstraction; VPL-VI; C.3; liveness; D.1; usability studies; D.2; cognitive and user-interface design issues;
D O I
10.1006/jvlc.2001.0216
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
It has been argued that most end-users would not want to write programs, no matter what kind of tools they had to help them. This paper analyzes the reasons for that situation, and addresses them with a new approach to programming. There is a large class of users who are very reluctant to create any kind of abstraction within a computer (they will not create a new directory, let alone a Word macro). But these people do create abstractions in their everyday lives outside the computer. The reason for their computer habits can be explained in terms of a simple investment and risk model. See What You Need (SWYN) is a system intended to deliver the capabilities-not just of Word macros, but of Perl-to end-users. The techniques it adopts are those of programming by example, visual program representation and direct manipulation. SWYN is not intended to be a complete system, and is unlikely to develop into a complete Perl replacement. It is a research vehicle, supporting the evaluation of specific features aimed at reducing the perceived risk of abstraction in programming. This paper describes two elements of the SWYN project, both of them related to the regular expressions that are central to pattern matching in Perl. The first is a prototype system that acquires regular expressions from the type of examples required in direct manipulation of data for programming. The second is an experimental evaluation of alternative notations for visualizing regular expressions. This experiment demonstrated a dear advantage for graphical notations in typical end-user tasks, both over conventional regular expressions, and over a tutorial-style textual notation. The overall effect is a reduction both in programming investment and in perceived risk. (C) 2001 Academic Press.
引用
收藏
页码:475 / 499
页数:25
相关论文
共 28 条
[1]   INDUCTIVE INFERENCE OF FORMAL LANGUAGES FROM POSITIVE DATA [J].
ANGLUIN, D .
INFORMATION AND CONTROL, 1980, 45 (02) :117-135
[2]  
[Anonymous], 1983, IEEE Computer
[3]  
Blackwell A., 1999, 11 ANN WORKSH PSYCH, P24
[4]   Pictorial representation and metaphor in visual language design [J].
Blackwell, AF .
JOURNAL OF VISUAL LANGUAGES AND COMPUTING, 2001, 12 (03) :223-252
[5]   Metacognitive theories of visual programming: What do we think we are doing? [J].
Blackwell, AF .
IEEE SYMPOSIUM ON VISUAL LANGUAGES, PROCEEDINGS, 1996, :240-246
[6]  
BLACKWELL AF, 1996, P 8 ANN WORKSH PSYCH, P105
[7]  
Burnett M. M., 1998, ACM Transactions on Computer-Human Interaction, V5, P1, DOI 10.1145/274444.274445
[8]   INTERACTIVE VISUAL DATA ABSTRACTION IN A DECLARATIVE VISUAL PROGRAMMING LANGUAGE [J].
BURNETT, MM ;
AMBLER, AL .
JOURNAL OF VISUAL LANGUAGES AND COMPUTING, 1994, 5 (01) :29-60
[9]  
BURNETT MM, 1998, 1998 IEEE S VIS LANG
[10]  
Card S. K., 1983, PSYCHOL HUMAN COMPUT