Whoooaaa Duuuuude: Why We Stretch Words in Tweets and Texts - WIRED

(Or that you’re a sarcastic talking raccoon.) These are known as stretchable or lengthened words, and now researchers from the University of Vermont have figured out just how pervasive they are on Twitter, uncovering fascinating patterns about their use.

Writing today in the journal PLOS One, the researchers detail how they combed through 100 billion tweets, mapping how often these words are stretched, and how far they are elongated—haha versus hahahahaaaa, for example.

“That can convey basically anything, like ‘Duuuuude, that's awful,’” says University of Vermont applied mathematician Peter Sheridan Dodds, one of the study’s coauthors.

“I hate using exclamation marks because they just don't fit my personality,” I tell Dodds and his coauthor, Chris Danforth, also an applied mathematician at University of Vermont.

And that can be particularly powerful on a platform like Twitter, whose inherent brevity doesn’t exactly encourage nuanced communication.

To quantify this, Dodds, Danforth, and the lead author of the paper, University of Vermont computational linguist Tyler Gray, randomly selected 10 percent of all tweets sent out between 2008 and 2016, around 100 billion in all.

(They have an arrangement with Twitter to obtain this data.) Gray wrote a program that searched the data for stretched words, specifically looking for repeated letters.

“So this seems like a candidate for a stretchable word,” continues Dodds.

This quantifies what the researchers call the “balance” of a stretchable word.

With the various stretched spellings of the word goal on Twitter, the G repeats maybe once or twice.

So in the case of the word goal, it’s the vowels that do the lengthening, and they tend to lengthen in lockstep with one another.

“What we didn't know beforehand is that those lines are pretty linear,” says Dodds.

In general, two-letter words stretch farther than regular words, like finallyyyy.

“People start with F, and then they lay on the Us,” says Danforth.

Because stretched words can be embedded with so much extra meaning beyond the words themselves, understanding them is critical for artificial intelligences that analyze text, like chatbots.

“If we're ever going to get to a point where an AI can understand the range of communication that people actually use in a day-to-day basis, this is one of the places where it's at,” says Sam Brody, who published his own research on Twitter word lengthening in 2011, prior to joining Bloomberg's AI group as a senior research scientist.

One quirk the researchers noticed was that when Twitter users were trying to be uber-emphatic, like to attract the attention of a celebrity, they elongated everything.

“There was a second kind of word,” says Dodds, “like: ‘fffffooooolllllllloooooowwwwww mmmmmmeeeeee, Justin Bieber.’ People would stretch out the F, the O, the L, or they would just stretch the whole thing out.