Software development Flashcards
1
Q
What are the 5 stages of software development in the Software Development Life Cycle?
A
- Analysis
- Design
- Implementation
- Evaluation
- Maintenance
2
Q
What is the analysis stage of the SDLC?
A
- this is where the problem to be solved is defined + system requirements
- for any problem takes into consideration: data to be used, procedures, problems with current solution and any future plans
- documentation produced = systems specification or user requirements - outlines what the system will do but not how to make it
- info gathered using interviews, questionnaires, observation at current system + looking at existing documentation
3
Q
what is the feasibility study?
A
- part of analysis stage
- assessment of whether the project is practical and makes sense to do
- considers: is there appropriate tech, can it be operated + maintained easily after production, can it be financed, is it legal, does it have a market, can it be produced reasonably quick
- useful as it identifies any issues and resolves them or drops the project completely if it isn’t deemed worth it
4
Q
what is requirements analysis?
A
- part of analysis stage
- outlines the requirements of the project
- gathered through interviews with users, questionnaires, observation of current system usage + looking at current system documentation
- important because it clearly and unambiguously defines what the user wants the software to do and creates a specification
- analyse strengths and weaknesses in the current system + consider inputs, outputs, stored data and amount of data
5
Q
what is the design stage of the SDLC?
A
- it’s when info gathered during the analysis stage is translated into a blueprint for programmers to follow during implementation
- test plan is formed here
- key aspects: description, database format, inputs, outputs, security features, hardware features and user interface
- features high level and low level design
6
Q
what is high level design?
A
- system architecture is defined
- components (modules) of the system are identified + relationships between them established
- descriptions of hardware + software interfaces + UI
7
Q
what is low level design?
A
- detailed guide formed after high level design
- much more comprehensive - covers specifics of each module + goes through each one’s logic in detail
- addresses different algorithms, data structures + interfaces
- ensures accurate + efficient implementation of required functionality
8
Q
what is the implementation stage of the SDLC?
A
- when code for the software is written, it is tested and installed on user’s machines
- can be very long or short depending on methodology used - complete when all code has been written to achieve required functionality
- problem is divided into modules that perform an individual task and assigned to a team member to code in the appropriate programming language
- code will continue to be edited until testing is complete
9
Q
what are some testing strategies?
A
- black box testing
- white box testing
- alpha testing
- beta testing
10
Q
what is black box testing?
A
- carried out independently of code (not using internal structure)
- produce and test a set of test data that covers all potential input and program functions based on specification and checks that outputs are valid
- done by company or end-users from user perspective
- good to check all requirements are satisfied + non-specialist can use it
11
Q
what is white box testing?
A
- test data is derived from source code (focus on structure not function)
- tests that program logic works throughout and makes sure that every possible path through the code is tested at least once
- good to make sure program works and is at peak efficiency
- doesn’t help find missing functions - you can only test what has been written
12
Q
what is alpha testing?
A
- done by in-house software development team + user
- reveals errors + omissions in system requirements definition or misunderstandings of requirements by the developer
- software might do something the user doesn’t want it to do
- aims to find and fix unspotted bugs before being released to the public
13
Q
what is beta testing?
A
- used when commercial software is being tested (eg. Windows, Word, etc)
- testing an unreleased program with a teeny bit of the intended audience - users agree to report any issues to developers
- good because it exposes product to real world and find bugs programmers didn’t find
- afterwards product is modified + further beta testing until satisfied
- done towards the end of testing period
14
Q
what is deployment?
A
- part of implementation stage
- done once other testing is complete to a satisfactory level
- product is installed on user systems + more testing is done
- any new issues are corrected
- may be done in a graduated manner so not all users face issues simultaneously
- do this when all target users get access to the product
15
Q
what is the evaluation stage of the SDLC?
A
- aka acceptance testing
- often involves post-implementation review - 6 months after product release
- delay between deployment + evaluation to allow users to get used to the product
- this and month end reviews allow the chance to see whether the product has fulfilled original requirements + identifies any shortcomings
- product can be adjusted if any issues found
- solution should be evaluated based on effectiveness, usability and maintainability
16
Q
What are the 3 types of maintenance?
A
- corrective maintenance
- adaptive maintenance
- perfective maintenance
17
Q
What is corrective maintenance?
A
This is when any bugs found after software is released to the general public have to be patched - this will always happen no matter how much testing is done
18
Q
What is adaptive maintenance?
A
This is when software is adapted to adjust to changing user requirements, for example as a result of development of new technology
19
Q
What is perfective maintenance?
A
This is when software is improved in non-essential ways even if the software still works - for example by improving functionality, UI, etc
20
Q
What is the waterfall software development methodology?
A
- one of the oldest models - not quite redundant but nearly there
- mainly used for small projects where requirements are well defined and there is no changing technology
- follows all the steps of the SDLC but only moves onto the next stage once the previous stage is done and documentation is produced
- it is possible to go back to a previous stage if needed but you can’t skip any
21
Q
What are the advantages of the waterfall model?
