the truth is in the back and forth 09.07.2010, 11:54 PM
posted by bob stein
James Bridle (designer and programmer of the Institute's Golden Notebook project in 2008) just published the complete history of the Wikipedia article on the Iraq War.
James writes on his blog:
This particular book--or rather, set of books--is every edit made to a single Wikipedia article, The Iraq War, during the five years between the article's inception in December 2004 and November 2009, a total of 12,000 changes and almost 7,000 pages. It amounts to twelve volumes: the size of a single old-style encyclopaedia. It contains arguments over numbers, differences of opinion on relevance and political standpoints, and frequent moments when someone erases the whole thing and just writes "Saddam Hussein was a dickhead".
As early as 2006, i wrote in if:book that the truth in Wikipedia articles lay in the edits, rather than the surface article:
In a traditional encyclopedia, experts write articles that are permanently encased in authoritative editions. The writing and editing goes on behind the scenes, effectively hiding the process that produces the published article. The standalone nature of print encyclopedias also means that any discussion about articles is essentially private and hidden from collective view. The Wikipedia is a quite different sort of publication, which frankly needs to be read in a new way. Jaron focuses on the "finished piece", ie. the latest version of a Wikipedia article. In fact what is most illuminative is the back-and-forth that occurs between a topic's many author/editors. I think there is a lot to be learned by studying the points of dissent; indeed the "truth" is likely to be found in the interstices, where different points of view collide. Network-authored works need to be read in a new way that allows one to focus on the process as well as the end product.
Four years later, we don't yet have the tools that would let people read Wikipedia articles in "a new way" but hopefully Bridle's very impressive experiment with this one article will spur efforts to develop new tools for reading online works which are constantly being changed and edited.
Posted by bob stein on September 7, 2010 11:54 PM
ianf on September 8, 2010 3:17 AM:
I am quite perplexed by apparent odd ordering of
the pictured volumes of this (I assume single-copy)
issue: rather than line up in ascending numerical
order, I to XII (1st to 12th volume), they come in
strange "apple-cart upsetting" order of
- that is, 8th volume is followed by the 10th, with
the 9th last in line, after the 11th. I can no believe
that a designer preparing the picture of his novel
project for worldwide distribution would have
missed this key aspect of it, so it has to have
been done on purpose. The meaning of which,
however, eludes me entirely.
Stefan Tobler on September 8, 2010 4:27 AM:
Brilliant. I'm gobsmacked. Every library and arts institute should have a copy!
ianf on September 9, 2010 3:06 AM:
Oh, come on, Stefan, what would that accomplish? It is clear that this project is more of a one-off art event, proof of a concept, happening-with-solid-printout, than a template for replication and everyday use.
jp on September 9, 2010 6:26 AM:
Hoax? I searched a lot of links on this one but can't find either a download, view or purchase option. At best poor promotion I would suggest... Where ARE these alleged volumes?
ianf on September 10, 2010 5:04 AM:
No hoax. Why does everything have to be downloaded
or come complete with purchase options (or else). It's
a ONE-OFF ARTS PROJECT meant to illustrate
the making of culture. Not a very deep/ serious
one at that, as can be extracted from this postscript
NOTE: Yes, the books are out of order in the photos.
Kindly do not draw inferences from this. It’s just
a photograph. Seriously.
bowerbird on September 10, 2010 5:41 AM:
are the contents of these 12 volumes the _changelog_ alone?
or is the article repeated in full for every edit that was made?
either way -- but _especially_ if it's the former -- this shows
how ludicrous it becomes to suggest studying all the changes.
and as for the idea that "the truth lay in the edits", hogwash...
it'd be more accurate to say that "the truth _lies_ in the edits".
this method of "moving the anchor-points" is precisely how
right-wing nut-jobs remove reason from public discourse;
they just keep dragging out more and more ridiculous _lies_,
and the american public -- who thinks that "the truth" must
fall somewhere roughly in the middle of the two extremes --
keeps moving itself towards an increasingly bizarre "truth"...
it's time to stand up against "the big lie", not accommodate it.
ianf on September 11, 2010 6:44 AM:
@bowerbird, not much is known about the capture-
and-display principles of the actual printout; two
small pictures of actual pages from it have been
published. It must've been extracted and laid out
fully programmatically, with some final editing out
of superfluous administrative Wikipedia page fluff.
I find the idea of physically demonstrating the
extent and range of edits of a single article quite
intriguing, but not in hardcopy format - which, no
matter how one looks at it, is outdated and basically
a mantelpiece object d'art the moment it has been
bound. On the other hand, suppose a nuclear (or
asteroidal Extinction Level Event-) winter occurs,
the Wikipedia infrastructure burns up with only
patchy backups, but this one-off impression survives.
Then we'd end up with this, an indisputable proof
of how things were in one aspect of life BEFORE
That said, I wish Bridle (or someone else) would
devote time to construct more generic "regurgitation
tools" - by which I mean post-processors allowing
anyone to see/fathom the extent and character of
changes in/of any one single Wikipedia (etc) page,
not solely that of interest only to those 12000-odd
individuals who have taken time off their busy
schedules to add/ subtract/ fold /spindle/ mutilate
the Iraq truth-or-"truth". And by "see/fathom" I mean
a visual front-end, something akin to now-and-then
presented dynamic visualisations of (e.g.) tweets
sent out during specific sports events and the like;
only here, after suitable analysis of cumulative
WP data, shown in some chronologically-concurrent
fashion (think Java applets; can not find the hrefs
to the samples of such tweets-visualisations that
Even such static-but-overviewable-at-a-glance graphic
as Wikipedia lamest edit wars provides more utility
than the entire 12 volumes of Bridle's.