We should never confuse text on the web for text on a printed page. While the accessibility of web text is awfully attractive -- I am, after all, writing this to be read online -- its ephemeral nature means that text on the web can easily be changed or simply disappeared at a website owner's whim. There need be no fingerprints left when a site changes the content of a page. When the changes are made to fix typos or or clean up phrasing, it's generally not a big issue; when the changes to the site eliminate or alter material now deemed to be politically unpalatable, it's reasonable to be concerned.
Sometimes, caches and archives created by search engines become impromptu evidence of changes; if caught soon enough, the original content can be held onto and posted elsewhere. Such is the mandate of the Memory Hole, a website dedicated to preserving things that the original sites decided needed to be disappeared. Often, these are government documents; sometimes, they're pieces from businesses or even newsmagazines.
The Memory Hole is a good example of the kind of transparency-enforcement possible in a networked world. It's far too easy to make documents disappear when they contain uncomfortable material. The one drawback is that the Memory Hole is just one place, one site. A well-placed cease & desist or denial-of-service attack could knock it out. What we need is a way to combine the content of the Memory Hole with the structure of, say, Gnutella...
Comments (2)
Hmmm...
The problem is one of trust: you need to know that the documents were really official texts. This *could* be done in a distributed way, I'm sure, but it would take a lot of thought. The memory hole takes the simple - but effective - way out.
What about FreeNet?
Posted by Dan | November 15, 2003 10:22 PM
Posted on November 15, 2003 22:22
I have an idea related to this which also has other advantages - basically distributed/peer-to-peer HTTP. As well as automatically sharing your bookmarks with your friends who are interested in similar stuff, it enables them to get cached versions of the pages from your machine (or the machine of someone else in the group who's cached them), thus reducing the "Slashdot effect". A side effect is that you can check for changes on the pages and highlight them to see how they are changing.
Don't know when/if I'll get around to developing it, though....
Posted by Mike | November 18, 2003 7:07 PM
Posted on November 18, 2003 19:07