Course 13 – 26th May
Previous courses…
◦ Testing
◦ SOLID Refactoring
Program Quality
Metrics
Copyright
Rigidity - the tendency for software to be difficult to change, even in simple ways
Fragility - the tendency of the software to break in many places every time it is changed
Immobility - the inability to reuse software from other projects or from parts of the same project
Viscosity - it is easy to do the wrong thing, but hard to do the right thing
Cohesion
◦ The Release Reuse Equivalency Principle
◦ The Common Closure Principle
◦ The Common Reuse Principle
Coupling
◦ Acyclic Dependencies Principle
◦ The Stable Dependencies Principle
◦ The Stable Abstractions Principle
How do we measure the quality of an item?
◦ Construction quality (how well it is built, whether the raw material has flaws, etc ...)
◦ Design quality (comfort, elegance...)
◦ A combination between quality of design and construction (sturdiness...)
In general, we can say that chair A is better than chair B regarding some particular aspect, but it is usually difficult to say by how much.
We do not assess construction quality (=>
unique among engineering applications)
All quality attributes refer to design.
Esthetic qualities:
◦ Software is mostly invisible, and esthetics only matter for the visible elements
◦ Apart from the GUI, observable aspects software are:
Notations for design and writing of code
Behavior of software when interacting with other
When discussing software quality we must:
◦ Define those attributes of quality that are of interest;
◦ Determine a way of measuring those attributes;
◦ Find a way of representing design;
◦ Write specifications that will guide developers (following and implementing design qualities).
Code that implements a given design is a representation of that design.
Performing quality assurance after writing code is expensive and possibly useless.
Usually, only the manner in which the code is written is taken into account (coding style, design patterns, adaptability, maintenance, reuse (coupling, cohesion), security)
Measures the appropriateness of software to the environment it is used in.
Variuos aspecte taken into account are:
◦ The software is running;
◦ The software performs according to specifications;
◦ The software is safe;
◦ The software can be adapted as requirements change.
Safety
Efficiency
Maintenance Usability
Is the software complete, correct and robust?
◦ Completeness – works for all possible inputs;
◦ Consistency – always behaves as expected;
◦ Robustness – behaves well in abnormal situations (eg. Lack of resources, lack of internet connection, etc.)
The software makes efficient use of available resources (CPU time, network connection, etc.)
Efficiency is always less important than safety. It is
easier to make safe software efficient than the reverse
How easily can the design be changed or adapted?
Types of maintenance:
◦ Corrective: error fixing;
◦ Perfective: adding features that should have been part of the product;
◦ Adaptive: updating software as requirements change.
How easily can the software be taught and used?
Simplicity
Modularity
The reverse of complexity.
Aspects of complexity:
◦ Control flow: counts all the possible execution paths for a program
◦ Information flow: measures amount of data transmitted within the program
◦ Understanding: counts the number of identifiers and operators
Can be measured by examining :
◦ Cohesion: how well the components of a module collaborate.
◦ Coupling: interaction between modules
We use metrics to
◦ understand
◦ control
◦ predict
Size of software
Complexity of software
Robustness of software
Amount of time required to develop some software
Resource allocation for development
Productivity of effort
Development costs
Intuitively, estimation seem subjective
◦ To inexperienced persons, it looks like predicting the future
◦ This is reinforced when estimation is incorrect and projects are delivered late
Formal estimation processes
◦ allows the project team to reach a consensus on the estimates
◦ improve the accuracy
Successful estimations take into account the following
◦ Work Breakdown Structure (WBS) – what are the tasks that need to be performed to finish the product?
◦ Assumptions – how to deal with incomplete information
◦ Trust – if stakeholders and engineers trust each other, the estimate will be more accurate
A list of tasks that, if completed, will produce the final product
◦ Broken down by feature
◦ By project phase (requirements tasks, design tasks, programming tasks, QA tasks, etc.)
◦ Some combination of the two
◦ Should reflect the way previous projects have been developed
A project should be broken down into 10 – 20 tasks
◦ Regardless of the size of the project
◦ For large projects (e.g. an operating system), the tasks are large
◦ For smaller projects, the tasks are correspondingly smaller
Create an estimate for the cost of each task
◦ Most accurate estimates are those that rely on prior experience
At the beginning of the development team members do not have all the information
◦ Assumptions are needed to fill in missing things
◦ Assumptions can also be placeholders which will be corrected later
◦ If an assumption is proven incorrect, the timeline of the project MUST be adjusted
For effective estimates, assumptions need to be written down
◦ If not, the team will need to have the same discussion again
Estimates can either be a source of trust or distrust between the project team and
managers.
