User:CharlesGillingham/Wikipedia/Arguments for and against citation formats

Standardization

There are arguments for and against standardization.

Arguments in favor of standardization

1. Simplicity: If Wikipedia uses multiple techniques to solve the same problem, Wikipedia (as a whole) is necessarily more complicated.
   1. Shorter learning curve: If Wikipedia uses multiple techniques to solve the same problem, editors who wish to choose a technique will spend more time studying the existing techniques and comparing their merits. The harms newwer editors more than senior editors.
   2. Simpler documentation: If each of these techniques requires documentation, the documentation becomes bloated and the language of the documentation becomes convoluted. This harms newer editors more than long-time editors.
2. Ease of maintenance: Unusual techniques tend to poorly maintained.
   1. Maintenance by later editors: Later editors will be familiar with a standard technique and will have less problems while maintaining it.
   2. Solutions for common problems: If there is a large number of examples of a particular technique, then chances are that most of the common problems have been found and that techniques to correct the problems have been developed.
   3. Maintenance by bots: Common problems with standard techniques can be repaired by bots. It is unlikely that anyone will produce a bot that repairs a problem with a an unusual or idiosyncratic technique.
3. Professionalism: Standardization is used in all major publications (such as encyclopedias and journals).
4. Standardization need not alienate new editors and content providers: Wikipedia allows new editors to make any bold edit they like, regardless if it obeys standards or not. Standardization is not something that should concern or inhibit editors who add content, whether they are new or not. Standardization is most effective when carried out by later editors (known as WP:GNOMEs) who specialize in fixing details. For this to be effective, WP:GNOMEs must always be WP:CIVIL and WP:ASSUMEGOODFAITH (editors who are aware of a standard should (1) avoid a critical tone (2) discuss their changes on the talk page and (3) avoid giving work to others, i.e. the should plan to fix it themselves). Also, content providers should should be aware they do not WP:OWN the articles they write, even if they have provided all the content on the page.

Arguments against standardization

1. Unanticipated exceptions and Procrustean solutions: All standardization schemes suffer from the fact that, in a large project like Wikipedia, there are always unusual cases that defy standardization. Forcing unusual cases into a standardized form can create unnecessary complexity and confusion. "A foolish consistency is the hobgoblin of little minds." (Emerson)
   1. Exceptions make the documentation of a standard unusable The documentation must necessarily explain all the exceptions that have ever been discovered. Solving an unusual problem requires sifting through complex documentation in order to find the special parameter or other work-around that Wikipedia has found to solve this problem. Without a standard, an adequate solution to a problem can usually be invented by any reasonably intelligent editor in a few minutes.
2. Some of our standards are poor choices and standardization stifles innovation: It is necessarily true that, at some point, some of Wikipedia's standards are poor choices and that better choices exist. If some articles use techniques that are better than Wikipedia's standards (at least arguably), these articles should be left alone because they may become the new standard. Any innovation is, by definition, non-standard.
   1. I'm sure there are dozens of examples
   3. ... etc
3. Different fields and nationalities require different standards For example, some fields prefer a particular citation format. Editors of these articles would prefer to use a citation format they are familiar with, rather than learning Wikipedia's standard.
4. Standardization is unnecessary: Standardization is an example of WP:CREEP. A lack of standards in certain areas does not harm the primary goals of the project. Arguing about a (some) standards is a waste of everyone's time.
5. Standardization alienates content providers and new editors unnecessarily: In practice, editors who are aware of standards are critical, arrogant and inflexible. In practice, content providers are unwilling to allow others to make substantial changes to the articles they write.

Current consensus

Main pages: WP:ENGVAR and WP:CITEVAR

In some cases, Wikipedia has chosen to standardize but in many other cases Wikipedia has chosen to allow editors to choose for themselves what to do in a particular article. In these cases, Wikipedia has chosen to standardize on an article-by-article basis. This gives precedence to the "existing method" of the article.

Citation templates

The basic argument is between those who think that citation templates should be used whenever possible and those who think that many different citation formats should be allowed. Other positions are given below under "alternatives".

