2004
DOI: 10.1007/978-3-540-27777-4_7
|View full text |Cite
|
Sign up to set email alerts
|

Suitability of FIT User Acceptance Tests for Specifying Functional Requirements: Developer Perspective

Abstract: Abstract. The paper outlines an experiment conducted in two different academic environments, in which FIT tests were used as a functional requirements specification. Common challenges for functional requirements specifications are identified, and a comparison is made between how well prose and FIT user acceptance tests are suited to overcoming these challenges from the developer's perspective. Experimental data and participant feedback are examined to evaluate whether developers can use requirements in the for… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

3
39
0

Year Published

2005
2005
2023
2023

Publication Types

Select...
3
3
2

Relationship

2
6

Authors

Journals

citations
Cited by 24 publications
(42 citation statements)
references
References 2 publications
3
39
0
Order By: Relevance
“…Although users can specify the test case tables, developers must later implement fixtures (lightweight classes calling business logic) that allow these tables to be executed. Suitability of acceptance tests for specifying functional requirements has been closely examined in our previous paper [4]. Our hypothesis that tests describing customer requirements can be easily understood and implemented by a developer who has little background on this framework was substantiated by the evidence gathered in our previous experiment.…”
Section: Introductionmentioning
confidence: 74%
“…Although users can specify the test case tables, developers must later implement fixtures (lightweight classes calling business logic) that allow these tables to be executed. Suitability of acceptance tests for specifying functional requirements has been closely examined in our previous paper [4]. Our hypothesis that tests describing customer requirements can be easily understood and implemented by a developer who has little background on this framework was substantiated by the evidence gathered in our previous experiment.…”
Section: Introductionmentioning
confidence: 74%
“…Training usually only covers the basics of the tool, so any advanced use of Fit is bound to be gained by continuous use. Melnik et al reports on a university course where 90% of the students managed to deliver their Fit test assignments after a brief introduction, indicating that the students found it easy to learn [10]. These students were however computer graduates.…”
Section: Learning Fitmentioning
confidence: 99%
“…[Ibid] This is potentially related to effects described by Melnik et al, where they claim Fit reduces noise, over-specification, and ambiguity [10].…”
Section: Using Fit To Create Requirements Specificationsmentioning
confidence: 99%
“…A well-defined test suite was provided by the customer (instructor) up front. In a separate experiment designed to evaluate the suitability of using FIT for specifying functional requirements for the developers, we have found that these tests can be easily understood, interpreted and implemented by developers [11].…”
Section: "I Think the Test Code Is More A Part Of Design Then It Is Jmentioning
confidence: 99%