Software Engineering Web

Software Engineering Web

1. What is software?

   • Instructions that when executed provide desired features, function, and performance
   • Data Structures that enable the programs to adequately manipulate information
   • Documentation that describes the operation and use of the programs (or sets of instructions)
   • Software Application Domains (or types of software)
     • System Software
       • operating systems, networking software, compilers etc.
     • Application Software
       • stand-alone programs for specific business need that facilitates business operations
         • something used by a business
     • Engineering/Scientific Software
       • Software used in scientific work. Characterized by number crunching and/or simulation
     • Embedded Software
       • software built into some physical device. Controls functions for end user and system. Ex) microwave, car
     • Product-line software
       • software that satisfies the needs of a particular market. Ex) word processing, databases
     • Web Applications
       • any application hosted on the internet.
         • Characterized by hypertext, graphics, interactive content, database, and business application access
     • Artificial Intelligence Software
       • characterized by non-numerical algorithms to solve complex tasks. Think robotics and pattern recognition
     • Open World Computing
       • pervasive, distributed computing that connects mobile devices, PCs, and enterprise systems via vast networks
     • Netsourcing
       • WWW as computing engine and content provider. Think search engine
     • Open Source Software
       • distribution of source code for systems applications, self descriptive with high quality change management
     • Legacy Software
       • developed a long time and continuously modified to meet changes in business requirements and computing platforms
         • can be costly to maintain an older systems but also very risky to evolve to a new one

2. Defintion of Software engineering

   • Fritz Bauer (1969)
     • is the establishment and use of sound engineering principles in order to obtain economical software that is reliable and works efficiently on real machines
   • IEEE (1993)
     • the application pf a systematic, disciplined, quantifiable approach to the development, operation, and maintenance of software. That is the application of engineering principles to software and the study of these principles.

3. Software Process

   • a collection of activities, actions, and tasks that are performed when a work product is to be created. AKA who is doing what and when and how to reach a certain goal.
     • an activity achieves a broader objective (a set of actions)
     • an action is a set of tasks for a work product
     • a task focuses on the completion of a small well-defined objective
   • Process Framework/Model
     • Framework activities (engineering activities) (many framework activities within one umbrella activity)
       • communication, planning, modeling, construction, deployment
       • task set within each action within one framework activity is decided by the software team based on the problem
       • Generic Process Flows
     • Umbrella Activities (management activities) many within one process framework
       • software project management, risk management, software quality assurance, formal technical reviews, measurement, software configuration management, reusability management, work product preparation and production
     • each the activities within a process resides in a framework or model that defines their relationship with the process and each other
     • Categories of Process Models
       • Prescriptive Process Models
         • defined by a prescribed set of process elements and a predictable process flow
           • highly structured and facilitate coordination and cohesion in software work
           • the rigid structure may not be well suited for a software world generally defined by change
         • Waterfall Model
           • follows a linear process flow
           • ideal when project requirements are well understood and not likely to change
           • assumes all of the skills needed to build the project are available
           • disadvantage is that no working product until the end
         • Waterfall Model with Feedback
           • same basic linear process flow except after each step there are feedback loops to all of the previous steps
         • V- Model
           • analogous linear process flow. Each testing phase (unit testing, integration testing, system testing, acceptance testing) is tied to each of the previous linear framewor activities
           • again no working product until very end
       • Incremental Process Models
         • delivers a series of releases called increments
           • each increment provides progressively more functionality for the customer as each increment is delivered. Each increment adds a feature that is observable by the customer
         • The work for each increment can follow any process flow. However subsequent increments can be started before prior ones finish
           • For example, in a linear process flow during the "Construction" phase of increment #1, the "Communication" phase can occur for increment #2
       • Evolutionary Process Models
         • releases increments similar to the incremental process model, however each increment does not necessarily add a new "feature." The system simply becomes a more complete version of the system. The customer may not notice any difference between two increments as things under the hood may have changed.
         • satisfies the need to introduce limited versions to meet competitive or business pressures
         • accounts for the fact that business and product requirements often change during development. This model explicitly accomodates that evolution
         • Prototyping
           • Rapidly produce prototype to present customer with some idea of what the system might be like
             • this helps to mitigate risks early as customer gets a better idea of requirements
           • Disadvantage is that customer thinks the product is further along than it actual is. They are unaware of shortcuts and compromises under the hood
         • Spiral Model
           • with each spiral the system grows
           • realistic model for large scale systems
           • risk driven process model (uses prototyping) that is used multi stake holder software systems
         • Concurrent Models
           • parallelizes tasks and is ideal for projects involving multiple engineering teams from different disciplines.
             • think embedded software or low level/close to hardware development
         • Drawbacks
           • project planning is hard but its difficult to project the number of cycles needed
           • The speed at which the system evolves is not defined
           • still prioritizes high/quality which sacrifices some flexibility and extrnsibility
   • Process Pattern
     • describes a consistent method for describing problem solutions within the context of a software process
   • Process Assessment
     • attempts to understand the current state of the software process with the intent of improving it

4. General Principles

   • George Polya (1945) "How to solve it". AKA the essence of problem solving
     • Step 1: Understand the problem (communication and analysis)
     • Step 2: Plan the solution (modeling and design)
     • Step 3: Carry out the plan (generation)
     • Step 4: Examine the result for accuracy (quality assurance)
   • Hooker (1996)
     • KEEP IT SIMPLE STUPID

Bilder und Dateien der Mindmap