Recent from talks
Nothing was collected or created yet.
History of wikis
View on Wikipedia
This article has multiple issues. Please help improve it or discuss these issues on the talk page. (Learn how and when to remove these messages)
|

The history of wikis began in 1994, when Ward Cunningham gave the name "WikiWikiWeb" to the knowledge base, which ran on his company's website at c2.com, and the wiki software that powered it. The wiki went public in March 1995, the date used in anniversary celebrations of the wiki's origins.[1] c2.com is thus the first true wiki, or a website with pages and links that can be easily edited via the browser, with a reliable version history for each page. He chose "WikiWikiWeb" as the name based on his memories of the "Wiki Wiki Shuttle" at Honolulu International Airport, and because "wiki" is the Hawaiian word for "quick".[2]
Wiki software has some conceptual origins in the version control and hypertext systems used for documentation and software in the 1980s, and some actualized origins in the 1970s "Journal" feature of NLS. Its distant ancestors include Vannevar Bush's proposed "memex" system in 1945, the collaborative hypertext database ZOG in 1972, the NoteCards system from Xerox, the Apple hypertext system HyperCard. As was typical of these earlier systems, Cunningham's motive was technical: to facilitate communication between software developers.
Many alternative wiki applications and websites appeared over the next five years. In the meantime, the first wiki, now known as "WardsWiki", evolved as features were added to the software and as the growing body of users developed a unique "wiki culture". By 2000, WardsWiki had developed a great deal of content outside its original stated purpose, which led to the spinoff of content into sister sites, most notably MeatballWiki.[3]
The website Wikipedia, a free content encyclopedia, was launched in January 2001,[4] and quickly became the most popular wiki, which it remains to this day. Its rise in popularity (it was in the top ten most popular sites in 2007[5]) played a large part in introducing wikis to the general public.[4] There now exist at least hundreds of thousands of wiki websites, and they have become increasingly prevalent in corporations and other organizations.[4]
Hypertext editing: pre-1994
[edit]A distant precursor of the wiki concept was Vannevar Bush's vision of the "memex", a microfilm reader which would create automated links between documents. In a 1945 essay in Atlantic Monthly titled "As We May Think", Bush described an imaginary future user interface: "Before him are the two items to be joined, projected onto adjacent viewing positions… The user taps a single key, and the items are permanently joined… Thereafter, at any time, when one of these items is in view, the other can be instantly recalled merely by tapping a button below the corresponding code space. Moreover, when numerous items have been thus joined together to form a trail, they can be reviewed in turn…"[6] This vision, though it has been described as predicting the World Wide Web, resembles wikis more than the web in one important way: the system being described is self-contained, not a loose network.
In 1972 Kristo Ivanov published a PhD dissertation on Quality-control of information,[7] containing a theoretical basis for what corresponds to the wiki-idea, in terms of systemic social interaction. Information turns into knowledge as a net of contributions and negotiations converge about a core concept, or entity. The emphasis is on a dynamically documented "agreement in the context of maximum possible disagreement", akin to the discussions in talk pages and the results of view history of Wikipedia.
Pre-World-Wide-Web hypertext systems
[edit]An indirect precursor of the wiki concept was the ZOG multi-user database system, developed in 1972 by researchers at Carnegie-Mellon University. The ZOG interface consisted of text-only frames, each containing a title, a description, a line with standard ZOG commands, and a set of selections (hypertext links) leading to other frames.
Two members of the ZOG team, Donald McCracken and Robert Akscyn, spun off a company from CMU in 1981 and developed an improved version of ZOG called Knowledge Management System (KMS). KMS was a collaborative tool based on direct manipulation, permitting users to modify the contents of frames, freely intermixing text, graphics and images, any of which could be linked to other frames. Because the database was distributed and accessible from any workstation on a network, changes became visible immediately to other users, enabling them to work concurrently on shared structures (documents, programs, ...).[8]
Three notable hypertext-based systems emerged in the 1980s, that may have been inspired by ZOG, KMS and/or one another: the NoteCards system, developed in 1984 and released by Xerox in 1985; Janet Walker's Symbolics Document Examiner, created in 1985 for the operation manuals of Symbolics computers; and Bill Atkinson's WildCard application, on which he began work in 1985, and which was released in 1987 as HyperCard.[8] Ward Cunningham has stated, that the wiki idea was influenced by his experience using HyperCard: he was shown the software by fellow programmer Kent Beck, before its official release (it was still called "WildCard" at the time), and, in his words, was "blown away" by it.[9][1]
Cunningham used HyperCard to make a stack holding three kinds of information: ideas, people who hold ideas, and projects where people share ideas. (He would later use this same architecture for the Patterns, People and Projects listed on the front page of his original wiki, the WikiWikiWeb.) Cunningham made a single card that would serve for all uses. It had three fields: Name, Description and Links. Cunningham configured the system so that links could be created to cards that did not exist yet; creating such a link would in turn create a new blank card.[9]
The World Wide Web
[edit]In 1990, Tim Berners-Lee of CERN built the first hypertext client, which he called World Wide Web (it was also a Web editor), and the first hypertext server (info.cern.ch). In 1991 he posted a short summary of the World Wide Web project on the alt.hypertext newsgroup, marking the debut of the Web as a publicly available service on the Internet.
Early adopters of the World Wide Web were primarily university-based scientific departments or physics laboratories. In May 1992 appeared ViolaWWW, a graphical browser providing features such as embedded graphics, scripting, and animation. However, the turning point for the World Wide Web was the introduction of the Mosaic graphical browser in 1993, which gained wide popularity due to its strong support of integrated multimedia. In April 1993, CERN agreed that anyone could use the Web protocol and code for free.[10]
WikiWikiWeb and the birth of wikis: 1994–2001
[edit]WikiWikiWeb, the first wiki
[edit]Ward Cunningham started developing the WikiWikiWeb in 1994 as a supplement to the Portland Pattern Repository, a website containing documentation about software design patterns, a particular approach to object-oriented programming.[1]
The WikiWikiWeb was intended as a collaborative database, in order to make the exchange of ideas between programmers easier; it was dedicated to "People, Projects and Patterns".[1] Cunningham wrote the software to run it using the Perl programming language. He considered calling the software "quick-web", but instead named it using the Hawaiian word "wiki-wiki", which means "quick-quick" or "very quick",[1] based on his memory of the Wiki Wiki Shuttle at Honolulu International Airport.
Cunningham installed a prototype of the software on his company Cunningham & Cunningham's website c2.com.[11] When the site was functioning, Cunningham sent the following email to a colleague:
Steve – I've put up a new database on my web server and I'd like you to take a look. It's a web of people, projects and patterns accessed through a cgi-bin script. It has a forms-based authoring capability that doesn't require familiarity with html. I'd be very pleased if you would get on and at least enter your name in RecentVisitors. I'm asking you because I think you might also add some interesting content. I'm going to advertise this a little more widely in a week or so. The URL is http://c2.com/cgi-bin/wiki. Thanks and best regards
— Ward Cunningham, [11]
Cunningham dates the official start of WikiWikiWeb as March 25, 1995.[1] On May 1, 1995, he sent an email about the website to a number of programmers, which caused an increase in participation.[1] This note was posted to the "Patterns" listserv, a group of software developers gathered under the name "The Hillside Group" to develop Erich Gamma's use of object-oriented patterns. Cunningham had noticed that the older contents of the listserv tended to get buried under the more recent posts, and he proposed instead to collect ideas in a set of pages which would be collectively edited. Cunningham's post stated: "The plan is to have interested parties write web pages about the People, Projects and Patterns that have changed the way they program." He added: "Think of it as a moderated list where anyone can be moderator and everything is archived. It's not quite a chat, still, conversation is possible."[12]
The site was immediately popular within the pattern community.[1]
CamelCase and internal links
[edit]Among Cunningham's innovations in creating WikiWikiWeb was the ability to easily link internally between pages, something that was often cumbersome to do in previous intranet and document management systems. Cunningham's solution to this was to automatically link any text expressed in CamelCase, including text for which a corresponding page did not yet exist.
This CamelCase convention was used by most wiki software for the first few years of wikis' existence. In 2001, the software UseModWiki, which at the time was in use on Wikipedia, switched to allow internal links to be done using standard spelling and double square bracket instead, in order to improve Wikipedia's usability.[13] This square bracket syntax has since become more of a default convention for internal links within wiki software in general.
Release of the Wiki Base software
[edit]Ward Cunningham wrote a version of his wiki software meant for public usage, called "Wiki Base". In his announcement, he wrote: "WikiWikiWeb is almost public. Actually, a pretty good clone of it is public at: https://web.archive.org/web/20030801073834/http://c2.com/cgi/wikibase. I've translated almost all of the actual wiki script into HyperPerl, a wiki-literate programming system that I think you will like." Visitors were requested to register on the wiki before they took the Wiki Base code.[1] Cunningham expected users to fold changes back into his editable version, but those who implemented changes generally chose to distribute the modified versions on their own sites.[1]
Alternate applications for wikis began to emerge, usually imitating the look-and-feel of WikiWikiWeb/Wiki Base; such applications were originally known as "WikiWikiClones". The first one was most likely created by IBM programmer Patrick Mueller, who wrote his in the REXX language, even before Wiki Base was released.[1]
Early wiki websites for software development
[edit]Inspired by the example of the WikiWikiWeb, programmers soon started several other wikis to build knowledge bases about programming topics. Wikis became popular in the free and open-source software (FOSS) community, where they were used for collaboratively discussing and documenting software.[citation needed] However, being used only by specialists, these early software-focused wikis failed to attract widespread public attention.[14]
Growth and innovations in WikiWikiWeb from 1995 to 2000
[edit]The WikiWikiWeb website approximately doubled in size every year 1995 to 2000, with disk usage rising from around 2 megabytes in 1995 to around 60 megabytes at the end of 2000.[1] During that time, various innovations were put in place, many suggested by users, to help with navigation and editing. These included:[1]
- 1995 – RecentVisitors, PeopleIndex: pages to help users know who was contributing
- 1995 – NotSoRecentChanges: excess lines from the RecentChanges page were (manually) copied to a file of "ChangesIn%lt;Month>"
- 1996 – EditCopy: offers the possibility to edit the backup copy of a page (this was replaced in 2002 with Page History)
- 1996 – ThreadMode: the form of a page where community members hold a discussion, each signing their own contribution
- 1996 – WikiCategories: categories can be added as an automatic index to pages
- 1997 – RoadMaps: proposed lists of pages to consult about specific topics, such as the Algorithms RoadMap or the Leadership RoadMap
- 1999 – ChangeSummary: an aid to telling which changes added interesting new content and which were only minor
- 2000 – UserName: the Wiki will accept a cookie that specifies a User Name to be used in place of the host name (IP identity) in the RecentChanges log
"ThreadMode" was defined as "a form of discussion where our community holds a conversation". It consists of a series of signed comments added down the page in chronological order. Ward Cunningham generally frowned on ThreadMode, writing: "Chronological is only one of many possible organizations of technical writing and rarely the best one at that."[15]
Cunningham encouraged contributors to "refactor" (rewrite) the ThreadMode discussions into DocumentMode discourse. In practice many pages started out at the top in DocumentMode and degenerated into ThreadMode further down. When ThreadMode became incomprehensible the result was called "ThreadMess".[16] (In most modern wikis the conflict between these two modes has been resolved by putting all document text on the main page of an article, and all discussion text on a talk page.)
