Please enable JavaScript.
Coggle requires JavaScript to display documents.
How long do Junior Developers take to Remove Technical Debt Items? -…
How long do Junior Developers take to Remove Technical Debt Items?
INTRODUCTION
Engineering is the field with
highest inflow of junior developers.
Automated static analysis tools
(ASAT)—are valuable to junior developers to improve their coding skills
with minimal consultations with senior colleagues
Analysis of the diffuseness of
TD items introduced by junior developers
Identification of the
type of TD items commonly refactored by junior developers
Comparison of the actual undestimated
TD Items refactoring
time
THE CASE STUDY
Guidelines
by Runeson
and Höst 2009
GOAL
What TD items are
introduced by junior developers
?
Which TD items are
commonly refactored by junior developers
?
How long
junior developers
spend to refactor TD items
?
What is the remediation
time accuracy
?
STUDY
DESIGN
Multiple case study
with a replication design
REPLICATION 1
133 second-year master students
at Tampere University.
REPLICATION 2
52 fourth-year bachelor students
at the University of Novi Sad.
CASE
SELECTION
teams
[ 4-8 members]
software project
for
three months
After 1.5 months,
SonarQube to detect TD items.
Fix TD items detected
by SonarQube
SUBJECT
SELECTION
experience
in software development.
at least two years of full-time
industrial experience
different countries and with different instructors was
to increase the generality of the results
DATA
COLLECTION
same version
of SonarQube
Collect data
using the same protocol
development
phase
completed
resolve as many
issues
as they can
report
effort spent
and
status of issue
fixes
SonarQube analysis
is performed to validate fixes
DATA
ANALYSIS
RE (Relative Error)
MMRE
(Mean Magnitude Relative Error)
MdMRE
(Median Magnitude Relative Error)
THREATS TO
VALIDITY
Construct Validity
default set of collected measures
considered by the sonarQube
Internal Validity
filtered data
Some issues detected
by SonarQube were
duplicated
External Validity
large number of
heterogeneous projects
RESULTS
Jr DEVs
add more
code smells
refactor more
code smells
focus equally on all TD items
,
independently from
the severity and type
spend no
more than 50%
of the SonarQube
estimated time
several cases
remediation time is
20 times lower
than the estimated
RELATED
WORK
The main works are related to
defining approaches that quantify TD
in terms of cost to fix
technical issues
remediation time is generally overestimated
by the tool as compared to the actual time for remediating TD items
the most
accurate estimations relate to code smells
, while the least accurate concern bugs.
CONCLUSION
Devs consistently introduce
TD items of
different types and severity
junior developers address them all
,
without considering the
severities and types.
very low remediation time
to remove TD items.
FUTURE
WORKS
replicate this study
non-academic context
Different
project
Context
adopt SonarQube
because of its support in
reducing technical debt
and they might help Jr Devs
to
refactor and clean the code efficiently
.