Avogadro Corp: The Singularity Is Closer Than It Appears (2 page)

Chapter 1

David arrived at the executive conference room ten minutes early, his throat dry and butterflies in his stomach. He tried without much success to keep his mind focused on getting ready for the presentation, pushing aside the nervousness that kept threatening to swallow him. It wasn’t often that project managers presented to the entire Avogadro Corp executive team.

It was a small relief that he was the first to arrive, so he could get ready without any pressure. Syncing his phone with the room’s display system took only a few seconds. There was no overhead projector here, just a flush mounted display panel in the wall behind him. He ran his hand over the polished hardwood desk, and leather chairs. A small step up from the plastic and fabric in the conference rooms he frequented.

David took some small comfort in the ritual of getting a coffee. As he poured two raw sugars into his coffee, he smiled at the lavishly stocked food table that contained everything from coffee and juice to artfully arranged breakfast danishes and lunch foods. Though Avogadro was an egalitarian geek culture company, the top executives of the company still had their perks.

Still no one had arrived, so David wandered around the room admiring the view. The dominant feature was of the Fremont bridge crossing the Willamette River. He could see the loft buildings in the Pearl District, and to the right downtown Portland. Directly to the East he could see Mt. Hood just below the cloud cover. The early morning sun was peaking through the clouds, sending shafts of sunlight toward the city. He was just wondering if he could see his own house in Northeast Portland when he heard a welcoming “Hello David.”

Turning around, he saw Sean Leonov and Kenneth Harrison entering the room. Sean came over to shake his hand, and then introduced him to Kenneth. David was excited to meet the other cofounder of Avogadro Corp. Dark haired and easy going, Kenneth was respected, even if he didn’t quite command the same awe as Sean.

Other vice-presidents of the company started to file in, Sean making brief introductions to each. David shook hands or gave nods as appropriate, his head swimming with the names and roles of each introduction.

For a few minutes there was a cocktail party atmosphere as people grabbed drinks and food and socialized. Then they gradually took their seats, arranging themselves in a semblance of a pecking order surrounding Sean and Kenneth. One seat was conspicuously empty at the head of the table.

When the bustle of arriving audience members finally died down, Sean stood. “I’ve already introduced you to David Ryan, the lead project manager for the ELOPe project. I hired David two years ago to prove the feasibility of a radical new feature for AvoMail. He’s done an incredible job, and I invited him here to give you the first look at what he’s developed. Prepare to be amazed.” He smiled to David, then sat down.

“
Thank you, Sean,” David said, standing up, and coming around to the front of the room. “Thanks everyone for coming.”

David thumbed his phone to project his first slide, a black and white photo of a secretary applying whiteout to a sheet of paper in a typewriter. “One of the first corrective technologies was whiteout,” he said, to chuckles from the audience. “It was highly innovative in its own time. That was nothing compared to the spell checker.” In the background, the slide changed to a photo of a man using a first-generation personal computer.

“
Years later as computer processing increased, grammar checkers were invented. First generation grammar checkers detected mistakes, and later versions helped fix them. Spell checkers and grammar checkers started out in word processors, and gradually made their way throughout the whole suite of communication tools: presentation editors, email.” David paused, enjoying the storytelling portion of his presentation.

As David spoke, he focused on one executive at a time, making eye contact with them before he moved onto the next. “Today the standards of business communication have changed. It’s not enough to have a grammatically checked, correctly spelled email to be an effective communicator. You must intimately know what your recipients care about and how they think to be persuasive. You must use just the right mix of compelling logic, data, and emotion to build your case.”

David paused again, and saw that he had the rapt attention of everyone there. “Sean hired me two years ago to see if I could build an unproven concept: an email language optimization tool that would help users craft more compelling, effective communications. I’m here today to show you the results of that work.”

He flipped slides again, popping up a timeline.

“
In the first twelve months, through data mining, language analysis, and recommendation algorithms we proved feasibility. Then we started implementing the Email Language Optimization Project, or ELOPe, in earnest.”

David clicked again, and now the wall display showed a screenshot of AvoMail, the popular Avogadro web based email. “From a user experience perspective, ELOPe works like a sophisticated grammar checker. As the user edits an email, we start to make suggestions about the wording to the user in the sidebar.”

“
Behind the scenes, complex analysis is taking place to understand the user intent, and map it to effective language patterns we’ve observed in other users. Let me give you a very simple example you might be familiar with. Have you ever received an email from someone in which they asked you to look at an attachment, but they forgot to attach it? Or perhaps you were the sender?”

Chuckles, and a few hands, went up from the audience.

“
It is embarrassing, of course, to make that mistake. Today nobody does make that mistake, because AvoMail looks for occurrences of the words ‘attachment’ or ‘attached’, and checks to see if a file is attached before sending the email. Through language analysis, we’ve improved the effectiveness of the user’s communications.”

A woman vice president raised her hand. David struggled and failed to recall her name, and settled for pointing to her. She asked, “But that’s a simple example of adding code to look for specific keywords. Are you talking about simple keyword detection?”

