Talk:Coupling (computer programming)

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Renaming to : Coupling (Computer science)[edit]

I propose to rename this entry to Coupling (Computer science) as this the most used terme by the specialists of the filed along with Cohesion, we need then to put a disambiguation link at the begining of Coupling. --Khalid hassani 14:36, 20 May 2005 (UTC)[reply]

Are you sure that coupling is used more than the dependency? Also I am not so sure what you mean by specialists. By that, do you mean computer scientists? computer programmers? hackers? -- Taku 20:10, May 20, 2005 (UTC)
by specialists I mean computer scientists and programmers, I have always heard about the couple coupling/cohesion as dual concepts et not dependency/cohesion. I think that there are in facts two different concepts in this entry, dependency which is in fact, rather related to RPM and packages, and coupling which is used in Object Oriented design, the two concepts are from two diffrents cultures, Linux hackers for the first and programmers using Object oriented languages for the second.--Khalid hassani 21:48, 21 May 2005 (UTC)[reply]
I would also rename this entry to Coupling (Computer science). In the article itself, the term "Coupling" occurs 28 times, whereas "Dependency" occurs only 6 times (not counting the name of the article). And Coupling#Computer programming points to here. I'm going to do the rename (maybe in about 8 or 10 hours). — Adrian | Talk 23:50, 9 November 2005 (UTC)[reply]
I have moved the article to Coupling (computer science). — Adrian | Talk 12:20, 10 November 2005 (UTC)[reply]

gains in the software development process[edit]

"...the gains in the software development process are greater than the value of the running performance gain."

I feel that this statement may be more opinionated than factual. Perhaps omission of this would be prudent?

I agree, and I've reworded this to make it clear that this is merely an opinion. Lunalot 21:44, 19 January 2007 (UTC)[reply]

Merge from the article "Low-Coupling / High-Cohesion pattern"[edit]

I have just merged the contents of the article "Low-Coupling / High-Cohesion pattern" into this article here under a new section with the same name. "Low-Coupling / High-Cohesion pattern" is now a redirect to here (see also Merging and moving pages). — Adrian | Talk 20:40, 10 November 2005 (UTC)[reply]

What about levelization and circular dependencies[edit]

John Lakos' book "Large-Scale C++ Software Design", although sometimes criticized for having some outdated C++ coding practices, has a good treatment of circular dependencies (a case of high coupling), as well as levelization. I believe this article is incomplete without a treatment of this subject.

Dependency hell[edit]

You are in Dependency Hell when you want connect to the Internet, try to install a package to stablish the access and the package has a lot of dependencies. You cannot download the packages because you are not connected and you cannot connect because you cannot dowload the packages.--Mac (talk) 16:58, 5 February 2008 (UTC)[reply]

--210.212.55.130 (talk)

Explicit and implicit dependencies[edit]

Can anyone include information about what are explicit and implicit dependencies ? And recursive implicit dependencies ?. --Nopetro (talk) 09:31, 11 February 2008 (UTC)[reply]

Attribution[edit]

The concepts of coupling and cohesion were invented by Larry Constantine, not Glenford Myers. Myers was a student of Constantine's at IBM's Systems Research Institute. Although Myers book was published before Yourdon and Constantine's classic, primary sources (e.g., 1974 IBM SYSTEMS J. aricle) make clear the source of the core ideas. Stirrer (talk) 17:14, 21 May 2008 (UTC)[reply]

Where is the Fenton Metric?[edit]

The Fenton Metric is missing. --phil (talk) 15:02, 21 June 2008 (UTC)[reply]

Inconsistent Terminology[edit]

The is some unfortunate inconsistency in the terminology used to describe different types of coupling that makes an already challenging subject even more difficult to follow. In the "Types of Coupling" section seven different types are discussed, and in the "Module Coupling" Section four types are mentioned. Unfortunately two of these latter four (global and environmental) are not described in the earlier section. I'm not being pedantic, just trying to understand some difficult concepts. Presumably there are alternative names in use here but this in not made clear anywhere. Inspeximus (talk) 07:15, 2 August 2011 (UTC)[reply]

Compiler as a Dependency[edit]

Some programs aren't packaged and require specific compilers, or are packaged and require compiling of OTA updates.

104.176.16.56 (talk) 00:48, 27 April 2018 (UTC)[reply]

Performance issues[edit]

This section seems to have nothing to do with coupling, and indeed the relationship between performance and coupling was much clearer before this section was introduced. The clarity of the section itself has deteriorated over time.

KeithC (talk) 13:51, 10 August 2018 (UTC)[reply]

The book name in reference [4] is incorrect.[edit]

The book name in the text of the article referenced in footnote 4 is not correct. The footnote is correct, the text reference lists the author, instead of the name of the book. Please fix. Gclose11 (talk) 21:55, 1 March 2023 (UTC)[reply]

This list of modes/types of coupling is extremely outdated[edit]

There is no mention of important types of coupling such as design-time versus runtime coupling, (execution) location coupling, creation coupling, call coupling, temporal coupling, data structure/data format coupling, inheritance coupling... 62.167.164.120 (talk) 14:54, 8 May 2023 (UTC)[reply]