On 7/9/06, Sarah slimvirgin@gmail.com wrote:
The policy is based on common sense, not dogma. When you send a letter to a newspaper for publication, you're expected to supply your name, address, and telephone number so that someone from the newspaper can check that you really did send it. Nothing like that exists for Usenet. It's all very well to say that if X didn't write the post, and we quote from it, X will tell us soon enough. But what happens if X claims that, in purporting to quote him, and in leaving that unchecked quote on Wikipedia for months until he spotted it, we have damaged him in some way? Newspapers have processes in place to avoid this scenario, and they have libel insurance for when things go wrong. We have none of those things, which is why we piggy-back on other people's, by using only material that has already been checked.
With groups.google.com (or previously to that, access to Dejanews, or previously to that being a close enough friend of Henry's to for him to be able to dig up the tapes) it is possible to (semi-reliably) go back and say "A person with this email account posted the following article at this time in history: [...]"
That is not to say that forgeries and frauds are unknown in Usenet. However, such were relatively rare, and are not unknown in other media which are taken generally to be acceptably accurate sources.
Wikipedia has not set the bar at unquestioned verifyable accuracy for every source. When someone adds in or uses a reference, there's no independent fact checking necessarily performed to see if the source really exists, or says what it is purported to say, or is not part of some sort of elaborate forgery project. And I do not doubt that within the million-odd articles, we have sources cited which don't exist, we have sources cited which say something else, and that we have sources cited which exist and say those things, but are forgeries or fraudulent.
Usenet posts, in verifyable archives, documenting Usenet activities, are as reliable as we need them to be. The existence and contents of such posts is easily and reliably validated. As a primary source for "the contents of Usenet", they're fine. Articles on Usenet and early Internet phenomena citing Usenet discussions on the topics (or contemporary mailing lists) should feel free to use those contents. If it turns out that a particuar cite was a forged article, we deal with that the usual way as with any other discredited source.