The use of categories was proposed by user Stan Silver on August 27, 1996.[17] His initial post suggested: "If everyone adds a category and topic to their page, then the category and topic pages themselves can be used as automatic indexes into the pages."[18] Initially Silver had proposed both categories and topics: categories denoted the specific nature of the page's subject (a book, a person, a pattern), while topics denoted the theme of the page (Java, extreme programming, Smalltalk). However, people ignored this separation, and topics were collapsed into the categories.[17]
The "ChangeSummary" option began as an aid to telling which changes added interesting new content, and which were just minor adjustments of spelling, punctuation, or correction of web links. It started when some users began taking the RecentChanges page, annotating each line with a brief description of each change, and posting the result to the ChangeSummary page. This practice was highly time-consuming and rapidly petered out, but was replaced by the "MinorEdit/RecentEdits" feature, designed to reduce the RecentChanges clutter.[1]
Tensions within WikiWikiWeb and the creation of sister sites
[edit]Between early 1998 and the end of 2000 participation in WikiWikiWeb snowballed, and the disk space consumed by wiki pages more than quadrupled. With increased participation tensions began to appear.
In 1998 proponents of Extreme Programming showed up on the site and started posting comments about Extreme Programming on most of the pages related to software development. This annoyed a number of people who wanted to talk about patterns, leading to the tag "XpFreeZone", which was put onto pages as a request not to talk about ExtremeProgramming on that page. Eventually most of the DesignPatterns people left to discuss patterns on their own wikis, and WikiWikiWeb began to be referred to as "WardsWiki" instead of the "PortlandPatternRepository".[1][better source needed]
Around the summer of 1999, user Sam Gentile posted the comment "I'm through here" on his user page, and began systematically removing his text from all pages on WikiWikiWeb that he had contributed to. Gentile worked at Microsoft and had been hurt by what he perceived as anti-Microsoft bias on WikiWikiWeb. His deletions led to controversy about whether he had the right to remove his own material, and whether others had the right to put it back in (which some began to do).[citation needed] This event became referred to as the "WikiMindWipe".[19][better source needed]
In April 2000, four WikiWikiWeb users independently tried to reduce the amount of text on the site via a large number of deletions. They mainly tried to delete material that was related to wikis and not software design patterns.[20] They considered this material to be dead weight, and would have preferred to see it all replaced by concise guidance to newcomers. Contributors who disagreed with these deletions began copying all of the deleted text back in again. A vote was taken on the issue, and it was proposed that any major reductions should be pre-announced, with an opportunity for response before action was taken.[21]
The longer-term result of the deletions was the formation of WikiWikiWeb "sister sites" later in 2000. Sunir Shah created a wiki called MeatballWiki, intended strictly for wiki-based documentation and discussions. A few months later, Richard Drake created the WhyClublet (or "Why?") wiki to host discussion of Christian issues, and Peter Merel created GreenCheeseWiki and The Reform Society to host, respectively, whimsical and political pages. Earle Martin subsequently created a catch-all for C2 off-topic pages called TheAdjunct. Many pages were moved from WikiWikiWeb to these alternative sites, with a stub of the moved page left on the WikiWikiWeb, containing a link to the new page and the message "This page exists only on SisterSites."
In 2001, Cunningham and user Bo Leuf published a book, The Wiki Way, which distilled the lessons learned during the collective experience of the first wiki.[22]
Other wiki websites, 1999–2000
[edit]While many early wiki websites were devoted to the development of open source software, one early wiki was created by the FoxPro user community. The FoxPro Wiki was founded in 1999 by Steven Black and evolved into a popular site with many pages.[23]
World66 was a Dutch company which tried to transform the open content idea into a profitable business. The website was founded in 1999 by Richard and Douwe Osinga, and contains travel-related articles covering destinations around the world.
A wiki forum was created in 1999 for discussion of the newly created PhpWiki software. This became one of the larger software-related wikis. Sensei's Library, a wiki dedicated to discussion of the game of Go, was created by Morten G. Pahle and Arno Hollosi in October 2000. For its first few years of operation, it was one of the largest and most active wikis outside Wikipedia.
Other wiki applications, 1997–2000
[edit]Clones of the WikiWikiWeb software began to be developed as soon as Ward Cunningham made the Wiki Base software available online. One of the early clones was CvWiki, developed in 1997 by Peter Merel, which was the first wiki application to have functioning transclusion, backlinks, and "WayBackMode".
JWiki[24] (short for JavaWiki), released in 1997, was the first implementation of WikiWikiWeb in the Java language, and the first to be back-ended by a database. It was developed by Ricardo Clements, a former co-worker of Cunningham's.
Another early wiki engine was JosWiki, developed by an international group of Java programmers who were trying to create a free and open "Java Operating System" (JOS).[25]
TWiki was created in Perl by Peter Thoeny in 1998, based on JosWiki. TWiki was aimed at large corporate intranets. It stored data in plain text files instead of in a database.
PikiPiki was created by Martin Pool in 1999 as a rewrite of WikiWikiWeb in Python.[citation needed] It was made to be a small program, using flat files and doing away with versioning (Pool felt that a wiki is not meant to be a document-management system).[26]
PhpWiki, created by Steve Wainstead in 1999, was the first wiki software written in PHP.[citation needed] The initial version was a feature-for-feature reimplementation of the WikiWikiWeb software. Subsequent versions adopted many features from UseModWiki.
Swiki was written in Squeak by Mark Guzdial and Jochen Rick in 1999.[citation needed]
Zwiki, written in Python in 1999, was based on the Zope web application server (it could also co-exist with the Plone content management system). It was initially developed by Simon Michael and Joyful Systems.[citation needed]
Traction TeamPage was released in December 1999; it was the first proprietary wiki application aimed at enterprise customers.
UseModWiki was developed from 1999 to 2000 by Clifford Adams. UseModWiki is a flat-file wiki written in Perl. It was based on Markus Denker's AtisWiki, which was in turn based on CvWiki. It introduced the square bracket syntax for linking words that was later adopted by many other wiki engines, such as MediaWiki.
MoinMoin, created in Python by Jürgen Hermann and Thomas Waldmann in mid-2000, was initially based on PikiPiki. It is a flat-file wiki with a simple code base but many possible extensions. MoinMoin uses the idea of separating the parsers (for parsing the wiki syntax) from the formatters (for outputting HTML code), with an interface between them, so that new parsers and output formatters can be written.
Wikipedia's early years: 2001–2003
[edit]The creation of Wikipedia
[edit]Until 2001, wikis were virtually unknown outside the restricted circles of computer programmers.[citation needed] Wikis were introduced to the general public by the success of Wikipedia, a free content encyclopedia that can be edited by anyone.[4]
Wikipedia was originally conceived as a complement to Nupedia, a free on-line encyclopedia founded by Jimmy Wales, with articles written by highly qualified contributors and evaluated by an elaborate peer review process. The writing of content for Nupedia proved to be extremely slow, with only 12 articles completed during the first year, despite a mailing-list of interested editors and the presence of a full-time editor-in-chief recruited by Wales, Larry Sanger.[4] Learning of the wiki concept, Wales and Sanger decided to try creating a collaborative website to provide an additional source of rapidly produced draft articles that could be polished for use on Nupedia.
Nupedia's editors and reviewers resisted the idea of associating Nupedia with a wiki-style website,[citation needed] so Wikipedia was launched on its own domain, wikipedia.com, on January 15, 2001.[4] It initially ran on UseModWiki software, with the original text stored in flat-files rather than in a database, and with articles named using the CamelCase convention. UseModWiki was replaced by a PHP wiki engine in January 2002 and by MediaWiki in July 2002.[citation needed]
Wikipedia attracted new participants after being mentioned on Slashdot as well as in an article on the community-edited website Kuro5hin.[27] It quickly overtook Nupedia. In the first year of its existence, over 20,000 encyclopedia entries were created.
Wikimedia Foundation and first Wikipedia sister projects
[edit]Wales, and other members of the Wikipedia user community, founded Wikipedia's first "sister site", "In Memoriam: September 11 Wiki",[28] in October 2002;[29] it detailed the September 11 attacks. (This project was closed in October 2006.) A second sister site, Wiktionary, was created in December 2002; the site was meant to be a collaboratively created dictionary.
In June 2003, Wales founded the Wikimedia Foundation, a non-profit organization, to manage Wikipedia and all its sister projects going forward. Two additional Wikimedia projects were added soon thereafter: Wikiquote, a reference for notable quotations, and Wikibooks, for collaboratively creating textbooks, both in July 2003. Another project, Wikisource, was added in November 2003; it was originally named "Project Sourceberg" (a play on Project Gutenberg), then renamed "Wikisource" in December 2003.
