Lecture 1 - introduction to software metrics.pptx

Software Metrics
Dr. Odongo Steven Eyobu
sodongo@cis.mak.ac.ug

Content
 Software Metrics
 Expectations
 Organization
 Reference
2

Organization
 Class representative + contact
 Enroll for course on MUELE
(http://muele.mak.ac.ug/), enrollment closes after two
weeks
 Communication: use email
 Compulsory lecture attendance: sign attendance
register at every lecture
 No use of mobile phones during class - either
SWITCH OFF or put on SILENCE
3

Organization cont’d
 Assessments
 Course work (Assignments, tests, class
discussions, participation and attendance) - 40%
 Final exam (60%)
 Groups of 5 for course assignments
4

Reference
 Software Metrics: A Rigorous and Practical Approach,
(2nd ed.) (638p.), N.E. Fenton and S.L. Peeger, PWS
Publishing, 1998. ISBN 0-534-95425-1.
 Metrics and Models in Software Quality Engineering,
Stephen H. Kan, 2nd ed. (560 p.),Addison-Wesley
Professional (2002). ISBN: 0201729156.
 Software Engineering: A Practitioners Approach, (5th
ed.), Roger S.Pressman
 Web/Internet, some links may be posted on MUELE
5

Teaser
 Why doesn't your weight increase when you eat a kg of food?
 Why does a 1GB ash thats full weigh the same as an empty
one?
6

Reference for Today’s lecture
 Material in this lecture is taken from chapters
1-3 of Software Metrics: A Rigorous and
Practical Approach (2nd
ed.), Norman E.
Fenton and Shari Lawrence Pfleeger, 1997,
PWS Publishing Company, Boston, MA, ISBN
0534954251
7

Overview
1. Measurement – what is it and why do we do
it?
2. Measurement basics
3. A goal-based software measurement
framework
8

Measurement – What Is It and
Why Do We Do It?
1. Measurement in Everyday Life
2. Measurement in Software Engineering
3. The Scope of Software Metrics
9

Measurement in Everyday Life
 Measurement governs many aspects of
everyday life:
 Economic indicators determine prices, pay raises
 Medical system measurements enable diagnosis
of specific illnesses
 Measurements in atmospheric systems are the
basis of weather prediction
10

How do we use measurement in
our lives?
 In a shop, price is a measure of the value of an item,
and we calculate the bill to make sure we get the
correct change.
 Height and size measurements ensure clothing will fit
correctly.
 When traveling, we calculate distance, choose a
route, measure speed, and predict when we’ll arrive
11

Measurement helps us to:
 Measurement helps us to:
 Understand our world
 Interact with our surroundings
 Improve our lives
 Examples
 Economics
 Radar
 Medical
 Weather
 Without measurements, technology cant
function 12

What is Measurement?
 some aspect of a thing is assigned a
descriptor that allows us to compare it with
other things.
 More formally – the process by which
 Numbers or symbols are assigned to attributes of
entities in the real world in such a way as to
describe them according to defined rules.
13

What is Measurement?
 Measure is a relation between an attribute
and a measurement scale
 Entity: object, event in real world
 Attribute: feature, property of entity
 We measure attributes of things
14

Entity in Software Engineering
15
Can be
any of the
following

Attribute
 An Attribute is a feature or property of an entity
 E.g. Blood pressure of a person, duration of the software
specification process
 Two general types
 Internal Attributes, can be measures based on the enity itself
• E.g code, coupling, modulariy, internal attribute
 External Attributes, can be measures only with respect to
how the entities relates to the environment
• Eg , reliability, maintainability
16

Software Metrics
 It refers to a broad range of quantitative
measurements for computer software that
enable to
 improve the software process continuously
 assist in quality control and productivity
 assess the quality of technical products
 assist in tactical decision-making
18

19
What Should Be Measured?
measurement
What do we
use as a
basis?
• size?
• function?
project metrics
process metrics
process
product
product metrics

Measurement Types
 Measurements are needed as:
 Descriptors of entities already in existence
 Prescriptors (standards, norms, failure intensity
objects, benchmarks) which entities of certain
class of category should satisfy
 Predictors to estimate properties of entries yet to
be designed or implemented
20

