Software development process

From Wikipedia, the free encyclopedia

This needs additional citations for verification.
Please help improve this article by adding reliable references. Unsourced material may be challenged and removed. (June 2006)

A software development process is a structure imposed on the development of a software product. Synonyms include software life cycle and software process. There are several models for such processes, each describing approaches to a variety of tasks or activities that take place during the process.

Activities and steps
Software development process
Requirements · Architecture Design · Implementation Testing · Deployment
Models
Agile · Cleanroom · Iterative · RAD RUP · Spiral · Waterfall · XP · Scrum
Supporting disciplines
Configuration management Documentation Quality assurance (SQA) Project management User experience design
This box: view • talk • edit

1 Processes and meta-processes
- 1.1 Waterfall processes
- 1.2 Iterative processes
2 Process models
- 2.1 Formal methods
3 See also
4 References
5 External links

[edit] Processes and meta-processes

A growing body of software development organizations implement process methodologies. Many of them are in the defense industry, which in the U.S. requires a rating based on 'process models' to obtain contracts. The international standard for describing the method of selecting, implementing and monitoring the life cycle for software is ISO 12207.

The Capability Maturity Model (CMM) is one of the leading models. Independent assessments grade organizations on how well they follow their defined processes, not on the quality of those processes or the software produced. CMM is gradually replaced by CMMI. ISO 9000 describes standards for formally organizing processes with documentation.

ISO 15504, also known as Software Process Improvement Capability Determination (SPICE), is a "framework for the assessment of software processes". This standard is aimed at setting out a clear model for process comparison. SPICE is used much like CMM and CMMI. It models processes to manage, control, guide and monitor software development. This model is then used to measure what a development organization or project team actually does during software development. This information is analyzed to identify weaknesses and drive improvement. It also identifies strengths that can be continued or integrated into common practice for that organization or team.

Six Sigma is a methodology to manage process variations that uses data and statistical analysis to measure and improve a company's operational performance. It works by identifying and eliminating defects in manufacturing and service-related processes. The maximum permissible defects is 3.4 per one million opportunities. However, Six Sigma is manufacturing-oriented and needs further research on its relevance to software development.

Domain Analysis: Often the first step in attempting to design a new piece of software, whether it be an addition to an existing software, a new application, a new subsystem or a whole new system, is, what is generally referred to as "Domain Analysis". Assuming that the developers (including the analysts) are not sufficiently knowledgeable in the subject area of the new software, the first task is to investigate the so-called "domain" of the software. The more knowledgeable they are about the domain already, the less work required. Another objective of this work is to make the analysts, who will later try to elicit and gather the requirements from the area experts, speak with them in the domain's own terminology, facilitating a better understanding of what is being said by these experts. If the analyst does not use the proper terminology it is likely that they will not be taken seriously, thus this phase is an important prelude to extracting and gathering the requirements. If an analyst hasn't done the appropriate work confusion may ensue: "I know you believe you understood what you think I said, but I am not sure you realize what you heard is not what I meant."^[1]

Software Elements Analysis: The most important task in creating a software product is extracting the requirements. Customers typically have an abstract idea of what they want as an end result, but not what software should do. Incomplete, ambiguous, or even contradictory requirements are recognized by skilled and experienced software engineers at this point. Frequently demonstrating live code may help reduce the risk that the requirements are incorrect.

Requirements Analysis: Once the general requirements are gleaned from the client, an analysis of the scope of the development should be determined and clearly stated. This is often called a scope document. Certain functionality may be out of scope of the project as a function of cost or as a result of unclear requirements at the start of development. If the development is done externally, this document can be considered a legal document so that if there are ever disputes, any ambiguity of what was promised to the client can be clarified.

Specification: Specification is the task of precisely describing the software to be written, possibly in a rigorous way. In practice, most successful specifications are written to understand and fine-tune applications that were already well-developed, although safety-critical software systems are often carefully specified prior to application development. Specifications are most important for external interfaces that must remain stable. A good way to determine whether the specifications are sufficiently precise is to have a third party review the documents making sure that the requirements are logically sound.

Software architecture: The architecture of a software system refers to an abstract representation of that system. Architecture is concerned with making sure the software system will meet the requirements of the product, as well as ensuring that future requirements can be addressed. The architecture step also addresses interfaces between the software system and other software products, as well as the underlying hardware or the host operating system.

Implementation: This is the part of the process where software engineers actually program the code for the project.

Testing: Testing software is an integral and important part of the software development process. This part of the process ensures that bugs are recognized as early as possible

Deployment: After the code is appropriately tested, it is approved for release and sold or otherwise distributed into a production environment.

Documentation: Documenting the internal design of software for the purpose of future maintenance and enhancement is done throughout development. This may also include the authoring of an API, be it external or internal.