Other wiki websites, 2001–2003
[edit]MeatballWiki rapidly became a popular wiki for discussions of wiki-related topics. The users of MeatballWiki came up with several ideas on the linking together of wikis, including:[30][31]
- InterWikiMap, a simple interwiki linking system (2000)[32]
- MetaWiki, the idea of a wiki that helps people find other wikis[33] (not to be confused with the Wikimedia Meta-Wiki)
- OneBigWiki (2002), the idea of having one wiki distributed across several servers[34]
- SwitchWiki (2003): the idea of having one site where one can switch between wikis[35]
- WikiNode, a way to link wikis via a standard "node" page on each.
- WackoWiki (2003), Fork from WikkaWiki one of the more than 100 Wiki Engines, that helped foster Wiki principles usage and development [36]
Some of these ideas were later implemented. For example, WikiIndex, a wiki that lists other wikis, was created in 2006, in an attempt to implement the MetaWiki and SwitchWiki ideas. The site also includes a WikiNode of its own and catalogs sites which include their own WikiNodes.
Disinfopedia was launched by Sheldon Rampton in March 2003. It aimed to produce a directory of public relations firms and industry-funded organizations that attempt to influence public opinion and public policy. It was later renamed SourceWatch, and is currently run by the Center for Media and Democracy (CMD).
Javapedia was a wiki inspired by Wikipedia. The project was launched in June 2003 during the JavaOne developer conference, and was intended to cover all aspects of the Java platform.
Wikitravel was started in July 2003 by Evan Prodromou and Michele Ann Jenkins.
Memory Alpha is a wiki devoted to the Star Trek fictional universe. It was launched by Harry Doddema and Dan Carlson in December 2003. For its first several years, it was one of the largest wiki projects.
This period also saw the creation of several other general wiki encyclopedias, created either independently of Wikipedia or meant to serve as an alternative to Wikipedia in order to fix some perceived weakness in it:
- Susning.nu was a Swedish-language wiki, created in October 2001, meant to serve as an encyclopedia, dictionary, and discussion forum.
- Enciclopedia Libre was created in February 2002 as a fork of the Spanish-language Wikipedia, by a group of contributors to the Spanish Wikipedia, who left because of fears of censorship and the possibility of the placement of advertising on Wikipedia.
- Wikinfo was launched in July 2003 as "Internet-Encyclopedia". It was a fork of the English-language Wikipedia, meant to hold original research and multiple articles on subjects from different points of view, instead of Wikipedia's policy of a single neutral-point-of-view article.[37] In 2013, after a period of downtime, the content was removed and a new version of Wikinfo was started at Wikia; however, as of June 2017, Wikinfo had been restored on wikinfo.org.[38] Sometime later, the wikinfo.org domain lapsed from existence once again (as of January 2021).
- WikiZnanie is a Russian-language wiki encyclopedia created in 2003; it took most of its content from the Russian Brockhaus and Efron Encyclopedic Dictionary of 1906.
Development of wiki software, 2001–2003
[edit]JSPWiki, created by Janne Jalkanen in 2001, is flat-file wiki software built around JavaServer Pages (JSP). JSPWiki adapted and extended the PhpWiki markup. It is primarily used for company and university intranets as a project wiki or a knowledge management application. It has also been used as a personal information manager (PIM).
MediaWiki was written for Wikipedia in 2002 by Lee Daniel Crocker, based on the user interface design of an earlier PHP wiki engine developed by Magnus Manske. Manske's PHP-based software suffered load problems due to increased use, so Crocker re-wrote the software with a more scalable MySQL database backend. As Wikipedia grew, achieving scalability through multiple layers of caching and database replication became a major concern for the developers. Internationalization was also a major concern[citation needed] (the user interface has been translated into more than 70 languages). One of the earliest differences between MediaWiki and other wiki engines was the use of freely formatted links instead of links in CamelCase. MediaWiki provides specialized syntax to support rich content, such as rendering mathematical formulas using LaTeX, graphical plotting, image galleries and thumbnails, and Exif metadata. MediaWiki lacks native WYSIWYG support, but comes with a graphical toolbar to simplify editing. One MediaWiki innovation for structuring content is "namespaces". Namespaces allow each article to contain multiple sheets with different standard names: one sheet presents the encyclopedic content, another contains the discussions surrounding it, and so on.
PmWiki was created in PHP by Patrick Michaud in 2002. It is a flat-file wiki engine that was designed to be easy to install and customize. PmWiki offers a template scheme that makes it possible to change the look and feel of the wiki. Customization is made easy through a wide selection of custom extensions, known as "recipes" available from the PmWiki Cookbook.
TikiWiki was created in PHP by Luis Argerich in 2002. It was later renamed "Tiki Wiki CMS Groupware", or simply "Tiki", as it gained additional CMS and groupware features. TikiWiki is modular with components that can be individually enabled and customized by the administrator, and extends customization to the user with selectable skins and themes.
Socialtext launched in 2002; it was the second proprietary enterprise-focused wiki application, after Traction TeamPage.
As they developed, wikis incorporated many of the features used on other websites and blogs, including:
- support for various wiki markup styles
- editing of pages with a GUI editor, WYSIWYG, and specific applications such as LaTeX
- optional use of external editors
- support for plugins and custom extensions
- use of RSS feeds
- integrated email discussion
- precise access control
- spam protection
Explosion in interest: 2004–2006
[edit]The period from 2004 to 2006 saw an explosion in interest in both wikis generally and Wikipedia in particular, and both started to become household terms. Corporations, organizations and other communities began to make increasing use of wikis. Many of the wiki-based sites, technologies and events that dominate today were started during that period.
XWiki is an open-source Java wiki application that was created by Ludovic Dubost in December 2003. Dubost set up hosting of XWiki-based wikis on the domain xwiki.com in January 2004—xwiki.com thus became an early (and possibly the first) wiki hosting service, or "wiki farm".
In October 2004, the wiki hosting service Wikicities launched, co-founded by Wikipedia founder Jimmy Wales and Wikimedia Foundation board member Angela Beesley. In March 2006, it changed its name to Wikia. Wikia remains one of the largest and most popular wiki farms.
A large number of other notable wiki farms were released soon afterward, including Wikispaces (launched March 2005), PBwiki (launched June 2005, later renamed PBworks), and Wetpaint (launched October 2005). Wikidot was launched in August 2006.
2005 marked the beginning of large-scale wiki-related meetings and conferences. August 2005 saw the first-ever Wikimania, an annual conference organized around Wikimedia Foundation projects, in Frankfurt, Germany. WikiSym, a more academic annual symposium about wikis in general, was first held a few months later, in October 2005, in San Diego, California; in 2014 it was renamed to OpenSym. RecentChangesCamp, an unconference dedicated to wikis, was first held in February 2006 in Portland, Oregon. Wikimania and OpenSym remain the two largest wiki-related gatherings.
Around 2005 wikis began to be massively confronted with "wiki spam", produced by spammers who enter website addresses onto wikis in order to improve the ranking of the displayed websites by search engines. Various strategies have been developed to counter wiki spam.[39]
Wikipedia and other Wikimedia projects
[edit]Wikipedia experienced exponential growth in usage and readership during the period from 2004 to 2006, rising in Alexa rankings from the top 1000 websites into the top 15.[40] It was named one of the top 5 global brands of 2006 in the Brandchannel Readers' Choice Awards.[41]
In 2004, the Wikimedia Foundation launched three new sites: Wikispecies, for cataloging species, in August 2004, Wikimedia Commons, to hold images and other media used by the Wikimedia projects, in September 2004, and Wikinews, for publishing collaborative news articles, in December 2004. Wikiversity, intended for tutorials and other courseware, was later launched in August 2006.
In November 2005, journalist John Seigenthaler wrote a much-publicized article in USA Today about Wikipedia's article about him, which for over four months had contained a false statement about him, inserted as a joke, stating that he had been a suspect in the assassinations of John F. Kennedy and Robert F. Kennedy. The USA Today article generated a subsequent controversy that both caused Wikipedia to tighten its standards for creating articles, especially articles about living people, and highlighted the growing importance of Wikipedia as a source of information.
During this period, Wikipedia also began to enter the popular culture. A prominent example was the Weird Al Yankovic parody song "White & Nerdy", which peaked at #9 on the Billboard Hot 100 in late 2006, and contained the lyric "I edit Wikipedia." In December 2006, Time magazine chose "You" as their Person of the Year, referring to the rise of Web 2.0 and web technologies that allow for user-contributed content, and cited Wikipedia as one of the key websites that allow for "community and collaboration on a scale never seen before".[42]
Other wiki websites, 2004–2006
[edit]In July 2004, OpenStreetMap, a website to create an open-source street map of the world using wiki functionality, was launched.
Most of the major parody-based wikis launched at around this time. Encyclopedia Dramatica, which mocked internet culture, was founded in December 2004 (it was shut down in April 2011). Stupidedia, a German-language wiki intended as a direct spoof of Wikipedia, was also founded in December 2004, a week later. Uncyclopedia, an English-language wiki also intended to parody Wikipedia, was founded the next month, in January 2005; it was later extended to dozens of other languages, and merged in other wikis, including Stupidedia. La Frikipedia, a Spanish-language parody of Wikipedia, was founded in October 2005.
In January 2005, the wiki wikiHow was created as a component of the how-to website eHow. In April 2006, its creators sold off eHow and focused full-time on wikiHow, which was launched at its own domain.[43] Both sites receive tens of millions of hits a month.
In April 2006, the company Internet Brands bought the sites WikiTravel and World66.[44] The purchase of WikiTravel by a commercial entity led to the creation of the non-commercial travel wiki Wikivoyage by some former WikiTravel authors and administrators in December 2006. Wikivoyage originally held content in German and Italian, and started as a fork of WikiTravel content; in 2013 it would become a WMF site.
Two major Chinese wiki encyclopedias began in 2006: Baidu Baike in April and Hudong in November. (Hudong's English-language name was changed to Baike.com in 2012.) Both currently hold millions of articles, and exceed the popularity of the Chinese Wikipedia within China. Both are for-profit wikis, that, unlike Wikipedia, hold the copyright to their own content.
Development of wiki software, 2004–2006
[edit]This period saw the launch of many wiki applications geared toward corporate/enterprise users. Such software was usually proprietary, and it usually included additional functionality, such as blogs and project management tools. Notable wiki software launched during this time includes:
- Trac (February 2004), created by Edgewall Software. Trac is an open source bug tracking and project management application, with wiki functionality.