Measurements: How to
 In order to make entities measurable
 What entities (objects) should be selected?
 What attributes should be selected?
 What values should be assigned to the attributes?
 What shall be the rules (relationships) ascribed to
the attributes and their entiries
 NOTE: assigned values and /or ascribed
rules can be quantitative of qualitative
21

quantitative vs qualitative
22

Understanding Measurement?
 we must ask questions that are difficult to answer:
 In a room with blue walls, is “blue” a measure of the color of the
room?
 A person’s height is a commonly understood attribute that can be
easily measured. What other attributes, such as intelligence?
 Some measurements (e.g., intelligence, wine quality) may have wide
error margins – is this a reason to reject them?
 How do we decide which error margins are acceptable?
 When is a measurement scale acceptable for the purpose to which it
is put (e.g., is it appropriate to measure a person’s height in
kilometers)?
 What types of manipulations can we apply to the results of
measurement?
23

Example 1: Code
 Entiry Code:
 What are the attributes?
 What are the possible measures?
24

Example 2:Availability
 Entity: Availability
 Attributes: ?
 Values: ?
 Relations: ? How are the attributes related?
26

Example 2:Availability
 Entity: availability
 Attributes, system uptime, system downtime
 Values: time is seconds
 Relations:
 Availability = uptimes/ (uptime + downtime)
27

Misleading Metrics!
 Fact (1): Knowledge is power
 Fact (2): Time is money
 Relation (rule): power = work/time
 Substituting ‘power’ & ‘time”
 Knowledge = work / money
 As knowledge approaches zero money approaches
infinity regardless of the amount for work done
 Conclusion
 The less you know, the more you make.
29
What went
wrong here!?
- Needs
validation

Making Things Measurable
 One aim of science is to find ways of measuring attributes of
things we’re interested in.
 Measurement makes concepts more visible, therefore more
understandable and controllable.
 Attributes previously thought to be unmeasurable now form
basis for decisions affecting our lives (e.g., air quality, inflation
index).
 Measuring the unmeasurable improves
understanding of particular entities, attributes
• Act of proposing a particular measure can open discussion
that will lead to greater understanding
• Making new measurement may requiring modifying
environment or practices (e.g., using a new tool, adding a step
in a process)
30
“What is not measurable, make measurable” (Galileo Galilei)

What is Software Measurement
 Software metrics are measures that are used
to quantify software, software development
resource and/or the software development
process
 This includes items which are directly
measureable, such as lines of code, as well as
item which are calculated from measurements,
such as software quality
 Involves selecting, measuring & putting together
many diff attributes..
• Not a trivial task, 300+ metrics have been defined
31

Measurement in SE
 In many instances, measurement is considered a
luxury.
 For many projects:
 Measurable targets are not set (e.g., products are supposed
to be user-friendly, reliable, and maintainable, but we don’t
quantify what that means).
 The component costs of projects are not quantified or
understood.
 Product quality is not quantified.
 Too much reliance on anecdotal evidence (e.g., try our
product and you’ll improve your productivity by 50%!). Most
of the time, there’s no measurable basis for the claims.
33

.. the effect
 According to American Programmer, 31.1% of
computer software projects get canceled
before they are completed,
 52.7% will overrun their initial cost estimates
by 189%.
 94% of project start-ups are restarts of
previously failed projects.
 Solution?
 systematic approach to software development and
measurement
34

Measurement in SE (cont’d)
 When measurements are made, they tend to be:
 Incomplete
 Inconsistent
 Infrequent
 Most of the time, we’re not told anything about:
 How experiments were designed
 What was measured and how
 Realistic error margins
 Without this information, can’t decide whether to apply
results to a development effort, and can’t do an objective
study to repeat the measurements.
 Lack of measurement in SW engineering is
compounded by lack of a rigorous approach.
35

Software Measurement Objectives
 Assessing status
 Projects
 Products for a specific project or projects
 Processes
 Resources
 Identifying trends
 Need to be able to differentiate between a healthy project
and one that’s in trouble
 Determine corrective action
 Measurements should indicate the appropriate corrective
action, if any is required.
36

Info to understand…
 Managers
 What does the process cost?
 How productive is the staff?
 How good is the code?
 Will the customer/user be satisfied?
 How can we improve?
 Engineers
 Are the requirements testable?
 Have all the faults been found?
 Have the product or process goals been met?
 What will happen in the future?
 Users??
37

Example – why do we need
measurement?
38

Lecture 1 - introduction to software metrics.pptx

More Related Content

Similar to Lecture 1 - introduction to software metrics.pptx (20)

Recently uploaded (20)

Lecture 1 - introduction to software metrics.pptx

Editor's Notes