In the CNN article, GLM responds to the objections of sane linguists and lexicographers everywhere in the person of Paul J.J. Payack, “president and chief word analyst for the Global Language Monitor” — chief word analyst? Nice title, if you can get it! He says, well, of course, it’s just an estimate, and the real point is merely to celebrate the enormity and continuing growth of English. But on the GLM website, au contraire, it’s all about this particular “word”, “Web 2.0” — nor is it really very surprising that a website should wish to aggrandize Internet-centric terminology. They make quite a big deal out of this particular term being their one-millionth word, even to the point of enumerating fifteen finalists, all of which were beaten out (how exactly?) by “Web 2.0” for the top honors. Of the fifteen finalists, by the way, only about half are really individual words and not phrases. And with only one or two exceptions, each is an ephemeron of one sort or other (usually political, technical, or pop-cultural). “Octomom”, we’re told, is now an English word. Please, say it ain’t so! “N00b” and “defriend”, okay, maybe — and yes, that’s “n00b” with two zeros. But “sexting”? Are you kidding me?!

By the way, the 1,000,0001st word, the GLM informs us, is “financial tsunami”. Again — *sigh* — not a word. And this is only one type of imprecision on this website (although, for me, it is the most annoying). Just start reading along, and you’ll see what I mean. For an organization that supposedly monitors English usage, the GLM could use some English lessons. Not to mention a copyeditor and a fact-checker. All of this is just vacuous hype (with the ulterior motive of promoting their marketing services, I would imagine). Being interested in words — and phrases — is a wonderful thing, but claiming any kind of authority in monitoring, branding, or counting them is simply balderdash. Now there’s a word for you, Mr. Payack!

Being interested in words — and phrases — is a wonderful thing, but claiming any kind of authority in monitoring, branding, or counting them is simply balderdash. (emphasis added)
While agreeing that this is just a silly P.R. stunt, there may be some value in counting words: take Michael Drout's recently-announced "lexomics" project, which attempts to reach conclusions about Old English texts by, among other things, counting the frequency of their words.

Thanks for the comment, N.E. Brigand. Yes, the “lexomics” project is indeed interesting, and potential quite valuable. Also, it’s a completely different animal. I don’t mean to suggest that one must never attempt to count words under any circumstances. But the Global Language Monitor differs from Mike Drout’s project in at least two important ways. The project: (1) counts the frequencies of words in a fixed, finite corpus; and (2) we are talking about a “dead” language (i.e., no longer a growing tongue). But naturally, attempting to count words in a living language is futile at best, and ascribing to them any kind of ordinal sequence is just silly. Your word there, quite apt, and a much better word than “Web 2.0” — we are in complete agreement there. :)

Surely -- if two word combinations like financial tsunami count -- then we're up to the 1,000,000,000,000,000,000,000th "word." Or at least we will be by 5:22 p.m.

-SamR