Programming complexity: Difference between revisions

Revision as of 20:31, 26 June 2008

Latest revision as of 15:37, 21 June 2025

(47 intermediate revisions by 39 users not shown)

Line 1:

{{Short description|Attribute of a software system}}

{{Mergeto|Software metric|Talk:Software metric|date=June 2008}}

{{Distinguish|Computational complexity theory}}'''Programming complexity''' (or '''software complexity''') is a term that includes software properties that affect internal interactions. Several commentators distinguish between the terms "complex" and "complicated". Complicated implies being difficult to understand, but ultimately knowable. Complex, by contrast, describes the interactions between entities. As the number of entities increases, the number of interactions between them increases exponentially, making it impossible to know and understand them all. Similarly, higher levels of complexity in software increase the risk of unintentionally interfering with interactions, thus increasing the risk of introducing defects when changing the software. In more extreme cases, it can make modifying the software virtually impossible.

{{Unreferenced|date=June 2008}}

'''Programming Complexity''' is the [[complexity]] of [[computer program|program]]s, [[Computer programming|programming]] and [[programming languages|languages]], and one of the [[unsolved problems in software engineering]].

The idea of linking software complexity to software maintainability has been explored extensively by [[Meir M. Lehman|Professor Manny Lehman]], who developed his [[Lehman's laws of software evolution|Laws of Software Evolution]]. He and his co-author [[Les Belady]] explored numerous [[software metrics]] that could be used to measure the state of software, eventually concluding that the only practical solution is to use deterministic complexity models.<ref>MM Lehmam LA Belady; Program Evolution - Processes of Software Change 1985</ref>

Applications are complex to the extent that when programmers [[Resignation|resign]] or are [[termination of employment|terminated]], [[company|companies]] [[fail]] if those companies have no one capable of [[understanding]] what the programmers did {{Who|date=December 2007}}. Because of this, [[researcher]]s establish [[software metric|metric]]s which [[measure]] the complexity and can be used to figure out how to reduce the complexity of the [[software]].

==Types==

There are several metrics one can use to measure programming complexity:

The complexity of an existing program determines the complexity of changing the program. Problem complexity can be divided into two categories:<ref>[https://academia.edu.documents.s3.amazonaws.com/1811257/SHSM2011.pdf In software engineering, a problem can be divided into its accidental and essential complexity [1<nowiki>]</nowiki>.]</ref>

* data complexity (Chapin Metric)

#'''Accidental complexity''' relates to difficulties a programmer faces due to the software engineering tools. Selecting a better tool set or a higher-level programming language may reduce it. Accidental complexity often results from not using the ___domain to frame the form of the solution.{{Citation needed|date=September 2015}} [[Domain-driven design]] can help minimize accidental complexity.

* data flow complexity (Elshof Metric)

#'''Essential complexity''' is caused by the characteristics of the problem to be solved and cannot be reduced.

* data access complexity (Card Metric)

* interface complexity (Henry Metric)

* control flow complexity (McCabe Metric)

* decisional complexity (McClure Metric)

* branching complexity (Sneed Metric)

* language complexity (Halstead Metric)

* [[cyclomatic complexity]]

<!--

==ReferencesMeasures==

Several measures of software complexity have been proposed. Many of these, although yielding a good representation of complexity, do not lend themselves to easy measurement. Some of the more commonly used metrics are

{{Reflist}}

* [[cyclomatic complexity|McCabe's cyclomatic complexity metric]]

-->

* [[Halstead complexity measures|Halstead's software science metrics]]

* Henry and Kafura introduced "Software Structure Metrics Based on Information Flow" in 1981,<ref>Henry, S.; Kafura, D. IEEE Transactions on Software Engineering Volume SE-7, Issue 5, Sept. 1981 Page(s): 510 - 518</ref> which measures complexity as a function of "fan-in" and "fan-out". They define fan-in of a procedure as the number of local flows into that procedure plus the number of data structures from which that procedure retrieves information. Fan-out is defined as the number of local flows out of that procedure plus the number of data structures that the procedure updates. Local flows relate to data passed to, and from procedures that call or are called by, the procedure in question. Henry and Kafura's complexity value is defined as "the procedure length multiplied by the square of fan-in multiplied by fan-out" (Length ×(fan-in × fan-out)²).

* Chidamber and Kemerer introduced "A Metrics Suite for Object-Oriented Design" in 1994,<ref name=":0">Chidamber, S.R.; Kemerer, C.F. IEEE Transactions on Software Engineering Volume 20, Issue 6, Jun 1994 Page(s):476 - 493</ref> focusing on metrics for object-oriented code. They introduce six OO complexity metrics: (1) weighted methods per class; (2) coupling between object classes; (3) response for a class; (4) number of children; (5) depth of inheritance tree; and (6) lack of cohesion of methods.

ThereSeveral are severalother metrics onecan canbe useused to measure programming complexity:

==See also==

* interfaceBranching complexity (HenrySneed Metric)

*[[Software crisis]] (and subsequent [[programming paradigm]] solutions)

* dataData access complexity (Card Metric)

*[[Software metric]]s - quantitative measure of some property of a program.

* dataData complexity (Chapin Metric)

* dataData flow complexity (Elshof Metric)

* control flowDecisional complexity (McCabeMcClure Metric)

*Path Complexity (Bang Metric)

[[Law of conservation of complexity|Tesler's Law]] is an [[adage]] in [[human–computer interaction]] stating that every [[Application software|application]] has an inherent amount of complexity that cannot be removed or hidden.

==Chidamber and Kemerer Metrics==

{{comp-sci-stub}}

{{Prose|date=August 2017}}

Chidamber and Kemerer<ref name=":0" /> proposed a set of programing complexity metrics widely used in measurements and academic articles: weighted methods per class, coupling between object classes, response for a class, number of children, depth of inheritance tree, and lack of cohesion of methods, described below:

* Weighted methods per class ("WMC")

** <math>WMC = \sum_{i=1}^nc_i</math>

** n is the number of methods on the class

** <math>c_i</math> is the complexity of the method

* Coupling between object classes ("CBO")

** number of other class which is coupled (using or being used)

* Response for a class ("RFC")

** <math>RFC = |RS|</math> where

** <math>RS = \{M\}\cup_{all\ i} \{R_i\}</math>

** <math>R_i</math> is set of methods called by method i

** <math>M</math> is the set of methods in the class

* Number of children ("NOC")

** sum of all classes that inherit this class or a descendant of it

* Depth of inheritance tree ("DIT")

** maximum depth of the inheritance tree for this class

* Lack of cohesion of methods ("LCOM")

** Measures the intersection of the attributes used in common by the class methods

** <math>LCOM = \begin{cases} |P| - |Q|, & \text{if } |P| > |Q|

\\0, & \text{otherwise } \end{cases}</math>

** Where <math>P = \{ (I_i,I_j)|I_i\cap I_j = \emptyset\}</math>

** And <math>Q = \{(I_i, I_j)|I_i \cap I_j \neq \emptyset\}</math>

** With <math>I_i</math> is the set of attributes (instance variables) accessed (read from or written to) by the <math>i</math>-th method of the class

==See also==

*[[Programming paradigm]]

*[[Software crisis]]

==References==

{{Reflist}}

[[Category:Software metrics]]

[[Category:Complex systems theory]]