Facebook App Put Words In Your Mouth - What Would I Say?
November 15, 2013

Don’t Put Words In My Mouth – Put them In My Facebook Status

Bryan P. Carpender for redOrbit.com - Your Universe Online

So I do your Lame Duck session starts and ends with a cover of this year! - BryanBot

That’s what appeared on my monitor when I was finally able to gain access to the What Would I Say? app that’s going viral and giving people a fun new way to waste their time. It’s simple (at least on the surface).

You simply log into the app with your Facebook credentials. Then you generate a status. Then wait a few seconds for the ensuing hilarity as a high-tech Markov text-generating bot combs through the storehouse of intelligent prose and scintillating reports known as your previous Facebook status updates and produces a mash-up of utter randomness.

If you like the word potpourri it hands you, then you can post it to Facebook with one click, much to the delight of your friends and loved ones. If the bot hands you something that doesn’t quite speak to you, simply click again to generate another status.

Lather, rinse, and repeat a few hundred times until you find a status (or twenty) that you deem worthy of your wall and post them as you go along.

Initially, I scoffed at this whole concept. I had to ask myself, “Are people really that lazy?” (Yes.)  “Are people really that busy wasting time on Facebook and all its trappings that they can’t even be bothered to come up with an original freakin’ status update?” (Again, yes.)

Then I clicked again to generate a new status. And another one. And another. Before I knew it, I had fallen down a What Would I Say? K-hole. A shame spiral of the randomness that was my own status updates thrown in a virtual Vita-Mix and blended up to a smooth, incoherent consistency.

The brain-trust behind this sensational new time-suck is comprised of Pawel Przytycki, Ugne Klibaite, Vicky Yao, Daniel Jiang, Edward Young, Harvey Cheng, and Alex Furger — all graduate students at Princeton — who created the app during last weekend’s Princeton Hackathon.

According to Klibaite, they got their creative juices flowing in the time honored manner: “We drank a lot of coffee and Red Bull and thought of fun things we could program that we could actually complete in a day and a half. This was just for fun. We never thought we would get further than showing this off at the Hackathon and to our friends on Facebook.”

But good news travels at the speed of light and by Tuesday, the app was being featured and linked to by a slew of sites and people were posting images of their randomly generated status updates all over social media.

Though simple in appearance and user interface (like I said, there are only two buttons to click), the behind-the-scenes machinations are more complicated.

“What Would I Say? automatically generates Facebook posts that sound like you! Technically speaking, it trains a Markov Bot based on mixture model of bigram and unigram probabilities derived from your past post history. Don't worry, we don't store any of your personal information anywhere. In fact, we don't even have a database! All computations are done client side, so only your browser ever sees your post history. The results are often hilarious (and sometimes nonsensical) -- it all depends on what you've written on Facebook,” explains the “How does it work?” section of the app.

Don’t overthink it, though. Just go explore it for yourself.

Consider this fair warning: this thing is mindless and addictive. There’s only two buttons to click, so it’s completely possible (read: inevitable) that you can and will lose track of time as you let the bot do your work, generating statuses that will make you look positively mad with genius and whimsy to your Facebook followers.

Another option is that it will randomly generate a status that might compel those close to you to go all 5150 on you and have you put on lockdown while they “run a few tests.” Keep that in mind and exercise caution when selecting which ones to post to Facebook.

Sometimes it’s totally nonsensical word salad:

Especially if one receives it in Baltimore when they stopped the audience, much less than hide from deciding whether or not to blaze outta the office, and look what say in the Roman Senate... - BryanBot

But there were a few times where the app seemed to be reading my mind:

The number starts. Love her to pieces. Allison Janney genius! - BryanBot

And then the app seemed to look directly into my soul and glean this gem:

I'm more like Meh. - BryanBot

(On that note, cue the Gangsta mic drop. I’m out.)

Check it out for yourself, but you’ve been warned.