- Confluence (March 2004), created by Atlassian.
- DokuWiki (July 2004), an open-source application intended for small companies' documentation needs.
- FlexWiki (September 2004), an open source application created by Microsoft.[45] Written primarily by David Ornstein, FlexWiki used .NET technology and had an integrated scripting language called WikiTalk (based on Smalltalk). It stored content in either text files or a SQL Server database.[46]
- JotSpot (October 2004), created by JotSpot, Inc. JotSpot was bought by Google in 2006 for an undisclosed amount; Google would later release the technology, in modified form, as Google Sites in 2008.
- SamePage (April 2006), created by eTouch Systems.
- Redmine (June 2006), an open source application similar to Trac.
- DekiWiki (July 2006), an open source application created by MindTouch, Inc.[47] It had started as a fork of MediaWiki, but was then significantly rewritten before its release. DekiWiki was later renamed to "Deki", then "MindTouch Core".
- Clearspace (December 2006), created by Jive Software. Clearspace was later renamed "Jive SBS", then "Jive Engage" and then Jive.
wikiCalc, a wiki-based spreadsheet application, was launched in November 2005 by spreadsheet pioneer Dan Bricklin.
In November 2006 Microsoft released Microsoft Office SharePoint Server 2007,[48] which included wiki and blog functionality for the first time.
In this period a number of semantic wiki applications were launched, including the currently most popular one, Semantic MediaWiki (which was first announced at Wikimania 2005).
Wikis enter the mainstream: 2007–2013
[edit]A milestone in public acceptance of wikis was reached in March 2007 when the word "wiki" entered the Oxford English Dictionary.[49]
In January 2007, Amazon.com released Amapedia, a product-review wiki on its own website; it was shut down in June 2010.
In 2007, ShoutWiki, a wiki farm, was founded as an alternative to Wikia. In the same year, Encyc was founded using PmWiki, and in 2008 it transitioned to MediaWiki. Another MediaWiki-based wiki farm, MyWikis, was launched in 2012.
In February 2007, Penguin Books launched a wiki to create the planned novel A Million Penguins, in a well-publicized experiment at creating a crowd-generated novel. The wiki was shut down a month later, not having created a coherent work.[50]
In March 2007, Larry Sanger, co-founder of Wikipedia, launched Citizendium, an "expert-guided" encyclopedia wiki requiring participants to use their real names.
In 2010, the site WikiLeaks (which had been founded in 2006) gained a great deal of both fame and notoriety as a result of a set of leaked documents the site published of classified materials from the United States government, notably footage of the July 12, 2007, Baghdad airstrike, the Afghan War documents leak and the United States diplomatic cables leak. The site was initially a true wiki editable by anyone, running on the MediaWiki application, though in 2010 had its permissions changed so that only its administrators could edit pages,[51] and later it moved off of MediaWiki and was no longer a wiki.
Wikipedia and other Wikimedia projects, 2007–2013
[edit]In 2007, Wikipedia entered the top 10 most popular websites in the world.[5] Wikipedia began to be heavily referenced in television and other media during 2007; see Wikipedia in culture.
In August 2008, then-U.S. presidential candidate John McCain was accused of plagiarizing Wikipedia in a speech about Georgia.[52] In June 2009, journalist Chris Anderson admitted to plagiarizing a series of Wikipedia articles in his book Free: The Future of a Radical Price. Anderson called it a "screwup", based on lack of clarity on how to cite a specific version of a Wikipedia article.[53]
January 2011 saw the 10th anniversary of the creation of Wikipedia, with a large round of coverage in the international media, most of it positive, including a cover story in German newspaper Die Zeit with the headline "The greatest work of mankind".[54]
In March 2012, a proposal was made to have the Wikimedia Foundation host a wiki containing travel advice,[55] which led to a discussion and vote lasting several months. In August 2012 the company Internet Brands, owners of the Wikitravel site, sued two of its contributors who had spoken out in favor of the proposal.[56] That lawsuit in turn led the Wikimedia Foundation to sue Internet Brands in September 2012 to not "impede, disrupt or block the creation of" such a site.[57] The community proposal to host a travel guide wiki was successful, and the decision was made to incorporate Wikivoyage, a Wikitravel spinoff site, as a project.
The Wikidata project, which had been discussed in various guises for some time, was officially launched on October 30, 2012.[58] Wikidata is a collaborative data repository meant to provide data that can be used by Wikipedia across all its languages.
January 14, 2013, saw the first official use of Wikidata: to enable the automatic display of "interlanguage links" within articles on the Hungarian Wikipedia.[59]
January 15, 2013, the 12th anniversary of Wikipedia's founding, saw the launch of the new, Wikimedia-hosted Wikivoyage.[60] The lawsuits regarding Wikivoyage between Wikimedia and Internet Brands were settled out of court the next month.[61]
Reuse of Wikipedia data
[edit]Several applications were developed during this time to make use of Wikipedia's data in order to improve both structured querying and natural-language searching. These included:
- DBpedia, launched in January 2007; a project to publish structured data from Wikipedia in machine-readable, queriable form. By 2008, it became a major component of the Linked Data initiative.[62]
- Freebase, launched by the company Metaweb in May 2007. Freebase stores information from Wikipedia and other sources in a structured, queriable format. Metaweb and Freebase were bought by Google in July 2010. The software later provided part of the technology behind the Google Knowledge Graph, launched in May 2012.
- Powerset, a company founded in 2005, in May 2008 released a search engine that used data from the English Wikipedia to answer natural-language queries. It was bought by Microsoft in July 2008, with the goal of improving results for the Bing search engine.
The Wikimedia Foundation project Wikidata is meant to provide a collaborative set of data that can make such querying easier, and available across all languages.
Wiki software, 2007–2013
[edit]This period saw consolidation in the set of wiki software being used. A large number of wiki engines ended development during this time, including FlexWiki, MojoMojo, UseModWiki and Zwiki.
In October 2008, most developers of TWiki left the project to work on a fork of the code, Foswiki, after creator Peter Thoeny took control of the TWiki code and trademark via his company, TWiki.net.[63]
In May 2009, Google announced its Google Wave platform (and its associated Federation protocol), which would combine the functionality of wikis with e-mail, instant messaging and social networking in order to provide a collaborative, real-time, server-hosted communication platform. Google Wave was released to the general public in May 2010, but development on it was ended several months later, in August 2010. In December 2010, Google transferred control of the software's development to the Apache Foundation, and it was renamed "Apache Wave".
In December 2010, wiki functionality was added to the SAP NetWeaver Portal application.[64]
In 2011, a wiki application, named Phriction, was added to the open-source collaboration suite Phabricator.
In April 2013, MindTouch Inc. abandoned its open-source product and moved to MindTouch being only a proprietary, hosted service.[47]
In July 2013, Wetpaint sold off its wiki service to WikiFoundry, having changed its business model to the hosting of professionally generated content.
Further consolidation and innovation: 2014–present
[edit]The period starting in 2014 saw a further decrease in wiki software offerings, as applications and hosted solutions including Apache Wave, MindTouch, SamePage, TWiki and Wikispaces stopped development. Conversely, a number of new proprietary hosted knowledge-management solutions, which include a wiki component, were launched, including Notion in 2016. The MediaWiki-based hosting site Miraheze launched in 2015.
The site Everipedia, a competitor to Wikipedia that stores its information on a blockchain, launched in 2015.
Wikia was gradually renamed to Fandom between 2016 and 2021.
See also
[edit]References
[edit]- ^ a b c d e f g h i j k l m n o "Wiki History". C2.com. Archived from the original on June 21, 2002. Retrieved October 8, 2010.
- ^ Andrews, Lorrin (1865). A dictionary of the Hawaiian language to which is appended an English-Hawaiian vocabulary and a chronological table of remarkable events (PDF). Henry M. Whitney. p. 514.
- ^ "Meatball Wiki: AboutThisSite". meatballwiki.org. Archived from the original on June 8, 2023. Retrieved June 24, 2023.
- ^ a b c d e f "Wikipedia : Definition, History, & Facts". Britannica.com. Archived from the original on January 21, 2022. Retrieved January 9, 2022.
- ^ a b Sabga, Chris (February 23, 2007). "Wikipedia Now Among Top 10 Most Popular Sites". infopackets. Archived from the original on October 24, 2010.
- ^ "Vannevar Bush, "As We May Think", July 1945". The Atlantic. September 9, 2010. Archived from the original on February 24, 2010. Retrieved October 8, 2010.
- ^ Ivanov, Kristo (1972). Quality-Control of Information (PhD). Stockholm: The Royal Institute of Technology (KTH). Archived from the original on November 17, 2011. (NTIS, # PB-219297.)
- ^ a b "Wiki Wiki Origin". C2.com. June 25, 2008. Archived from the original on May 14, 2011. Retrieved October 8, 2010.
- ^ a b "Wiki Wiki HyperCard". C2.com. June 11, 2008. Archived from the original on April 6, 2007. Retrieved October 8, 2010.
- ^ "The birth of the Web". CERN. Archived from the original on June 1, 2023. Retrieved June 24, 2023.
- ^ a b "New Service on PPR". C2.com. March 16, 1995. Archived from the original on October 13, 2007.
- ^ Cummings, Robert E. (September 8, 2009). "What Was a Wiki, and Why Do I Care? A Short and Usable History of Wikis". Wildwiki.net. Archived from the original on September 21, 2009.
- ^ Adams, Clifford (January 27, 2001). "Hello, "FreeLinks" work, and redirected pages". Wikipedia-l (Mailing list). Archived from the original on June 20, 2014.
- ^ Szybalski, Andy (March 14, 2005). "Why it's not a wiki world (yet)" (PDF). bigwhitebox.org.
- ^ "Thread Mode". C2.com. March 31, 2010. Archived from the original on October 19, 2010. Retrieved October 8, 2010.
- ^ "Thread Mess". C2.com. October 30, 2006. Archived from the original on May 14, 2011. Retrieved October 8, 2010.
- ^ a b "History Of Categories". C2.com. Archived from the original on June 4, 2011. Retrieved October 8, 2010.
- ^ "AboutCategoriesAndTopics". C2.com. August 27, 1996. Archived from the original on November 29, 1996. Retrieved October 8, 2010.
