Retraction and Revision of Online Works

[packbat]

Recently, a webcomic artist whose comic I read posted a strip which, unbeknownst to ... em*, duplicated a punchline from another strip, several weeks earlier. Ey got many emails informing em of this, and therefore changed the punchline panel an hour or two after it went up. Ey also posted a note, informing eir readers of the edit.

I could tell who this comic artist was, and what the change was ey made, but I won't – it's irrelevant to this essay. What I'm interested in is the idea that one can publish a work on the web and then change it after it's already been viewed.

The idea of revising already-published works, and of retracting what has already been publicly said, is not new, or even controversial. A 'corrections' box in the newspaper is standard practice, poems published more than once are often altered to some degree between printings, and essays and editorials are occasionally followed by literal retractions. The Gettysburg Address underwent multiple revisions even after it was delivered; the version in the textbooks is a modification of the one Lincoln delivered on the battlefield. What makes the online case unusual is that the work is only published in a single place, and it is often easy for the artist to unpublish the old version, leaving only the altered piece with no evidence of tampering.

This secret (or semi-secret) rewriting angers me. It is sometimes hard to make a case for what harm the creator of a work does in revising or removing it from a webpage; while in specific cases additional factors may apply, in general the harm it does lies only in the loss of the original work, either to deletion or revision. However, further harm can lie in the corruption of responses to the original work, especially when it is deleted altogether; when element of a discussion is removed, it eliminates the capacity for the audience to judge any part of the discussion that involves that element.

Even if the alteration of a online work is announced and explained, there remains some value in the original version which is lost if the revision is the only version that can be viewed. Admittedly, while this chain of reasoning can lead to extremes – like requiring that all first drafts be posted, and posted before the actual work, when possible – it is not unreasonable to be interested in the version which was associated with the initial release, and mere information on the differences does not suffice in this regard.

I think Eric Burns of Websnark had the right idea when he stated his policy on mistakes on Websnark (original statement heavily mangled to remove context):

[...] So, I thought I'd type for a few minutes on one of my unwritten policies for Websnark.

If I make a mistake, or blow something, or post something I later regret... I'll own up to it, but I won't delete it. This is a record of... well, something, at least. Of my thoughts and blatherings, if nothing else. And when I blow it, I'm going to leave it up and own up to it.

[...]

We live in an age where the historical record is easy to alter. We combat that not with technology, but with integrity. When I overreact or blow a call, that mistake becomes a part of this archive. That just seems fair to everyone.

This is an attitude I admire. I admire it partly because it shows bravery and honesty, but I also admire it because it means that (for example) when he writes an eloquent rant that later turns out to be unjustified, the rant stays available for reading anyway. It means that points he made in the rant are still available for consideration. And just because the central thesis gets refuted doesn't mean that all the ancillary points become invalid.

I admit that, under many circumstances, the original work is almost irrelevant. Furthermore, under some circumstances, change is almost demanded, as in the case of the misplaced rant, and the case of the duplicated punchline. I admit that sometimes changes are almost unambiguously improvements, and that the value that was in the original is sometimes entirely subsumed in the revision. But even in this case, the best case, a certain amount of history is destroyed by the vanishment of the original version.

And that's probably what bothers me about it. Knowledge gets destroyed, however useless that knowledge was, and in a world where nothing is perfect, we need all the knowledge we can get.

Comments

[nanakikun.livejournal.com]

heh heh eh

a virtual experience

[nanakikun.livejournal.com]

Of course, we'd have to suggest, as being nanakikun is being, that one's history, especially when in a media such as the internet where you can save your own copies of everything you experience* , part of the experience is going back to where you grew up and finding that all the buildings have moved around, or are completely different.
If it were the web's responsibility to keep everything the same, then there wouldn't be any way to get from the older websites to newer ones... : /

This topic is discussed at length in the somewhat confusingly translated Metal Gear Solid 2: Sons of Liberty, available for Sony PlayStation 2 and Microsoft X-Box. : D

* come back later and insert some link here to the firefox plugin that provides disk storage of everything browsed

newspaper errata & corrections

[(Anonymous)]

Byron Calame in his web journal (http://forums.nytimes.com/top/opinion/readersopinions/forums/thepubliceditor/publiceditorswebjournal/index.html) and in his columns (Sunday New York Times op-ed pages) sometimes writes about how to get factual corrections into (or at least appended to online versions of) the newspaper ... e.g.:

Ms. Collins’ existing written policy calls for uniformly publishing corrections at the bottom of opinion columns, which leads to their being appended to the original article in nytimes.com and various electronic databases. The approach taken by Mr. Rich means that users of nytimes.com who find themselves reading the Sept. 18 column—something they now must pay to do—get no warning that it contains any error at all. As I’ve said before, I think a crucial function of a correction these days is to get the right information appended to the increasingly referenced digital versions of articles as soon as possible.

If things are out-and-out wrong (e.g., a three-order-of-magnitude error such as the one Calame mentions occurred in the 9 Sep NYT, where somebody inadvertently wrote megabyte instead of gigabyte!) then it seems legitimate to me to "edit history" and put the right words in place --- esp. since sometimes folks (or automated data-harvesters?) won't read down to the correction.

But if something is said inadroitly and the author wants to fix it, I dunno ... it depends perhaps on the importance of the changes, and who "owns" the original ... I am inclined to improve my ^zhurnal (http://zhurnaly.com) when I see something that can be significantly enhanced, and certainly to fix typos and factual errors and inadvertent insults or invasions of privacy ... and people in the Wiki world have wrestled with this sort of thing for quite a while ...

BTW, from the C-SPAN Congressional Glossary (http://www.c-span.org/guide/congress/glossary/revise.htm)

Asking to Revise and Extend allows a member to add to or edit his/her floor remarks in the Congressional Record.

Making changes to the words actually spoken on the floor requires the consent of the entire House.

Revisions are limited to those that make technical or grammatical corrections.

Extensions are usually the text of articles, letters, or reports to accompany the floor statement.

- ^z -