Editing Multiple dispatch (section)

=== Use in practice ===
To estimate how often multiple dispatch is used in practice, Muschevici et al.<ref name=Muschevici>{{cite book |last1=Muschevici |first1=Radu |last2=Potanin |first2=Alex |last3=Tempero |first3=Ewan |last4=Noble |first4=James |title=Proceedings of the 23rd ACM SIGPLAN conference on Object-oriented programming systems languages and applications |chapter=Multiple dispatch in practice |year=2008 |series=OOPSLA '08 |pages=563–582 |doi=10.1145/1449764.1449808 |publisher=ACM |location=Nashville, TN, USA |isbn=9781605582153|s2cid=7605233 |url=https://figshare.com/articles/thesis/Multiple_Dispatch_in_Practice/16959112 }}</ref> studied programs that use dynamic dispatch. They analyzed nine applications, mostly compilers, written in six different languages: [[Common Lisp Object System]], [[Dylan (programming language)|Dylan]], [[Cecil (programming language)|Cecil]], MultiJava, Diesel, and Nice. Their results show that 13–32% of generic functions use the dynamic type of one argument, while 2.7–6.5% of them use the dynamic type of multiple arguments. The remaining 65–93% of generic functions have one concrete method (overrider), and thus are not considered to use the dynamic types of their arguments. Further, the study reports that 2–20% of generic functions had two and 3–6% had three concrete function implementations. The numbers decrease rapidly for functions with more concrete overriders.

Multiple dispatch is used much more heavily in [[Julia (programming language)|Julia]], where multiple dispatch was a central design concept from the origin of the language: collecting the same statistics as Muschevici on the average number of methods per generic function, it was found that the Julia [[standard library]] uses more than double the amount of overloading than in the other languages analyzed by Muschevici, and more than 10 times in the case of [[Operator (computer programming)|binary operators]].<ref name=julia-review/>

The data from these papers is summarized in the following table, where the dispatch ratio <code>DR</code> is the average number of methods per generic function; the choice ratio <code>CR</code> is the mean of the square of the number of methods (to better measure the frequency of functions with a large number of methods);<ref name=Muschevici/><ref name=julia-review/> and the degree of specialization <code>DoS</code> is the average number of type-specialized arguments per method (i.e., the number of arguments that are dispatched on):

{| class="wikitable sortable"
|-
! Language
! Average # methods (DR)
! Choice ratio (CR)
! Degree of specialization (DoS)
|-
| [[Cecil (programming language)|Cecil]]<ref name=Muschevici/> 
| 2.33 
| 63.30 
| 1.06
|- 
| [[Common Lisp]] ([[CMU Common Lisp|CMU]])<ref name=Muschevici/> 
| 2.03 
| 6.34 
| 1.17 
|- 
| Common Lisp ([[McCLIM]])<ref name=Muschevici/> 
| 2.32 
| 15.43 
| 1.17 
|-
| Common Lisp ([[Steel Bank Common Lisp|Steel Bank]])<ref name=Muschevici/> 
| 2.37 
| 26.57 
| 1.11 
|-
| Diesel<ref name=Muschevici/> 
| 2.07 
| 31.65 
| 0.71
|- 
| [[Dylan (programming language)|Dylan]] (Gwydion)<ref name=Muschevici/> 
| 1.74 
| 18.27 
| 2.14
|-
| Dylan (OpenDylan)<ref name=Muschevici/> 
| 2.51 
| 43.84 
| 1.23 
|- 
| [[Julia (programming language)|Julia]]<ref name=julia-review/> 
| 5.86 
| 51.44 
| 1.54 
|- 
| Julia (operators only)<ref name=julia-review/> 
| 28.13 
| 78.06 
| 2.01 
|-
| MultiJava<ref name=Muschevici/> 
| 1.50 
| 8.92 
| 1.02
|-
| Nice<ref name=Muschevici/> 
| 1.36 
| 3.46 
| 0.33
|- 
|}