- ^ "Wiki Mind Wipe Discussion". C2.com. Archived from the original on August 24, 2010. Retrieved October 8, 2010.
- ^ "Wiki On Wiki". C2.com. August 31, 2005. Archived from the original on October 28, 2010. Retrieved October 8, 2010.
- ^ "Wiki Reductionists". C2.com. Archived from the original on May 15, 2011. Retrieved October 8, 2010.
- ^ Cunningham, Ward & Leuf, Bo (2001). The Wiki Way: Quick Collaboration on the Web. Boston: Addison-Wesley. ISBN 978-0-201714999.
- ^ "Wiki FAQ". Visual FoxPro Wiki. March 3, 1999. Archived from the original on June 14, 2011. Retrieved October 8, 2010.
- ^ "JWiki". C2.com. Archived from the original on May 22, 2010. Retrieved June 8, 2010.
- ^ "Jos Wiki". C2.com. December 25, 2004. Archived from the original on July 28, 2010. Retrieved October 8, 2010.
- ^ "Piki Piki". C2.com. October 22, 2009. Archived from the original on October 19, 2010. Retrieved October 8, 2010.
- ^ "Britannica or Nupedia? The Future of Free Encyclopedias". Kuro5hin.org. Archived from the original on November 16, 2011. Retrieved October 8, 2010.
- ^ "In Memoriam: September 11, 2001". sep11memories.org. Archived from the original on February 6, 2007. Retrieved February 6, 2007.
- ^ "First edit to the wiki". In Memoriam: September 11 wiki. October 28, 2002. Archived from the original on September 29, 2011.
- ^ "Progression of Wiki Organization". WikiNodes. Archived from the original on February 23, 2007.
- ^ "Inter Wiki". C2.com. Archived from the original on October 19, 2010. Retrieved October 8, 2010.
- ^ "Inter Wiki Map". C2.com. Archived from the original on February 2, 2010. Retrieved October 8, 2010.
- ^ "Meatball Wiki: MetaWiki". MeatballWiki. Retrieved October 8, 2010.
- ^ "Meatball Wiki: OneBigWiki". MeatballWiki. Archived from the original on September 17, 2009. Retrieved October 8, 2010.
- ^ "Switch Wiki". C2.com. April 24, 2007. Archived from the original on February 2, 2010. Retrieved October 8, 2010.
- ^ "WackoWiki". C2.com. March 31, 2025. Archived from the original on December 12, 2016. Retrieved March 31, 2025.
- ^ Klobas, Jane E. & Beesley, Angela (2006). Wikis: tools for information work and collaboration. Chandos. p. 46. ISBN 1-84334-179-4.
- ^ Bauder, Fred. "Wikinfo". wikinfo.org. Archived from the original on February 6, 2018. Retrieved June 13, 2017.
- ^ "WikiHome". Wiki.chongqed.org. Archived from the original on August 24, 2004. Retrieved October 8, 2010.
- ^ Wikipedia's Alexa ranking milestones (3 month average), Wikimedia Meta.
- ^ Zumpano, Anthony (January 29, 2007). "Similar Search Results: Google Wins". Brandchannel. Archived from the original on February 20, 2007.
- ^ Grossman, Lev (December 13, 2006). "Time's Person of the Year: You". Time. Archived from the original on December 18, 2006. Retrieved February 7, 2010.
- ^ Herrick, Jack (September 2006). "History of eHow and wikiHow". wikiHow. Archived from the original on January 3, 2007. Retrieved July 23, 2012.
- ^ "Internet Brands Acquires Wikitravel and World66 Online Travel Guides; Company Will Fund Growth of Two of the World's Fastest Growing Wiki Travel Sites (press release)". BusinessWire. April 20, 2006. Archived from the original on June 24, 2012. Retrieved July 23, 2012.
- ^ "FlexWiki: Microsofts Third Open Software Project". eWeek. September 28, 2004. Retrieved April 5, 2012.
- ^ Lee, Thomas (October 26, 2006). "A working wiki". PC Pro. Archived from the original on January 15, 2010. Retrieved April 5, 2012.
- ^ a b Fulkerson, Aaron (April 9, 2013). "MindTouch Core and Platform: 'This is the End, Beautiful Friend'". MindTouch. Archived from the original on August 8, 2013. Retrieved September 3, 2013.
- ^ Dunwoodie, Brice (November 7, 2006). "SharePoint (MOSS) and Office 2007 Released". CMSWire. Archived from the original on January 13, 2010. Retrieved February 16, 2010.
- ^ Haines, Lester (March 16, 2007). "Wiki elevated to Oxford English Dictionary". The Register. Archived from the original on August 10, 2017.
- ^ Mason, Bruce & Thomas, Sue (April 24, 2008). "A Million Penguins Research Report" (PDF). De Montfort University. Archived (PDF) from the original on December 5, 2014. Retrieved December 15, 2014.
- ^ Gilson, Dave (May 19, 2010). "WikiLeaks Gets A Facelift". Mother Jones. Archived from the original on April 29, 2014. Retrieved June 17, 2010.
- ^ "McCain faces accusations of Wikipedia plagiarism". The Calgary Herald. August 19, 2008. Archived from the original on May 10, 2011.
- ^ Jaquith, Waldo (June 23, 2009). "Chris Anderson's Free Contains Apparent Plagiarism". The Virginia Quarterly Review. Archived from the original on July 7, 2009. Retrieved July 7, 2009.
- ^ "Photo of January 12, 2011 Die Zeit cover". TwitPic. Archived from the original on October 15, 2012.
- ^ Heilman, James (March 11, 2012). "Earliest edit to page now called "Wikivoyage"". Wikimedia Meta-Wiki. Archived from the original on July 11, 2016.
- ^ Cohen, Noam (September 9, 2012). "Travel Site Built on Wiki Ethos Now Bedevils Its Owner". The New York Times. Archived from the original on November 14, 2012.
- ^ Kay, Kelly (September 5, 2012). "Wikimedia Foundation seeks declaratory relief in response to legal threats from Internet Brands". Wikimedia Blog. Archived from the original on November 8, 2012.
- ^ Pintscher, Lydia (October 30, 2012). "wikidata.org is live (with some caveats)". wikidata-l (Mailing list). Archived from the original on June 17, 2014. Retrieved November 3, 2012.
- ^ Pintscher, Lydia (January 14, 2013). "First steps of Wikidata in the Hungarian Wikipedia". Wikimedia Deutschland Blog. Archived from the original on December 14, 2015. Retrieved January 15, 2013.
- ^ "Wikimedia Foundation launches Wikivoyage, a free, worldwide travel guide that anyone can edit". Wikimedia Foundation. Archived from the original on January 17, 2013. Retrieved January 15, 2012.
- ^ Musil, Steven (February 17, 2013). "Wikimedia, Internet Brands settle Wikivoyage lawsuits". CNET. Archived from the original on December 27, 2013. Retrieved February 19, 2013.
- ^ Miller, Paul (February 7, 2008). "Sir Tim Berners-Lee Talks with Talis about the Semantic Web". talis-podcasts. Archived from the original on May 10, 2013. (Transcript of interview.)
- ^ Asay, Matt (October 29, 2008). "TWiki's hunt for cash fractures its community". CNET News. Archived from the original on December 1, 2010.
- ^ "SAP Netweaver Portal 7.3 ramp-up released". Ameya's Blog. December 3, 2010. Archived from the original on July 30, 2012.
