TIME Magazine's Claire Suddath commented on Яolcats last week, claiming that it is funny (she's right) but that lolcats are not (she's wrong). Suddaths's (tongue-in-cheek?) complaint about lolcats is most relevant here on Language Log:
Lolcats is stupid. There, I said it. People who attribute grammatically incorrect statements to unsuspecting housecats are the same people who speak to children in baby voices and pat pregnant women's bellies without asking permission. Besides, even if your cat could speak, and it happened to ask for a cheeseburger, why would it spell "cheez" with a "Z?" Why? It's one thing to pretend that your cat can talk, but it's another thing to pretend that it has a debilitating speech impediment.
If Suddath pronounces "cheese" in any way that sounds different from what "cheez" is meant to represent, then I hate to tell her, but she's the one with the speech impediment.
And let's look at the text of the lolcat example in her "speech impediment" link:
Dis iz da spoon u wil now uze to measure ma foodz
The only noteworthy deviation from Standard English in terms of pronunciation here is "th" being represented with "d". (I take the "a" in "da" and "ma" to simply represent schwa, which is not at all an atypical pronunciation of the vowel in these function words.) This is a very typical dialect variant, not a speech impediment. Of course, it's also a prototypical child's pronunciation of "th", but not because they have a "speech impediment". (Children are cute, hence the cat is cute — that's supposed to be part of the appeal of lolcats, innit?) Yet another sad example of how the media grossly misrepresents linguistics.
[ Hat tip: Emily Matthews ]