“
That’s a good question,” David answered, “but no, we don’t rely on any keywords at all. I’ll explain how, but I’d like to use a more complicated example. Imagine that a manager is asking for more funding for their project. Before handing over money, a decision maker is going to want to understand the justification of that funding request. What’s the benefit to the company of providing more funding? Maybe it’s a quicker time to market, or a higher return on investment. Perhaps the project has run short of funds and is in danger of being unable to complete.”

David saw nods in the audience, and relaxed a little. He was glad his hand-picked example resonated with his audience of business executives. He continued, “ELOPe can analyze the email, determine that the user is making a funding request, know that it should be accompanied by a justification, and provide examples of what effective justifications might look like.”

David flipped to a slide showing this example. The short video capture demonstrated a user typing a request for funding, as example justifications popped up on the right hand side. Each example justification already incorporated details gleaned from the original email, like the project name and timeline. David waited quietly while the thirty second video played. He heard some soft exclamations in the background from the group. He knew this was incredibly impressive the first time someone saw it. He smiled to himself. It was Arthur C. Clarke who said, “Any sufficiently advanced technology is indistinguishable from magic.” Well, this was magic.

David paused to let the video sink in before resuming. “It’s not enough to provide a general set of recommendations. Different people are motivated by different kinds of language, different styles of communication, different reasons. Let’s use another example. An employee is going to ask his manager for extended vacation time. He’d probably like to make a compelling case for granting that vacation request. What will motivate his manager? Should he mention that he’s been working overtime? Should he mention that he needs to spend time with his kids? Or that he’s planning to visit the Grand Canyon, a place that his manager happens to associate with good memories?”

“
The answer,” David went on, as he paced back and forth in the front of the room, “is that it depends on the person you’re sending it to. So ELOPe customizes its analysis not just to what the sender is asking for, but for what the recipient is motivated by.”

David noticed that Rebecca Smith was standing in the doorway listening to the presentation. In a sharp tailored suit, and with her reputation hovering about her like an invisible aura, the Avogadro CEO made for an imposing presence. Only her warm smile left a welcoming space in which an ordinary guy like David could stand.

She nodded to David as she came in and took her seat at the head of the table.

Kenneth asked, “But what you’re describing, how does it work? Natural language processing ability of computers doesn’t even come close to being able to understand the semantics of human language. Have you had some miracle breakthrough?”

“
At the heart of how this works is the field of recommendation algorithms,” David explained. “Sean hired me not because I knew anything about language analysis but because I was a leading competitor in the Netflix competition. Netflix recommends movies that you’d enjoy watching. The better Netflix can do this, the more you as a customer enjoy using Netflix’s movie rental service. Several years ago, Netflix offered a million dollar prize to anyone who could beat their own algorithm by ten percent.”

“
What’s amazing and even counterintuitive about recommendation algorithms is that they don’t depend on understanding anything about the movie. Netflix does not, for example, have a staff of people watching movies to categorize and rate them, just to find the latest sci-fi space action thriller that I happen to like. Instead, they rely on a technique called collaborative filtering, where they find other customers just like me, and then see how those customers rated a given movie to predict how I’ll rate it. Sean’s insight was that since natural language analysis struggles to understand semantics, it would be best to start with an approach that doesn’t rely on understanding, but instead one which utilizes patterns.”

When David received nods from the audience, he went on. “That’s what ELOPe does. It looks at the language used by millions of email users. It looks at the language received by people, and how they reacted. Did they react positively or negatively? Compiled over thousands of emails per person, and millions of people, we can find a cluster of users just like the intended recipient of an email, and see how they respond to variations of language and ideas to find the best way to present information and make compelling arguments.”

Now there were some puzzled looks and half raised hands as people around the room tried to ask questions. David forestalled them with a raised hand, and went on. “Hold the questions for a second, and let me give you a simple example. Let’s imagine that a person called Abe, whenever he received an email mentioning kids, responded with a negative response.”

David gestured back and forth with his hands, getting into the example. “Now imagine that ELOPe has to predict whether a new email about to be sent would be received positively or negatively by Abe. If that new email also mentioned kids, it’s a good bet that it will be received negatively. If Abe was your boss, and you were going to ask him for vacation time, it’s probably not a good idea to use spending time with your kids as justification.”

He heard a few chuckles.

“
So is there is no semantic analysis?” Rebecca asked. “We don’t know why Abe dislikes kids?”

“
No, we have no idea why Abe feels the way he does,” David answered with a smile. “We just observe the pattern of behavior.”

“
What if my manager hadn’t received any emails about kids?” Sean protested. “How could we predict how he would respond?”

David smiled, knowing that Sean knew the answer, and was just helping him along. “Let’s say we have another user, Bob. Bob hasn’t received any emails about kids. However, ELOPe notices that Bob, Abe, and about a hundred other people have responded similarly to most topics, topics such as the activities they do on the weekend, the vacations they take, how they choose to spend their time. Let’s say that this group of people are ninety-five percent similar. That is, across all the topics they’ve responded to, they are ninety-five percent likely to have similar sentiment in their response: negative or positive. This is what we call a user cluster.”