History of wikis
View on GrokipediaHypertext and Collaborative Editing Precursors
Pre-Web Hypertext Systems
The concept of hypertext, involving nonlinear linking of text and media, originated in the 1960s as a means to create interconnected knowledge structures beyond linear documents. Ted Nelson coined the term "hypertext" in 1963 and formalized it in his 1965 publication Literary Machines, envisioning systems where content could be dynamically referenced and reused across documents.[8] Project Xanadu, initiated by Nelson in 1960, proposed a global hypertext repository with "transclusion"—a mechanism for embedding live, versioned excerpts from source documents rather than static copies—to enable precise attribution and prevent duplication while maintaining causal links to originals.[9] Despite its theoretical influence on ideas of persistent, royalty-tracked content sharing, Xanadu remained largely unimplemented due to technical complexities in versioning and micropayments, with prototypes emerging only decades later.[10] Douglas Engelbart's oN-Line System (NLS), developed at the Stanford Research Institute starting in the early 1960s, represented an early practical implementation of hypertext linking in a computer-based environment. NLS allowed users to create and navigate structured documents with explicit links between elements, such as jumping between related journal entries or outlines, using a chorded keyboard and the newly invented mouse for manipulation.[11] In its December 9, 1968, public demonstration at the Fall Joint Computer Conference—known as the "Mother of All Demos"—Engelbart showcased NLS's capabilities for on-screen document linking and dynamic views, enabling shared access to evolving knowledge bases over networked terminals.[12] These features laid foundational techniques for associative information retrieval, though NLS emphasized individual augmentation over widespread editable collaboration, limited by its mainframe-centric architecture and specialized input devices.[13] Apple's HyperCard, released on August 11, 1987, democratized hypermedia for personal computing by providing an intuitive authoring tool for Macintosh users. Created by Bill Atkinson, it organized information into navigable "stacks" of virtual cards connected via buttons and hyperlinks, supporting embedded text, images, and scripts for simple interactivity.[14] HyperCard's card-based metaphor facilitated non-technical users in building personal knowledge networks, such as custom databases or tutorials, with over 100,000 stacks distributed by the early 1990s.[15] Its ease of use influenced subsequent interface designs for linked content exploration, though it operated in a standalone, single-user mode without inherent network distribution.[16]World Wide Web and Early Collaborative Tools
Tim Berners-Lee, a researcher at CERN, submitted the initial proposal for the World Wide Web on March 12, 1989, outlining a system for managing hypertext-based information sharing among scientists.[17] This was revised in May 1990, introducing Hypertext Markup Language (HTML) as a simple markup for creating linked documents, with the tag enabling hyperlinks to connect pages across distributed servers.[18] The first web server and browser, both named WorldWideWeb, went online in 1991, allowing users to navigate interconnected static HTML pages via Hypertext Transfer Protocol (HTTP), but without mechanisms for direct user modification of content.[19] Pre-web systems like Usenet, launched in 1979 by Duke University students Tom Truscott and Jim Ellis, provided early models of distributed, communal content updates through threaded discussions in newsgroups, where participants could post and reply to messages propagated across networked servers.[20] Usenet's decentralized structure foreshadowed web-era forums by enabling collective knowledge building via asynchronous contributions, though it relied on non-hypertext protocols like NNTP and lacked seamless integration with graphical browsing.[21] On the emerging web, rudimentary collaborative elements appeared in the form of static discussion archives or early email-linked feedback, but these were not native to HTTP and required manual aggregation by site administrators.[22] Early web pages, constrained to static HTML until the introduction of Common Gateway Interface (CGI) scripts around 1993, imposed significant limitations on dynamic collaboration: content updates demanded server-side file edits by privileged users, precluding broad, real-time communal editing.[23] Hyperlinks facilitated navigation and reference but not alteration, resulting in read-only experiences that hindered the evolution of shared, evolving knowledge bases akin to later wikis.[24] These constraints persisted through 1994, as browsers like NCSA Mosaic emphasized display over interactivity, underscoring the gap between hyperlinked information retrieval and editable, collaborative hypertext systems.[18]Invention of the Wiki Concept
Ward Cunningham's WikiWikiWeb Launch
Ward Cunningham initiated development of the first wiki in 1994 to address the need for a simple, collaborative platform supporting the Portland Pattern Repository, a project aimed at documenting and sharing software design patterns among programmers.[4] The repository sought to enable rapid, unstructured discussions and contributions without the barriers of formal editing processes prevalent in early web forums.[25] This motivation stemmed from Cunningham's experience in software engineering, where he recognized the value of lightweight, version-controlled hypertext for pattern-sharing in professional communities.[26] The wiki software, termed "Wiki Base," was implemented using Perl and deployed on the c2.com domain, registered by Cunningham's firm, Cunningham & Cunningham, on October 23, 1994.[4] Public launch occurred on March 25, 1995, marking the debut of WikiWikiWeb as an open, editable knowledge base integrated with the pattern repository site.[27] This deployment emphasized accessibility, allowing any visitor to contribute or modify content directly via web forms, fostering emergent organization through community participation.[26] Cunningham named the system "WikiWikiWeb" after the Hawaiian phrase "wiki wiki," signifying "very quick" or "hurried," drawn from signage for the Wiki Wiki Shuttle bus at Honolulu International Airport.[28] The term encapsulated the design philosophy of speed in editing and iteration, prioritizing ease and immediacy over polished structure to encourage frequent, small-scale updates by users.[28] This naming choice reflected Cunningham's intent to create a tool that minimized friction in collaborative knowledge building, distinct from static web pages or rigid content management systems of the era.[4]Core Technical Innovations
The WikiWikiWeb, launched by Ward Cunningham on March 25, 1995, introduced several technical mechanisms that distinguished it from prior hypertext systems by enabling barrier-free, real-time collaborative authoring directly via web browsers.[29] Central to this was the automation of internal hyperlinks through CamelCase (also termed WikiCase), where sequences of capitalized words—such as PortlandPatternRepository—were programmatically detected and converted into clickable links to new or existing pages without necessitating manual markup, HTML tags, or predefined link structures.[30] This innovation leveraged pattern matching in the server-side Perl script to parse page text and generate dynamic navigation, reducing the cognitive and technical load for contributors compared to earlier systems requiring explicit link coding.[31] Editing interfaces featured prominent "Edit this page" links atop every viewable page, allowing immediate access to a plain-text editor for in-line modifications, previews, and saves without login prompts or client-side tools.[1] This design supported atomic updates, where changes propagated instantly upon submission, fostering fluid content evolution among dispersed users. Complementing this was built-in version control, which automatically archived each revision with timestamps and author notes (often pseudonymous), enabling reversion via diff views to restore prior states and undo erroneous or malicious alterations.[32] The system deliberately omitted user registration and authentication requirements, enforcing open anonymity to prioritize accessibility over access controls; edits were attributable only by optional signatures or IP tracking, which lowered entry thresholds but relied on communal self-correction for integrity.[32] These features, implemented in a lightweight CGI-based architecture, emphasized simplicity and trust in collective maintenance, setting the template for subsequent wiki engines despite exposing pages to potential disruption.[33]Early Adoption in Niche Communities
Software Development and Programming Wikis
The WikiWikiWeb found initial application among software developers for documenting design patterns and collaborative practices in extreme programming, methodologies pioneered by Ward Cunningham and associates like Kent Beck.[1] Launched in March 1995 on c2.com as a supplement to the Portland Pattern Repository, it enabled programmers to interlink discussions on code structures, project experiences, and iterative development techniques through simple edit links and version histories.[4] This niche use emphasized wikis as lightweight tools for maintaining living documentation of programming heuristics, distinct from static code repositories or formal manuals. From 1995 to 2000, the platform's content grew from initial pattern entries to encompass extreme programming debates, with features like RecentVisitors alerts (added 1995) and ThreadMode threading (1996) supporting real-time engineer interactions.[4] Adoption remained restricted to Cunningham's professional network and affiliated communities, such as those exploring agile methods, with page volume doubling annually to about 60 MB by 2000 yet attracting minimal attention beyond specialist circles.[4] Tensions over content focus, including extreme programming advocacy, prompted internal zoning like XpFreeZone tags, underscoring its role in fostering targeted technical discourse rather than broad dissemination.[4] In 1998, Peter Thoeny created TWiki in Perl, drawing from Ward Cunningham's model and JosWiki, to address documentation needs in corporate software environments.[34] Targeted at enterprise intranets, TWiki supported issue tracking, bug documentation, and project wikis for teams, as seen in early uses like Javapedia for Java developer resources.[34] This extension highlighted wikis' adaptability for structured code-related knowledge management in professional settings, though uptake stayed limited to tech firms and remained obscure outside programming subcultures through 2000.[34]Initial Forks and Community Expansions
As participation in the WikiWikiWeb grew beyond its original focus on software design patterns, contributors encountered challenges in balancing technical documentation with broader discussions on collaboration practices, prompting the emergence of dedicated sister sites. In 2000, Sunir Shah established MeatballWiki using Clifford Adams's UseMod software, explicitly intended for documenting and debating wiki methodology, culture, and community dynamics rather than pattern-specific content.[4] This fork addressed accumulating tensions over page edits and content direction on the original site, where open editing sometimes led to disputes between pattern enthusiasts and those advocating for meta-level explorations of wiki principles.[35] MeatballWiki's creation exemplified early community-driven divergences, preserving the WikiWikiWeb's pattern repository while providing an outlet for philosophical and operational critiques without centralized moderation.[36] Parallel to these content forks, the release of open-source wiki engines accelerated technical expansions and independent installations. PhpWiki, developed by Steve Wainstead, marked the first publicly available wiki implementation in PHP with its debut in December 1999, enabling simpler hosting on standard web servers compared to the original Perl-based system.[37] This portability spurred community adaptations, as developers forked and modified the codebase to suit varied hosting environments, reducing barriers to entry for non-experts. By mid-2000, such engines facilitated the proliferation of standalone wikis for niche groups, including initial experiments in personal knowledge organization outside programming domains, where users adapted the hyperlinked, editable format for individual note-taking and ad-hoc information synthesis.[4] These developments underscored the wiki model's inherent forking potential, driven by the absence of proprietary locks and the emphasis on copyleft-like sharing of both content and code. Early adopters leveraged engines like PhpWiki to create private or semi-public sites for knowledge curation, marking a shift from collective software repositories to versatile tools for decentralized information management, though adoption remained limited to tech-savvy circles pre-2001.[37] Conflicts over edit histories and site philosophies in forks like MeatballWiki highlighted causal trade-offs in unrestricted collaboration: enhanced innovation through divergence, tempered by fragmentation risks, without yet scaling to mass audiences.[35]Philosophical Underpinnings and Early Critiques