A
- it’s simple to understand and straightforward to manage the project
- self contained stages mean team members clearly know what to do
22
Q
What are the disadvantages of the waterfall model?
A
- time consuming + expensive to go back and fix an issue
- not much client input beyond analysis stage so they may end up with a product that doesn’t satisfy their requirements
- no working software produced until late into the cycle
23
Q
What is the spiral model?
A
- made up of 4 basic steps: analyse, design, implement, evaluate
- repeatedly go through these steps and produce a prototype at the end of each iteration
- evaluating each prototype helps produce a plan for the next prototype
- requirements are defined at the start and each successive loop around the spiral produces a more refined product until it’s complete
- mainly used for medium to high risk objects with evolving or complex requirements and for long term projects
24
Q
Advantages of the spiral model?
A
- well defined steps make it easy to manage
- software is produced at an early stage - issues identified quickly + easily
- user feedbacks fairly often - changes made early on + end result is closer to what they expected
25
Disadvantages of the spiral model?
- time consuming + expensive to do each iteration and produce a prototype
- not good for small projects
- needs a large team
26
What is the agile software development methodology?
- often used for smaller projects as a faster more efficient variation of the spiral model
- software is developed in rapid incremental cycles known as sprints
- each version builds on the previous version and is thoroughly tested
- doesn't need much planning to get started - lack of emphasis on documentation since client is heavily involved
- best used for small projects that are part of a bigger project or in developing business where customer needs are constantly changing
- 4 stages: plan, develop, test, demonstrate
27
What are the advantages of the agile model?
- rapid continuous delivery of useful software + lots of engagement between customers, developer and testers = high customer satisfaction
- much quicker than other methodologies - software is produced in weeks, not months
- high customer involvement = software produced makes the customer happy
28
what are the disadvantages of the agile model?
- requires an experienced team of programmers (novices must be mentored)
- lack of emphasis on documentation can become an issue
- customer dissatisfaction if they haven't been clear about requirements
- can't incorporate features that would be good for later because the focus is on what is needed now
29
what is extreme programming software development methodology?
- type of agile software development but working in even shorter cycles where working code "releases" are made rapidly and the user feedbacks on it
- the speed aims to improve productivity + quality + introduce checkpoints here new/changed customer requirements can be introduced
- 5 values: courage, simplicity, communication, feedback, respect
- 5 rules: planning, managing, designing, coding, testing (emphasis on coding stage)
- you need a really small team of up to 12 people for this to work
- best for large projects with a short deadline and changing requirements
30
what are the advantages of extreme programming model?
- saves money + time on project realisation - focus on delivering products, not producing documentation
- simpler methodology - code can be improved at any point
- constant feedback from customer - they get a product they are happy with quicker
31
what are the disadvantages of extreme programming model?
- lack of focus on design + documentation may lead to issues with code + mistakes being repeated in the future
- doesn't really work if programmers are separated geographically
32
what is rapid application development (RAD) software development methodology?
- short development cycles with a focus on making working software and a prototype more refined than the previous one (refined based on user feedback)
- workshops and focus groups gather requirements - no formal requirement document
- each part of the system will be produced in a short time frame - not perfect, but good enough
- reuse any software components possible
- best used for long projects with both user and technology requirement changes - it is much easier to incorporate changes at a late stage with this methodology
33
advantages of RAD model?
- cheaper - fewer developers needed
- faster - reuseable components of code
higher customer satisfaction - their feedback is incorporated
- shorter development cycles make it easier to incorporate changes
34
disadvantages of RAD model?
- uses powerful tools to speed up development process - needs skilled professionals + too expensive for small projects
- if there are no reusable components to e project could fail
- if the project can't be modularised properly RAD can't be used
35
what could go wrong during a long project?
- personnel change
- requirements change
- technology changes
- costs increase
- project could get cancelled
36
What is an algorithm?
A set of instructions to solve a problem or complete some well defined task in a finite number of steps
Eg. making a cup of tea
37
what are the properties of a good algorithm?
- clear and precise steps that produce the correct outputs for any valid input
- should allow for invalid inputs
- must terminate at some point
- execute efficiently in as few steps as possible
- should be designed in such a way that other people would be able to understand it and modify it if necessary
38
what are some examples of computational algorithms?
- Internet related algorithms - to manage and manipulate all the data stored on the internet
- route-finding algorithms - eg. to find the best route to transmit packets of data from A to B over a network when there are 1000s of options
- compression algorithms - used to compress data files so they can be transmitted faster or held in a smaller amount of storage space - eg. MP3 files
- encryption algorithms - you have to encrypts someone's personal details when they do online shopping to make sure it isn't intercepted or read because it's sensitive data
39
What is a trace table?
- it is used to dry run a program and see how the algorithm flows
- the table has columns for each variable + processes + output
- as you go through the algorithm in the way the computer does, note down in the trace table what values change and how