Software Training and Support: A large percentage of software projects fail because the developers fail to realize that it doesn't matter how much time and planning a development team puts into creating software if nobody in an organization ends up using it. People are often resistant to change and avoid venturing into an unfamiliar area, so as a part of the deployment phase, it is very important to have training classes for new clients of your software.

Maintenance: Maintaining and enhancing software to cope with newly discovered problems or new requirements can take far more time than the initial development of the software. It may be necessary to add code that does not fit the original design to correct an unforeseen problem or it may be that a customer is requesting more functionality and code can be added to accommodate their requests. It is during this phase that customer calls come in and you see whether your testing was extensive enough to uncover the problems before customers do.

[edit] Waterfall processes

Main article: Waterfall model

The best-known and oldest process is the waterfall model, where developers are to follow these steps in order:

Requirements specification (AKA Verification)
Design
Construction (AKA implementation or coding)
Integration
Testing and debugging (AKA validation)
Installation (AKA deployment)
Maintenance

After each step is finished, the process proceeds to the next step, just as builders don't revise the foundation of a house after the framing has been erected.

There is a misconception that the process has no provision for correcting errors in early steps (for example, in the requirements). In fact this is where the domain of requirements management comes in which includes change control.

This approach is used in high risk projects, particularly large defense contracts. The problems in waterfall do not arise from "immature engineering practices, particularly in requirements analysis and requirements management." Studies of the failure rate of the DOD-STD-2167 specification, which enforced waterfall, have shown that the more closely a project follows its process, specifically in up-front requirements gathering, the more likely the project is to release features that are not used in their current form^{[citation needed]}.

Often the supposed stages are part of joint review between customer and supplier, the supplier can, in fact, develop at risk and evolve the design but must sell off the design at a key milestone called Critical Design Review (CDR). This shifts engineering burdens from engineers to customers who may have other skills.

[edit] Iterative processes

Iterative development [1] prescribes the construction of initially small but even larger portions of a software project to help all those involved to uncover important issues early before problems or faulty assumptions can lead to disaster. Iterative processes are preferred by commercial developers because it allows a potential of reaching the design goals of a customer who does not know how to define what they want.

Agile software development processes are built on the foundation of iterative development. To that foundation they add a lighter, more people-centric viewpoint than traditional approaches. Agile processes use feedback, rather than planning, as their primary control mechanism. The feedback is driven by regular tests and releases of the evolving software.

Interestingly, surveys have shown the potential for significant efficiency gains over the waterfall method. For example, a survey, published in August 2006 by VersionOne and Agile Alliance and based on polling more than 700 companies claims the following benefits for an Agile approach.^[2] The survey was repeated in August 2007 with about 1,700 respondents.^[3]

Here are some results:

Accelerated time to market
Increased productivity:
- 2006: 10% or higher improvements reported by 87% of respondents
- 2007: 10% or higher improvements reported by 83% of respondents
- 2006: 25% or higher improvements reported by 55% of respondents
- 2007: 25% or higher improvements reported by 55% of respondents
Reduced software defects:
- 2006: 10% or higher improvements reported by 86% of respondents
- 2007: 10% or higher improvements reported by 85% of respondents
- 2006: 25% or higher improvements reported by 55% of respondents
- 2007: 25% or higher improvements reported by 54% of respondents
Reduced cost:
- 2006: 10% or higher improvements reported by 63% of respondents
- 2007: 10% or higher improvements reported by 28% of respondents
- 2006: 25% or higher improvements reported by 26% of respondents
- 2007: 25% or higher improvements reported by 28% of respondents

Other interesting improvements reported, among them:

Enhanced ability to manage changing priorities
Alignment between IT and business goals
Improved team moral
Reduced project risk

There is also an interesting chart at http://versionone.com/Resources/AgileBenefits.asp that shows Agile development value proposition in comparison to traditional development.

Extreme Programming (XP) is the best-known iterative process. In XP, the phases are carried out in extremely small (or "continuous") steps compared to the older, "batch" processes. The (intentionally incomplete) first pass through the steps might take a day or a week, rather than the months or years of each complete step in the Waterfall model. First, one writes automated tests, to provide concrete goals for development. Next is coding (by a pair of programmers), which is complete when all the tests pass, and the programmers can't think of any more tests that are needed. Design and architecture emerge out of refactoring, and come after coding. Design is done by the same people who do the coding. (Only the last feature - merging design and code - is common to all the other agile processes.) The incomplete but functional system is deployed or demonstrated for (some subset of) the users (at least one of which is on the development team). At this point, the practitioners start again on writing tests for the next most important part of the system.