Ward Cunningham's invention of the wiki in 1995 was grounded in a philosophy of radical trust, positing that collaborative editing could thrive through an assumption of good faith among participants, obviating the need for stringent verification or credentials prior to contributions. This ethos enabled the WikiWikiWeb, initially serving as a dynamic knowledge base for software pattern discussions among a small group of programmers, where emergent consensus would refine content over time. Cunningham emphasized creating an environment fostering human connection in digital spaces, allowing strangers to build trust sufficient for co-creating meaningful artifacts without centralized control.[38][3][39] This trust-based model starkly contrasted with established hierarchical systems of knowledge production, such as peer-reviewed publications or curated databases, which prioritized expert authority and top-down validation to ensure accuracy. By democratizing edits via a simple "edit" button on every page, wikis embodied a first-come, collective refinement process, betting on the collective intelligence of motivated contributors to surpass individual gatekeepers in agility and coverage. Cunningham's approach drew from agile software principles he co-developed, applying iterative, incremental improvements to information management itself.[40][41] Even in its nascent phase within niche communities, the open-editing paradigm elicited early skepticism regarding its robustness against misuse, particularly vandalism that could degrade content quality absent proactive moderation. Observers noted that while reversion tools mitigated damage in small, cohesive groups familiar with participants, the lack of barriers risked proliferation of errors or deliberate disruptions in less controlled settings. These concerns underscored doubts about the model's scalability for diverse, anonymous users, prefiguring broader debates on reliability in expansive, unvetted collaborative endeavors, though empirical issues remained minimal in the initial programmer-centric wikis.[42][25]Wikipedia's Emergence and Acceleration
Transition from Nupedia
Nupedia, an expert-driven online encyclopedia launched on March 9, 2000, by Jimmy Wales and Larry Sanger under Bomis, Inc., employed a multi-stage peer-review process requiring assignment to authors, expert drafting, internal review, external review, copyediting, final approval, and release, which constrained production to approximately 12 completed articles in its first year.[43][44] This methodical approach, modeled on traditional encyclopedias, prioritized quality through credentialed contributors but yielded slow progress, with only around two dozen articles fully approved by early 2001 despite recruitment efforts.[45] The turning point came on January 2, 2001, when Sanger, during a conversation with software developer Ben Kovitz, learned of wiki software's collaborative potential; inspired, he emailed Wales that evening proposing a wiki side project to Nupedia for unrestricted drafting of articles that could later feed into the peer-review pipeline.[46][47] Wales authorized the initiative, leading to Wikipedia's launch on January 10, 2001, via UseMod wiki software, with the first edit to its homepage occurring on January 15.[48] This marked a deliberate pivot to open editing, where anyone could contribute without credentials or approval, contrasting Nupedia's gated model and aiming to accelerate content generation through emergent community self-correction. Wikipedia's permissive structure triggered immediate expansion: starting from zero articles, it amassed dozens within weeks, then surged to hundreds by summer and thousands by year's end, vastly outpacing Nupedia's output and validating the wiki's efficacy for rapid encyclopedic accumulation despite initial skepticism over unvetted contributions.[49] This growth stemmed from viral promotion on mailing lists and early adopter enthusiasm, shifting focus from elite authorship to decentralized participation while retaining Nupedia as a quality benchmark—though the wiki soon eclipsed its predecessor in scale and momentum.[6]Wikimedia Foundation and Sister Projects
In December 2002, Wiktionary launched as the initial sister project to Wikipedia, functioning as a multilingual, collaborative dictionary compiling definitions, etymologies, and translations edited by volunteers.[50] This extension built on wiki principles to aggregate lexical data beyond encyclopedic content, with initial development proposed by community members including Daniel Alston.[50] By early 2002, Wikipedia had migrated from the Perl-based UseModWiki software to MediaWiki, a new open-source platform written in PHP and MySQL, originally developed by Magnus Manske to handle increasing edit volumes and introduce features like enhanced searching and revision tracking.[51] This shift to MediaWiki, first deployed on January 25, 2002, enabled better performance for distributed editing without requiring proprietary tools.[51] The Wikimedia Foundation was incorporated on June 20, 2003, in St. Petersburg, Florida, as a 501(c)(3) nonprofit by Jimmy Wales to centralize legal liability, secure funding through donations, and manage infrastructure for Wikipedia alongside nascent sister initiatives.[52] Prior to this, projects operated informally under Bomis, Inc., but the foundation's structure facilitated tax-exempt operations and resource allocation independent of commercial interests.[53] Wikibooks debuted on July 10, 2003, under the foundation's umbrella as a repository for open-licensed textbooks and instructional modules across subjects, editable by contributors to foster modular educational content.[54] This project emphasized structured, hierarchical organization distinct from Wikipedia's article format, targeting self-paced learning materials.[54]Rapid Article Growth and Media Recognition
The English Wikipedia, launched on January 15, 2001, demonstrated remarkable growth through volunteer contributions, accumulating 200 articles within the first month and expanding to approximately 18,000 by early 2002.[55] This pace reflected the advantages of crowd-sourcing, where unrestricted editing by non-experts accelerated article creation far beyond the timelines of peer-reviewed projects like Nupedia, which produced only 12 articles in its first year despite expert oversight.[55] By 2003, the English edition had exceeded 100,000 articles, underscoring the empirical efficiency of distributed volunteer labor in scaling encyclopedic content. Parallel to this expansion, multilingual editions proliferated, with the German version launching on March 16, 2001, followed by French on May 23, 2001, and others including Swedish, Dutch, and Norwegian in the same period.[56] These early language variants grew through analogous volunteer-driven processes, fostering a global network of interdependent wikis that collectively amplified content velocity and diversity. Early media attention validated the model's viability, beginning with tech community outlets like Slashdot, which featured discussions on Wikipedia's free-content approach alongside Britannica in July 2001. Mainstream recognition followed, including a New York Times article on September 20, 2001, highlighting the project's collaborative innovation and rapid maturation. This coverage attracted further volunteers, reinforcing the self-sustaining cycle of growth via publicized empirical successes in open knowledge production.Peak Expansion and Diversification
Proliferation of Specialized Wikis
In the mid-2000s, the wiki model diversified beyond general encyclopedias through the emergence of wiki farms and domain-specific sites, enabling communities to document niche topics like fandoms, hobbies, and technical interests. Wiki farms, platforms hosting multiple independent wikis, facilitated this proliferation by lowering technical barriers to creation and maintenance. In October 2004, Wikicities—later rebranded as Wikia—was launched as a for-profit service co-founded by Jimmy Wales and Angela Beesley, specifically designed to host specialized wikis on subjects often deemed unsuitable for Wikipedia's neutral point of view policy, such as fan fiction, gaming guides, and entertainment trivia.[57][58] Domain-specific wikis exemplified this trend, with communities building comprehensive repositories for targeted audiences. Memory Alpha, focused on canonical details from the Star Trek franchise, initiated setup on November 11, 2003, and rapidly grew as a collaborative canon reference, attracting dedicated contributors uninterested in broader encyclopedic constraints.[59] Similar efforts included hobbyist sites for media franchises and subcultures, where editors prioritized depth over verifiability standards applied in general wikis. This period marked a shift toward user-driven specialization, with wiki farms like Wikia supporting international and vernacular content for localized interests, such as European gaming communities or Asian pop culture databases, thus complementing rather than competing with dominant platforms. By 2006, these initiatives had spawned hundreds of active niche wikis, underscoring the wiki's adaptability for passionate, non-academic collaboration while highlighting tensions over content moderation and commercial viability in for-profit models.[60]Wiki Software Ecosystem Maturation
Following the deployment of MediaWiki for Wikipedia in 2002, the software achieved dominance in the wiki ecosystem through targeted releases and infrastructure upgrades in 2003. In June 2003, developers implemented database and web server separation on a second server, alongside filesystem-based caching for anonymous users to boost performance under load.[51] The inaugural official release occurred in August 2003, stabilizing the PHP-based core with features like interwiki links and side-by-side diffs for revision comparison.[51] By July 2003, additions such as auto-generated tables of contents and section-level editing further matured the editing interface.[51] Alternatives to MediaWiki's database-reliant model emerged to address diverse needs, exemplified by DokuWiki's initial release in July 2004. Developed by Andreas Gohr, it stored content in plain text files without a backend database, prioritizing simplicity and portability for smaller-scale or resource-constrained environments.[61] Extensions for advanced functionality proliferated, with Semantic MediaWiki's first version (0.1) released in 2005 following a foundational proposal in August 2005. This extension enabled semantic annotations, typed links, and query capabilities, transforming wikis into structured knowledge bases queryable via SPARQL-like syntax.[62] Concurrently, version control mechanisms evolved with refinements to diff algorithms and rollback tools, leveraging MediaWiki's inherent revision history to facilitate precise change tracking across large corpora.[51] Open-source contributions from volunteer developers, coordinated by leads like Brooke Vibber from 2003 onward, enhanced scalability through iterative optimizations in caching layers and query handling, preparing the software for corpora exceeding millions of pages.[51][63] These efforts emphasized modular extensibility via hooks and plugins, fostering a robust ecosystem adaptable to high-volume operations.[64]Onset of Scalability Challenges
As Wikipedia's English edition expanded from approximately 100,000 articles in early 2004 to over 1 million by late 2006, the open-editing model began revealing inherent scalability limits tied to its reliance on volunteer moderation and decentralized consensus.[65] The surge in contributions amplified low-quality inputs, with vandalism—defined as deliberate deleterious edits—estimated at 5-7% of total revisions during this period, necessitating the development of manual patrol systems to monitor recent changes and revert harmful alterations.[66] Reversion logs from these years documented thousands of daily undo operations, particularly on high-visibility pages, as the volume outpaced the capacity of human overseers to detect and correct issues in real time.[67] Edit wars, characterized by repeated contentious reversions, intensified on politically charged topics amid heightened public interest, such as the 2004 U.S. presidential election. The George W. Bush article, for instance, amassed over 5,500 edits in 2004 alone, many involving disputes over biographical details and policy interpretations, reflecting how the model's permissionless structure invited cycles of ideological tug-of-war without built-in arbitration thresholds.[68] Similar patterns emerged on pages like John Kerry's, where rival factions engaged in protracted reversions, underscoring a causal vulnerability: as edit volume scaled quadratically with visibility, resolution relied on ad hoc community intervention rather than scalable mechanisms, leading to temporary content instability.[69] Coverage disparities further highlighted the model's uneven scaling, with empirical data showing a power-law distribution where popular subjects garnered disproportionate attention and depth, while obscure topics received sparse, unverified stubs. By 2005, biographies and current events dominated edit activity—evident in the top-edited articles' focus on figures like Bush—leaving niche historical or scientific entries with minimal revisions and persistent gaps, as volunteer interest correlated inversely with topic obscurity.[70] This imbalance stemmed from the absence of incentives for comprehensive curation, resulting in a corpus skewed toward culturally salient but causally overrepresented domains, with analyses confirming geographic and thematic biases favoring North American and Western popular culture over global or specialized knowledge.[71]Enterprise Adoption and Institutional Integration
Commercial and Corporate Wiki Implementations