Arguments in favor of recommending citation templates as the standard for Wikipedia's citations

1. Standardization: The plurality of wikipedia's completed citations are formatted with citation templates; citation templates are the de-facto standard. As such, citation templates have all the benefits associated with standardization:
   1. Simplicity
      1. Shorter learning curve
      2. Simpler documentation
   2. Maintenance
      1. Maintenance by later editors
      2. Solutions to common problems
      3. Maintenance by bots: This is particularly easy for citation templates, because they have a machine-readable semantics.
   3. Professionalism

1. Machine readable semantics: Citation templates blah blah

Arguments against the use of citation templates

These arguments favor allowing any citation format that uniquely identifies a source, regardless of how it is formatted. If an article uses an identifiable citation system, then all the citations in that article should use the same system. This is the current consensus (as of February 2011).

1. Clutter in edit-mode:
   • which can be alleviated by the use of list-defined references.
2. Load time: Large numbers of templates cause pages to load considerably slower, as of spring 2011.
   • which could be resolved by the adoption of an enhanced cite.php that extended the ref-tag syntax.
3. Standardization: Citations templates have all the problems associated with standardization, described above.
   1. Unanticipated exceptions and Procrustean solutions
      1. Exceptions make the documentation of a standard unusable
   2. Some of our standards are poor choices and standardization stifles innovation
   3. Different fields and nationalities require different standards
   4. Standardization is unnecessary
   5. Standardization alienates content providers and new editors unnecessarily
4. There are excellent alternatives to using citation templates

List defined references

Arguments in favor of list defined references

• Reduces clutter in edit-mode.

Arguments against list defined references

• Encourages editors to use opaque computerese names for footnotes
• Difficult to maintain, especially when there are large numbers of footnotes. See, for example, Artificial intelligence, where fixing a misplaced comma in a footnote can be a very difficult chore; you have to find the name go to the reference section and search for the name.

The Issues

Adjacent References

Maintaining a citation is easier if the citation is directly adjacent to the material it supports. This was one of the motivations for cite.php, and a strong argument against shortened footnotes or list defined references.

Load time issue

See Wikipedia talk:Featured article candidates/Citation templates (technical)

Clutter in Edit Mode

Opponents of citation templates argue that citations with citation templates cause more edit-mode cluttered. For example, they argue that this:

"The philosophy of AI", writes [[John McCarthy (computer scientist)|John McCarthy]], "is unlikely to have any more effect on the practice of AI research than philosophy of science generally has on the practice of science."<ref>[[John McCarthy (computer scientist)|McCarthy, John]] (1996), "[http://www-formal.stanford.edu/jmc/aiphil/node2.html#SECTION00020000000000000000 The Philosophy of Artificial Intelligence]], ''What has AI in Common with Philosophy''</ref>

is less cluttered than this:

"The philosophy of AI", writes [[John McCarthy (computer scientist)|John McCarthy]], "is unlikely to have any more effect on the practice of AI research than philosophy of science generally has on the practice of science."<ref>{{Citation | last=McCarthy | first=John | author-link=John McCarthy (computer scientist) | title=The Philosophy of Artificial Intelligence | work=What has AI in Common with Philosophy? | year=1996 | url=http://www-formal.stanford.edu/jmc/aiphil/node2.html#SECTION00020000000000000000}}</ref>

Or this:

"The philosophy of AI", writes [[John McCarthy (computer scientist)|John McCarthy]], "is unlikely to have any more effect on the practice of AI research than philosophy of science generally has on the practice of science."<ref>
{{Citation
| last=McCarthy | first=John | author-link=John McCarthy (computer scientist)
| title=The Philosophy of Artificial Intelligence
| work=What has AI in Common with Philosophy?
| year=1996
| url=http://www-formal.stanford.edu/jmc/aiphil/node2.html#SECTION00020000000000000000
}}</ref>

The problem of edit-mode clutter is directly addressed by both short citations

the philosophy of AI, writes [[John McCarthy (computer scientist)|John McCarthy]], "is unlikely to have any more effect on the practice of AI research than philosophy of science generally has on the practice of science."{{sfn|McCarthy|1996}}

and list-defined references.

the philosophy of AI, writes [[John McCarthy (computer scientist)|John McCarthy]], "is unlikely to have any more effect on the practice of AI research than philosophy of science generally has on the practice of science."<ref name=McCarthy1996/>