Dating Language

Written by Nexcerpt on April 6th, 2011 in Dating & Online, Patterns & News.

Turnstyle describes Online Dating Lingo Tapped and Mapped, visualizing the language of online dating profiles. Well… allegedly visualizing the language. Sort of. If you don’t care about language.

The project is by R. Luke DuBois. It’s a promising concept, weakly implemented. It shouts, in pale grey blue and mauve, why artists who lack analytical skills should stay out of the infographic business.

DuBois says, “I joined twenty-one dating sites in order to make my own census of the United States in 2010.” Considering that his list mentions “,” you gotta wonder which one he left out when naming only twenty:,,,,,,,,,,,,,,,,,,, and

Virgins.  Or Not Virgins.  You decide.

Virgins. Or Not Virgins. You decide.

At some time, I have been a member of over one-third of those sites. I assure you: in their profiles (as in every communication) different people use words in very different contexts, with very different intent.

In the first map, Dubois jumbles together M percent of profiles that open by asserting, “I’ve never been lonely!” along with the N percent that sign off by admitting, “Being lonely sucks.”

Another map representing appearances of “Shy” mixes together (and thus presumes equivalent?) the X percent of profiles that begin by whispering, “I can be shy,” with the Y percent that end shouting, “Don’t be shy!”

A few maps reflect some shadow of what they purport to reveal. True, most profiles that invoke the word “funny” do so in describing the seeker or their preferred prey date. Usage is less clear about a word like “sarcastic,” though it’s usually mentioned as a favorable trait.

Check the entire panel of representations of online dating language. As to meaning, they’re all over the… ummm… map. Turnstyle misses this fundamental problem entirely, by calling attention to maps of Naughty vs. Nice:

“DuBois said that from this image he can tell that no one in Wyoming used “naughty” in their profile, but bigger amounts of women in Colorado used “naughty. In addition, all the purple on the Nice map suggests that both men and women use “nice” in their profile.”

Yes: the word “naughty” is rife with meaning — almost always subtly sexual — but… “nice”? I just displayed (randomly) three women’s profiles from, each containing the word “nice.” Sample usage: “A nice mixture of Irish, German and English…” “I enjoy a nice Merlot…” and “Be Nice — I am FBI smart!

I’m unsure what to make of that last one — but I don’t see any of them contrasting with “naughty.”

Granted, it’s difficult to imagine working the words “anal” or “oral” — much less “spanking” into an online profile in other than a sexual context. (You’d rarely see any of those terms at, and never at eHarmony!)

But… what if those words correlate with a forceful “no”? “No [activity]” paints an incredibly different picture than “Must dig [activity].” Here, pro and con — top and bottom, spanker and spankee — are all stirred together as equivalents.

Dubois also offers a mathematically vague explanation of why he associates certain words with certain cities. But, again, any value of the data — beyond the frequency of sports teams’ names — is vacated by his silence on methodology. He tags the capital of Alabama “Conservative.” Duh. Yes, some singles might identify themselves that way (though perhaps not if actually hoping to get a date). I suspect that many people in Montgomery wrote some variation on, “OMFG, how can everybody around here be so freakin’ conservative?!?

Are the maps useful? No — at least without details of the methodology, and of some flavor of semantic analysis. But, wait… apparently there was no semantic analysis. Uh-oh. Context matters. Meaning matters. Ignoring the things that matter is a strange approach to any task. OK, let’s say it’s merely art — maybe it doesn’t need to matter? Yikes!

On the technology front, the project mixes both the progressive and the archaic. In a video describing the project (you may safely skip the wasted first 30 seconds), DuBois offers a very cool scan of hundreds of dating profile headshots — starting at 1:20. It echoes that ridiculous loop the imagineers insert into every episode of CSI (while “running the fingerprints through AFIS”). Except… it contains more than ten photos, and they belong to real people.

Dubois undermines respect for that clever implementation by admitting: “I took an old Rand McNally, and I scanned it into my computer, and I traced it by hand, and traced all the borders and the rivers and the lakes and the highways, and stuck in all the towns…”

At least we know where the first 18 months of the project went! (I will resist the urge to mention the first ten public domain sources for such digital maps.)

Looking at any of these “representations” is a bit like reading a horoscope: it means absolutely nothing in itself — yet it can mean absolutely anything to anyone.

In such cases — horoscopes and dating maps — the author/artist presumes that every member of the audience will arrive holding all the necessary meaning in mind beforehand. No one involved in creation — or observation — need show any recognition of reality. It’s not intended to guide a legitimate understanding. It’s just… some stuff put together to look at. There’s no need for it to communicate anything meaningful.

Geez. Nice work if you can get it.

Don’t get me wrong: these maps are kinda pretty. They might make you think. But, don’t imagine for a moment that what you think is what’s really happening.

Maybe Dubois would submit the dataset to a group with the credentials and analytical skills to tease out some useful information.

Hehe… I said “tease.” And we all know what that means…

Comments are closed.

Recent Posts and Other Categories