Atlassian's Confluence, released on March 25, 2004, emerged as a proprietary wiki-based collaboration platform tailored for enterprise knowledge management, featuring structured pages, version history, and user permissions to facilitate controlled content creation within teams.[72] By 2007, it had gained mainstream adoption among corporations seeking alternatives to open wikis, with integrations for tools like Jira enabling seamless documentation of workflows and project insights, amassing thousands of deployments as companies prioritized auditable edits over unrestricted access.[73] This period marked Confluence's evolution into a scalable solution for internal intranets, where space-level restrictions prevented broad edit rights, mitigating risks inherent in public wiki models such as vandalism or inconsistent quality.[74] Microsoft integrated wiki functionalities into SharePoint Server 2007, allowing organizations to deploy enterprise wikis as site collections with built-in libraries for linked pages and metadata, embedded within broader document management ecosystems.[75] These implementations emphasized permissioned editing through Active Directory synchronization and role-based controls, enabling granular access—such as read-only views for executives or edit rights for contributors—directly addressing open wikis' vulnerabilities to unverified changes in sensitive corporate contexts.[76] From 2008 to 2013, enhancements in SharePoint 2010 and 2013 further refined these features with improved search and workflow automation, supporting compliance requirements like audit trails in regulated industries.[77] This era's proprietary adaptations collectively shifted enterprise wiki paradigms toward restricted collaboration, driven by needs for data security, intellectual property protection, and accountability, as open editing proved untenable for proprietary information amid rising cyber threats and regulatory scrutiny.[74] Vendors like Atlassian and Microsoft marketed these systems as fortified alternatives, with Confluence reporting enterprise-scale usage for over 100,000 organizations by 2013, underscoring a preference for vetted contributors to maintain content integrity over anonymous contributions.[72]Refinements in Access Controls and Features
During the late 2000s, MediaWiki introduced the FlaggedRevs extension on June 4, 2008, enabling reviewers to flag stable versions of articles for public display while pending changes underwent verification, thereby enhancing content stability against rapid alterations.[78] This feature saw deployment trials on English Wikipedia starting June 14, 2010, as pending changes, allowing deferred review to mitigate unvetted edits without halting all contributions.[79] Complementing this, semi-protection mechanisms, which restrict editing to autoconfirmed users (those with accounts at least four days old and ten edits), were routinely applied to high-vandalism pages; empirical analyses indicate such protections curb vandalism by safeguarding existing content while preserving editorial access for established contributors.[80] Anti-spam measures advanced through the ConfirmEdit extension, which integrates CAPTCHA challenges—such as image recognition or simple math questions—to block automated bots during account creation and edits, a staple refinement deployed across Wikimedia sites to filter malicious inputs without broadly impeding human users.[81] By 2013, access controls matured with the rollout of OAuth support on Wikimedia wikis in November, permitting secure third-party applications to interact with user accounts via delegated permissions, obviating password sharing and bolstering API-driven integrations.[82] Usability features expanded concurrently, including native mobile applications: the official Android app launched in February 2012, followed by an iOS redesign in April, optimizing offline reading, search, and lightweight editing for portable devices.[83] MediaWiki's Action API underwent iterative enhancements across versions 1.10 (December 2007) to 1.21 (November 2012), adding modules for querying revisions, users, and categories, which facilitated third-party tools for data extraction and automation while maintaining granular permission controls.[84] These developments collectively hardened wikis against abuse while broadening programmatic and mobile accessibility.Intensifying Debates on Reliability
In the years following the 2005 Nature study, which identified an average of four errors per Wikipedia science article compared to three in Encyclopædia Britannica, researchers and critics revisited these findings amid growing article volume, uncovering persistent inaccuracies and methodological limitations in the original comparison.[65] Analyses highlighted that the study's expert reviewers frequently noted omissions, misleading presentations, and uncounted minor errors beyond the tallied factual mistakes, raising questions about Wikipedia's overall verifiability in technical domains.[85] By 2008, examinations of specific content areas, such as historical topics, confirmed substantive flaws; for instance, a review of nine U.S. history articles found inaccuracies in eight Wikipedia entries, including major distortions in two cases where open editing amplified unvetted claims over established scholarship. Concerns extended to ideological imbalances, with editor self-reports and contribution patterns indicating a left-leaning demographic that correlated with underrepresentation of conservative perspectives in politically sensitive articles.[86] This skew, evident in surveys of active contributors showing disproportionate progressive affiliations, manifested in content disputes where alternative viewpoints faced higher scrutiny or reversion rates.[87] Such patterns fueled debates on systemic bias, as empirical audits of political entries revealed selective sourcing and framing that privileged certain narratives, undermining claims of neutral point of view. Quantitative analyses of editing dynamics further underscored reliability challenges, particularly in politics and history. A 2011 study by Yasseri et al. developed an algorithm to detect "edit wars"—repeated reverts exceeding four cycles—and applied it to millions of revisions across six Wikipedia language editions from 2007 onward, finding conflict rates up to ten times higher in contentious categories like politics (e.g., articles on elections or ideologies) and history (e.g., war narratives or national events).[88] These wars, often involving a small cadre of persistent editors rather than broad consensus, persisted without resolution in over 20% of cases, highlighting causal vulnerabilities: anonymous contributions and low barriers to entry enabled factual distortions and ideological entrenchment, eroding trust in disputed topics.[89]Modern Stagnation, Innovations, and Alternatives
Declining Editor Engagement
Active editor participation on the English Wikipedia, defined as registered users making at least five edits per month, peaked at more than 51,000 in 2007 before entering a sustained decline that persisted through the 2010s and into the 2020s, stabilizing at approximately 30,000 by the mid-2010s and remaining at similar levels thereafter.[90] [91] This post-2014 stagnation reflects broader retention challenges, with newcomer survival rates dropping sharply; empirical survival analysis indicates that only 12-15% of new editors remain active after one year, a figure that held steady or worsened amid efforts to combat vandalism and maintain quality.[92] [93] Causal factors include structural barriers introduced in response to rapid growth, such as automated abuse filters, CAPTCHA requirements, and heightened revert rates for novice contributions, which reject up to 75% of first edits from newcomers—often good-faith attempts—leading to demotivation and early exit.[94] [95] These measures, while effective against spam, inadvertently prioritize existing content protection over influx, reducing overall editor productivity by limiting fresh participation.[96] Verification hurdles, including mandatory reliable sourcing and draft review processes implemented around 2013-2014, further deter entry-level contributors by imposing expert-level standards prematurely, exacerbating the feedback loop of low retention. Burnout among veteran editors compounds the issue, driven by chronic administrative burdens like dispute resolution, policy enforcement, and endless maintenance against reverting bots and aging content, with studies showing revert-heavy environments correlating with higher attrition even for experienced users.[97] [93] Harassment policies, intended to foster civility, have also contributed to barriers through escalated oversight and sanctions, though empirical data attributes primary decline to systemic quality controls rather than abuse alone; online harassment incidents, while documented as repelling some editors, affect a minority and lack causal primacy in aggregate participation metrics.[98] [99] In contrast to wiki models reliant on unbounded volunteerism, which foster burnout via infinite scalability demands, the Encyclopædia Britannica's expert-staffed system sustains engagement through compensated, finite roles, avoiding mass attrition by design—though direct empirical comparisons on retention remain limited, the volunteer model's vulnerabilities are evident in Wikipedia's stalled growth.[90]Incremental Technical Updates
In January 2023, the Wikimedia Foundation rolled out a redesigned desktop interface for Wikipedia, marking the first significant update since 2014 and addressing a decade of relative stasis in visual and navigational elements. The changes emphasized usability through a minimalist layout, expanded whitespace to delineate sections, larger typography for better readability, and simplified menus to reduce cognitive load, particularly for new users accustomed to modern web standards.[100][101] Wikidata's technical integration within MediaWiki ecosystems advanced incrementally post-2014, building on its 2012 launch to support more robust structured data handling. Key evolutions included refinements to the query service infrastructure, shifting from early HTTP API polling mechanisms—initially implemented around 2014–2015—to optimized updaters enabling faster synchronization of edits and enhanced SPARQL querying for dynamic data reuse across wiki articles, such as automated infobox population.[102][103] MediaWiki's mobile capabilities received targeted optimizations, with the MobileFrontend extension undergoing continuous refinements to bolster responsive design, editing tools, and performance on devices, including adaptations for touch interfaces and reduced data usage. Concurrently, bot frameworks for site maintenance evolved to incorporate API compatibility updates—such as those responding to 2015 changes—and coordination protocols to minimize inter-bot conflicts during tasks like link verification and categorization, ensuring sustained operational efficiency without overhauls.[104][105]Rise of Decentralized and AI-Influenced Models
In response to the centralization vulnerabilities of traditional wikis, such as single points of failure and susceptibility to coordinated edits or censorship, developers began exploring decentralized architectures post-2014, leveraging blockchain and peer-to-peer (P2P) technologies to enable immutable records and distributed control.[106] Blockchain-based projects like WikiChain, proposed around 2021, aimed to create tamper-resistant edit histories by storing contributions on a distributed ledger, theoretically mitigating issues like revisionism in centralized systems while allowing collaborative verification through consensus mechanisms.[107] These models promised greater resilience against administrative overreach but faced challenges in scalability and user adoption, as blockchain overhead often slowed editing compared to conventional databases.[106] Federated and P2P wiki systems further advanced decentralization by distributing authority across independent nodes, reducing reliance on a central server. P2Pedia, introduced in a 2014 framework and refined in subsequent research, enabled direct peer synchronization of wiki pages, accommodating diverse viewpoints without a governing body imposing consistency rules, though it struggled with conflict resolution in replicated data.[108] More recent efforts, such as Ibis launched in March 2024, adopted ActivityPub federation protocols—originally for social networks—to interconnect autonomous wiki instances, allowing articles to propagate across servers while preserving local moderation and versioning to counter centralized biases.[109] These P2P approaches addressed scalability flaws in monolithic wikis by enabling organic growth through voluntary federation, yet empirical tests revealed ongoing hurdles like eventual consistency delays and fragmented user bases.[110] The rise of large language models (LLMs) amplified scrutiny of centralized wikis' open-access model, exposing exploitation risks from unchecked data scraping that fueled AI training without reciprocal contributions or attribution. By 2023, surges in automated LLM crawlers increased Wikipedia's bandwidth costs by orders of magnitude, straining nonprofit infrastructure while enabling proprietary models to ingest public knowledge en masse, often amplifying biases or inaccuracies from source data without accountability.[111] Analyses from 2025 highlighted how such scraping undermined the sustainability of open wikis, prompting hybrid AI-influenced decentralized proposals that integrate LLMs for automated verification or generation under blockchain immutability to reclaim value from exploited datasets.[112] However, these experiments remain nascent, with risks of AI hallucinations persisting unless tethered to verifiable, distributed ledgers, underscoring a causal shift toward models prioritizing provenance over unchecked openness.[112]References
- http://meatballwiki.org/wiki/WikiHistory
- https://www.mediawiki.org/wiki/MediaWiki_history
- https://meta.wikimedia.org/wiki/Multilingual_Wikipedia
- https://www.semantic-mediawiki.org/wiki/Semantic_MediaWiki_Version_History
- https://www.mediawiki.org/wiki/Extension:FlaggedRevs
- https://www.mediawiki.org/wiki/Extension:ConfirmEdit
- https://www.mediawiki.org/wiki/API:Action_API
- https://meta.wikimedia.org/wiki/Research:The_Rise_and_Decline
- https://www.mediawiki.org/wiki/Extension:MobileFrontend