DMCA
Why HyperTalk Debugging is More Painful than it Ought To Be (1993)
BibTeX
@MISC{Kleiman93whyhypertalk,
author = {Ruben Kleiman and Marc Eisenstadt},
title = {Why HyperTalk Debugging is More Painful than it Ought To Be},
year = {1993}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
As part of a series of investigations on the nature of programming and debugging environments, this paper looks in detail at what it's like to work with an apparently `modern' and `friendly' environment: HyperCard. I kept a detailed diary of several lengthy debugging sessions, and then analysed the problems and difficulties I experienced. Eight fundamental problems were observed in the use of HyperTalk's debugging facilities: indirect access to troublesome source code; disruptive intermediate actions required; poor interpreter access during breaks; poor monitoring of built-in functions; no coarse-grained view of execution; no data flow analysis; no control flow analysis; deceptive view of inner states. The paper discusses the broader implications of these eight problems, as well as possible ways to alleviate the problems. Background and motivation Despite the large number of psychological studies of computer programming and debugging over the past two decades (e.g. Sime, et al., 1973;...
Keyphrases
hypertalk debugging source code several lengthy debugging session built-in function disruptive intermediate action deceptive view psychological study coarse-grained view detailed diary inner state computer programming control flow analysis eight fundamental problem large number debugging environment poor interpreter access possible way data flow analysis poor monitoring indirect access
