In my recent go rogue posting, I reported a comment on an earlier posting from Daniel Gustav Anderson on go rogue as a sexual euphemism, saying that at first I suspected the comment of being spam, but decided it was legit. Then Jake Townhead commented on my posting, questioning my use of the word spam and suggesting that Anderson's comment was merely "bespoke mischief". So now some words on spam.
First, a note on why I decided DGA's comment wasn't spam. Anderson appears to be a graduate student in Cultural Studies at George Mason University (in Virginia). He has a blog "For the Turnstiles … on critical theory and integral theory", and there's a recent interview with him ("Nonviolence of Nonmetaphysics") here and mentions of him and his work on a number of sites.
(Of course, this could all be an elaborate fiction, an invented persona and backstory, but at some point you have to trust some of the things you read. Another possibility is that someone has appropriated his name, e-mail address, and blog link; I've written him to check that he was in fact the source of the go rogue comment. Later: he has now written back to say that he was the author but that he's learned he was the victim of a prank. He apologizes for repeating misinformation, and asks that his comment be deleted — but that wouldn't really help much, now that I've posted about it.)
The upshot of all this is that I'm closing my books on sexual go rogue. I started out dubious, but now I'm convinced that it's a sheer invention. DGA is not an invention, but this story about go rogue is. Alas, the story seems to have spread, in the way that amazing stories do — probably because it digs at Sarah Palin.
On to spam. There's an OED draft entry of 2001 for spam as a verb: transitive 'to flood (a network, esp. the Internet, a newsgroup, or individuals) with a large number of unsolicited postings, or multiple copies of the same posting', intransitive 'to send large numbers of unsolicited messages or advertisements'. The first cite is from 1991, but that's from the New Hacker's Dictionary, so the word (in a non-food sense) goes back before 1991.
The OED has, at the moment, no entry for a noun spam, but that's also been around for some time.
[Added 11/18: Actually, it does have a sub-entry (a draft addition of 2001) for the computing noun spam, but it's under the main entry for Spam (the food name), where I didn't think to look. See discussion by Alex Boulton and me in the comments on this posting.]
In fact, it's been around as a mass noun ("how much spam?") and a count noun as well, meaning 'spam message, piece of spam' — a count noun that can have an ordinary s-plural ("how many spams?") or a zero plural ("how many spam?"). By 2001, when I discussed the noun in section 10 of this paper, all three of these uses were attested, and some people were using two or all three of the variants, sometimes in the same text, while other people had very strong preferences for one of the variants (I myself strongly prefer the mass variant).
(Entertaining fact: the spam filter that WordPress uses — on both Language Log and my blog — used to announce
Akismet has caught N spam for you since you first installed it. [with the zero-plural count variant]
but at some point it shifted to
Akismet has protected your site from N spam comments already … [with the noun as the first element in a noun-noun compound, a position where the count/mass distinction is normally neutralized]
The count of spam comments on Language Log is over 160,000 at the moment. Oi.)
But what counts as a spam comment? A bulk mailing of a comment, for sure. Your classic spam comment has no real content — instead, it has nothing that relates to the posting it's "commenting" on ("Great blog!"; "How can I subscribe to this blog?"; "This simply prodigy!"; "Can you explain your point?"; a URI or list of URIs; a quotation from some random work; an advertisement for some product; etc.). The point of such comments is to send the unwary reader to a commercial URI or to jack up the sender's hit count (or, of course, both). Others, in particular those that are all random gibberish, seem to be just malicious.
Spammers have gotten cleverer over the years, as I noted a while back. Comments now often seem to be tailored to some degree to the posting they're commenting on: comments linking to porn sites attached to postings with sexual content, for instance, or comments purporting to come from people the blog has mentioned, but linking to a commercial site. Some actually refer to the content of the posting ("That is an old cartoon" is a recent one on my blog, following up on a posting with a cartoon in it), but then it turns out that the URI provided is a commercial one, concerned with this content in some way (with the subject of the old cartoon, in this recent case).
It becomes increasingly difficult for bloggers to decide whether they're dealing with bulk mailings (with the targets selected by software crafted for the purpose) or with hand-crafted comments designed to send the reader to a URI or just to make mischief. As a result, many people — clearly, I am one — have extended the word spam to cover a variety of annoying, mischievous, or malicious postings and comments that might not have been mailed in bulk. Marking such postings and comments as spam for the purposes of a spam filter is then a useful step, since the malefactors are likely to repeat their offense, and one of the things spam filters are sensitive to is the sender's address. This isn't dangerous. In WordPress, the Akismet filter doesn't automatically delete things that appear in the spam queue, but merely offers them up for moderation, so misclassifications (in either direction) can be corrected.