“But it wouldn’t be an encyclopedia; it would be a wiki”: the changing imagined affordances of wikis, 1995-2002

Stuart Geiger (@staeiou, stuart@stuartgeiger.com)

UC-Berkeley, Berkeley Institute for Data Science

Association of Internet Researchers (IR17)

Tartu, Estonia | 19 Oct 2017

The pre-history of Wikipedia

1999-2000: Nupedia: Jimmy Wales & Larry Sanger's volunteer-based project to write a freely licensed encyclopedia.

It was highly-structured, expert-based, and...

...a failure. After 18 months & $250,000 USD, only 12 articles.

(From archive.org, layout edited to fit)

Winter 2000-2001: Larry Sanger & Jimmy Wales learn about WikiWikiWeb, Jimmy asks them a simple question:

(archived from WikiWikiWeb/WikiPedia, as told in Sanger's memoir)

(They go ahead and make Wikipedia anyway)

Socio-technical governance of Wikipedia

Why does Wikipedia look like this?

Instead of this?

Wikipedians started out with the same platform as late 1990s wikis

(archived from archive.org)

But Wikipedia became a quite different kind of wiki

Wikipedia was neither the first encyclopedia nor the first wiki

Diderot and D'Alembert's Encyclopedie (1751) Ward Cunningham's WikiWikiWeb (1994-5)

Encyclopedia: "circle of knowledge" broadly summarizing what is known

Wiki: web hypertext documents that anyone can edit

Wikipedia is produced out of the tension to be both a “wiki-” and a “-pedia”

(also see Rosenzweig 2006, Lih 2009, Reagle 2010, Niederer & van Dijck 2010, Ford 2014, Wadewitz 2014, Tkacz 2015)

So what was a wiki pre-Wikipedia?

The fast, quick, read-write version of the WorldWideWeb

(the php wiki, archived by John Abell)

Originally, wikis did not:

  • keep full revision histories (with UIs for browsing history)
  • separate article content from discussions about content
  • be able to temporarily protect pages from public editing
  • use or have built-in support academic-style references
  • templates for specialized formatting and task signalling [citation needed]

Pre-Wikipedia "wikizens" linked software features (or lack thereof) to ideas about what "the wiki way" demanded from them

As Wikipedians worked out how to write an encyclopedia in a wiki, they revised the meanings and materialities of both.

Imagined affordances (Nagy & Neff 2015):

“Imagined affordances emerge between users’ perceptions, attitudes, and expectations; between the materiality and functionality of technologies; and between the intentions and perceptions of designers.”

Imagined affordance #1:

The "WikiNow" versus a full version control system

"WhyWikiWorks" from WikiWikiWeb:

(archived on archive.org)

(archived on archive.org)

To many Wikipedians, the lack of a full revision history was a bug

(from a wikipedia-l discussion)

Controversies over admins deleting pages from the database with no record

(from a wikipedia-l discussion)

Soon, the codebase was patched to support full histories and logging

(archived on archive.org)

Imagined affordance #2:

Wiki pages as stream of consciousness vs. separate pages for content and discussion

Wikipedia's talk pages, separating content from discussion

Wikipedia article Wikipedia talk page

I'm going to have to skip over some slides now

Wikipedia would become both an encyclopedia and a wiki, as Wikipedians revised both what encyclopedias and wikis were.

We modify the technology to make it determine the society we thought the technology had always determined