Stakeholders need the project completed but usually do not have software engineering
experience
Project managers must take care to make the estimation process as open and honest as
possible
It is common for nontechnical people to assume that programmers pad their
estimates
◦ They have a “rule” by which they cut off a third or half of any estimate
◦ This lack of trust causes engineers to automatically pad their estimates
An important part of running successful software projects is reaching a common understanding between the engineers, managers, and stakeholders.
KLOC: Kilo Lines Of Code
Effort, PM: Person – Month
Boehm 1995
Takes into consideration high level development tools and techniques
◦ Prototyping
◦ Modular development
◦ 4GL (fourth generation language)
Allows for estimates from the very first stages of development
Effort required to create a prototype of the application
Based on the Number of Object Points (NOP)
Formula for computing effort:
Investigate the screens and dialogs that are needed
◦ Simple: 1
◦ Complex: 2
◦ Very complex: 3
Reports that need generated
◦ Simple: 2
◦ Complex: 5
◦ Very complex: 8
Each lower level module (eg. 3GL): 10
The sum of all of the above represents the NOP.
Estimate the total lines of code (ESLOC)
Takes into account
◦ Requirements instability
◦ Possibilities of code reuse
Product attributes
◦ Safety, module complexity, size of user manual, size of the required database, amount of reusable
components
Platform attributes
◦ Constraints referring to execution time; platform volatility, memory constraints
Personnel attributes
◦ Analyst experience; developer experience; personnel continuity; knowledge of the domain of the problem to be solved with regards to analysts and developers;
knowledge of the programming language and development tools
Project attributes
◦ Required tools; distance between development teams (eg. different countries) and quality of communication;
A planning method from Extreme Programming (XP)
A method used to manage the negotiation
between the engineering team (Development) and the stakeholders (Business)
◦ Treats the planning process as a game
◦ playing pieces are “user stories” written on cards
◦ the goal is to assign value to stories and put them into production over time
Unlike other planning methodologies, it does not require a documented description of the scope of the project to be estimated
The Planning Game combines
◦ estimation
◦ identifying the scope of the project
◦ Identifying the tasks required to complete the software
The planning process is highly iterative. Each iterations looks like this:
◦ Scope is established by having Development and Business work together to interactively write the stories.
◦ Each story is given an estimate of 1, 2, or 3 weeks.
Larger stories are split into multiple iterations
◦ Business is given an opportunity to steer the project between iterations.
◦ The estimates are created by the programmers, based on the stories that are created.
◦ Commitments are agreed upon
20 PM. Are the following correct?
◦ 20 people working 1 month
◦ 4 people working 5 months
◦ 1 person working 20 months
Individual productivity decreases as team size increases
◦ Communication overhead
◦ On adding new members, productivity decreases initially
Adding people to a team behind of schedule
For a team with P members, one can have between P-1 and P(P-1)/2 communication channels
Each channel is a decrease in efficiency
We have 12 months to finish the job, so it will take 12 months.
A competitor asked for $1.000.000. We wil ask for $900.000.
The client budget is $500.000. That will be the exact cost of development.
Development takes 1 year, but we say it will take 10 months. A delay of 2 months is not
Lack of accuracy
Employee pushback
Use for other purposes than intended
Animosity within the development team
The rights enjoyed by authors with regards to their work;
Copyright is the instrument of protection of authors and their work;
Copyright gives the creator of an original work exclusive right for a certain time period in
relation to that work, including its publication,
distribution and adaptation; after which time the
work is said to enter the public domain.
Several exclusive rights typically attach to the holder of a copyright:
◦ to produce copies or reproductions of the work and to sell those copies (mechanical rights; including,
sometimes, electronic copies: distribution rights)
◦ to create derivative works (works that adapt the original work)
◦ to perform or display the work publicly (performance rights)
◦ to sell or assign these rights to others
◦ to transmit or display by radio or video (broadcasting rights)
Extreme Programming Explained by Kent Beck (Addison Wesley, 2000)
Applied Software Project Management, by Andrew Stellman and Jennifer Greene
(O’Reilly, 2006)
Bug Life Cycle: http://www.buzzle.com/editorials/4- 6-2005-68177.asp,
http://qastation.wordpress.com/2008/06/13/process -for-bug-life-cycle/
COCOMO: http://en.wikipedia.org/wiki/COCOMO
Curs 12, Ovidiu si Adriana Gheorghies:
http://www.info.uaic.ro/~ogh/files/ip/curs-12.pdf