Virginia Heffernan at the NYT takes on the difficult issue of quoting casual web texts in more formal publications like the New York Times.

Consider another example. To show that Web users are curious about human reproduction, I might quote kavya on Yahoo Answers, word for word: “How is babby formed? How girl get pragnent?”

But that makes kavya look like an idiot. Readers might miss the sweet earnestness of his question. Maybe he (or she) is 7 or a native speaker of Hungarian. I should cut the kid a typographical break; that’s not an easy question to ask. The cockamamie diction and syntax of Internet English is, possibly, only incidental to his inquiry. A reporter could paraphrase or revise his question — “How is a baby formed?” — lest readers get blinded to the intent of the question by moronizing typos.

But “How is babby formed?” is funny. And who wants to deny readers a chance to laugh and to get the full flavor of Internet-culture wackiness? It’s flat-out lying to pretend that everyone (or anyone) spells well online.

This is something you've run into if you write about the Web much: Do you preserve spelling errors? Do you correct them in brackets? Do you just paraphrase? When and why? It all comes down to (predictably) rhetorical purpose, tempered by maintaining the validity of your data. As Times editor Daniel Okrent says in Heffernan's piece, we've struggled with this issue for a long time in terms of quoting verbal statements (do you spell out accents to give readers the flavor? do you preserve the "and ... uh ... um"?), and there aren't simple answers. (Not a direct quote.)

[via Fimoculous.com]