While Iterative development approaches have their advantages, software architects are still faced with the challenge of creating a reliable foundation upon which to develop. Such a foundation often requires a fair amount of upfront analysis and prototyping to build a development model. The development model often relies upon specific design patterns and entity relationship diagrams (ERD). Without this upfront foundation, Iterative development can create long term challenges that are significant in terms of cost and quality.

Critics of iterative development approaches point out that these processes place what may be an unreasonable expectation upon the recipient of the software: that they must possess the skills and experience of a seasoned software developer. The approach can also be very expensive if iterations are not small enough to mitigate risk; akin to... "If you don't know what kind of house you want, let me build you one and see if you like it. If you don't, we'll tear it all down and start over." By analogy the critic argues that up-front design is as necessary for software development as it is for architecture. The problem with this criticism is that the whole point of iterative programming is that you don't have to build the whole house before you get feedback from the recipient. Indeed, in a sense conventional programming places more of this burden on the recipient, as the requirements and planning phases take place entirely before the development begins, and testing only occurs after development is officially over.

In fact, a relatively quiet turn around in the Agile community has occurred on the notion of "evolving" the software without the requirements locked down. In the old world this was called requirements creep and never made commercial sense. The Agile community has similarly been "burnt" because, in the end, when the customer asks for something that breaks the architecture, and won't pay for the re-work, the project terminates in an Agile manner.

These approaches have been developed along with web based technologies. As such, they are actually more akin to maintenance life cycles given that most of the architecture and capability of the solutions is embodied within the technology selected as the back bone of the application.

Refactoring is claimed, by the Agile community, as their alternative to cogitating and documenting a design. No equivalent claim is made of re-engineering - which is an artifact of the wrong technology being chosen, therefore the wrong architecture. Both are relatively costly. Claims that 10%-15% must be added to an iteration to account for refactoring of old code exist. However, there is no detail as to whether this value accounts for the re-testing or regression testing that must happen where old code is touched. Of course, throwing away the architecture is more costly again. In fact, a survey of the "designless" approach paints a picture of the cost incurred where this class of approach is used (Software Development at Microsoft Observed). Note the heavy emphasis here on constant reverse engineering by programming staff rather than managing a central design.

Test Driven Development (TDD) is a useful output of the Agile camp but some suggest that it raises a conundrum. TDD requires that a unit test be written for a class before the class is written. It might be thought, then, that the class firstly has to be "discovered" and secondly defined in sufficient detail to allow the write-test-once-and-code-until-class-passes model that TDD actually uses. This would be actually counter to Agile approaches, particularly (so-called) Agile Modeling, where developers are still encouraged to code early, with light design. However, to get the claimed benefits of TDD a full design down to class and responsibilities (captured using, for example, Design By Contract) is not necessary. This would count towards iterative development, with a design locked down, but not iterative design - as heavy refactoring and re-engineering might negate the usefulness of TDD.

TDD proponents tend to dispute the above analysis. Classic TDD suggests that any serious approach to TDD uses heavy refactoring, and that the primary value of TDD is to enable the kind of refactoring and light-up-front-design that critics might suggest to be problematic.

[edit] Process models

A decades-long goal has been to find repeatable, predictable processes that improve productivity and quality. Some try to systematize or formalize the seemingly unruly task of writing software. Others apply project management techniques to writing software. Without project management, software projects can easily be delivered late or over budget. With large numbers of software projects not meeting their expectations in terms of functionality, cost, or delivery schedule, effective project management appears to be lacking.

[edit] Formal methods

Formal methods are mathematical approaches to solving software (and hardware) problems at the requirements, specification and design levels. Examples of formal methods include the B-Method, Petri nets, RAISE and VDM. Various formal specification notations are available, such as the Z notation. More generally, automata theory can be used to build up and validate application behavior by designing a system of finite state machines.

Finite state machine (FSM) based methodologies allow executable software specification and by-passing of conventional coding (see virtual finite state machine or event driven finite state machine).

Formal methods are most likely to be applied in avionics software, particularly where the software is safety critical. Software safety assurance standards, such as DO178B demand formal methods at the highest level of categorization (Level A).

Formalization of software development is creeping in, in other places, with the application of OCL (and specializations such as JML) and especially with MDA allowing execution of designs, if not specifications.

Another emerging trend in software development is to write a specification in some form of logic (usually a variation of FOL), and then to directly execute the logic as though it were a program. The OWL language, based on Description Logic, is an example. There is also work on mapping some version of English (or another natural language) automatically to and from logic, and executing the logic directly. Examples are Attempto Controlled English, and Internet Business Logic, which does not seek to control the vocabulary or syntax. A feature of systems that support bidirectional English-logic mapping and direct execution of the logic is that they can be made to explain their results, in English, at the business or scientific level.