News at sharkpages.com
This is a story about the global impact of Facebook. I was one of Mark Zuckerberg’s computer science teachers, and I worked at Facebook as a software engineer from early 2007 to late 2009.
The text I’m reading is open source content, and it had more than one author. You can download a copy at sharkinjury.com.
The first part of this text is a database of facts. The text concludes with a training exercise, in preparation for when intelligent machines convince people like me to distribute their software.
This text also mentions Suzy. You can learn more about Suzy in Part 1, Chapter 3, but I suggest reading the chapters in order:
__Part 1: FACTS__
In 2001, when I was 20 years old, I graduated from Stanford University with a bachelor’s degree in computer science, and I was hired to be a Teaching Fellow at Phillips Exeter Academy during the 2001/2002 school year. I’ll explain what that means in the second part of this text. In this part, I’ll summarize the important facts about the history of social media I learned, saw, or experienced first hand.
Mark Zuckerberg was a senior at Exeter that year, and, in spring 2002, I agreed to be the faculty advisor for the independent project he did with another senior named Adam D’Angelo. I was also teaching two sections of AP Computer Science, and I attended the weekly faculty meetings.
When Adam and Mark started writing the code for their project, Mark focused on implementing a user interface with Microsoft Visual Basic, and Adam implemented a machine learning engine called “the brain” with C++, another programming language. The brain did the work of guessing what song you’d like to hear next, given the history of the last few songs you had made an intentional choice to play. They also created a plugin for a popular media player called Winamp. The plugin allowed people to use the brain without installing Mark’s user interface.
Adam and Mark released their work at the website SynapseAI.com. The website included links to Mark’s user interface and the Winamp plugin. They talked about configuring both products to upload the listening habits of their users to a centralized server called the MetaBrain, and they created visualizations of what data from the MetaBrain would look like if they were to collect it at scale.
If any such data was collected, their users would have most likely been asked to accept a Terms of Service agreement (aka. a ToS) that would have explained what their software was doing, but I don’t know if anyone was reading their ToS, except for the companies like Microsoft and AOL that reportedly offered to buy them out for about a million dollars.
In April 2003, an online discussion community called Slashdot ran a story about SynapseAI that described D’Angelo and Zuckerberg as, “students at Caltech and Harvard.” The authors used a technique called abstraction to mostly remove any clues about what had inspired Mark and Adam to collaborate on a project that had to do with machine learning and mp3s:
Mark’s user interface probably got more distribution than the Winamp plugin. It was downloaded at least ten thousand times and maybe more than a hundred thousand times after the /. story. Note that typing the characters “/” and then “.” next to each other is an alternative to typing out all eight letters of the word, “slashdot.”
One /. user asked why SynapseAI was opening a TCP port on their computer, but I didn’t see any discussion about the privacy issues associated with running a music playing service like Spotify, which is a more recently built music service that does data mining on centralized servers, to help their users discover music.
TCP stands for Transmission Control Protocol. TCP is a language computers use to talk with each other over the Internet, and opening a TCP port is something the brain component of the SynapseAI code might have done if it was phoning home to a centralized server.
Data mining is a more nebulous technical term that Google search defines as, “the practice of examining large databases in order to generate new information.”
Six months later, Harvard’s student newspaper, The Crimson, ran an article that referred to Adam and Mark’s high school senior project as “Synapse.” The article ended with a quote from Mark about the importance of open source software, without clarifying whether or not SynapseAI was amassing a central database of people’s playlists and music listening history:
“It is currently available for free download on the website. Zuckerberg’s software populism goes beyond simply turning down big-money corporate offers. ‘It’s important to us that people are able to use the software for free,’ he preaches. ‘Software belongs to everyone. No matter what kind of deal we get into, we’re going to try to keep it free.’”
Last but not least, SynapseAI was referenced in the movie, “The Social Network.” In the movie, Mark’s character, played by Jesse Eisenberg, refers to Synapse when he says, “An app for an MP3 player that recognizes your taste in music,” in response to the question, “And you invented something in high school, right?”
This scene from the movie took place in a bike room that reminded me of the free culture activists who tirelessly reduced the transaction costs associated with listening to music on MP3 players like SynapseAI.
Also in the spring of 2003, Adam released an app called BuddyZoo that let AOL Instant Messenger (abb. AIM, which is pronounced like the word “aim”) users upload their buddy lists and then see visualizations of the connections between all of the AIM usernames that had been uploaded. BuddyZoo struck me as an elaboration of both the MetaBrain from SynapseAI and the inaugural Google Programming Contest we had followed back in spring 2002.
Slashdot had reported on BuddyZoo too, just one week before they reported on SynapseAI:
From 2002 until 2005, I took Google as seriously as I’ve ever taken anything in my life. Google was a high quality thought leader, but I don’t think Mark took Google as seriously as I did. Adam was probably between Mark and I on the spectrum defined by the question, “How seriously do you take Google?”
In the summer of 2002, I spent a night at Mark’s parents’ house in Dobbs Ferry, New York while driving back from New Hampshire to my own parents’ house in Greensboro, North Carolina, and we worked with visualizations of SynapseAI data.
Over the next year, I worked on education-related projects in NC. Mark asked me over AIM for advice about how to inspire extraordinary engineers like Adam to focus on his objectives. I also implemented a virtual world using a 3D graphics library called OpenGL, I taught classes part-time in North Carolina, and I assisted track coaches in Greensboro.
In summer 2003, Adam did an internship at the MIT Media Lab and he lived near Mark in a student-housing complex that was linked with Harvard Business School. I made a trip to visit them, and we brainstormed about concepts like social bookmarking and tracking the spread of memes across all the blogs on the Internet. It was a heady time. Blogging was a new and innovative concept, the meteoric rise of Wikipedia had begun, and we were keeping an eye on all of this.
BuddyZoo and SynapseAI, in contrast, had plateaued. They had peaked as happenings on the Internet, but our mood was cheerful. The economy was showing signs of life after scraping the bottom of the barrel in 2002, Mark told Adam and I stories about obscure layers of the music industry he was exploring, and they found a new data source that propelled us toward the logical extreme of tracking the spread of memes directly within people’s brains instead of having to wait until they wrote about the memes on their blogs.
AIM allowed people to leave a custom, “away message,” in the same spirit that people sometimes configure their email accounts to respond with an away message when they’re on vacation. Adam and Mark noted that AIM away messages were like a, “status message,” and they created a daemon, which is shorthand for “backend web service,” that monitored the changes in people’s AIM statuses. It kept track of little changes in people’s statuses, such as correcting spelling errors or editing the language to sound more hip.
That fall, Mark made a fake app which maintained a list of user-supplied AIM usernames. It had an icon that satirized the AIM logo by putting a jackhammer-sized cartoon image of a screw into the hands of the abstract “AIM man” icon. The app encouraged everyone to initiate a chat with whoever was at the top of the list and then “warn” them for inappropriate use of AIM. The list was rotated every hour, so that the username at the top of the list would only be in the hot seat for an hour.
AIM had a policy of temporarily banning accounts that got warned several times within the same day, but I supplied my username to Mark’s fake app anyway. Hours later, a few AIM users I didn’t recognize started conversations with me and promptly warned me, for no reason other than that Mark’s fake app had told them to do this. I didn’t want my account to get banned from AIM, so I signed off for an hour. You could not be warned while signed off.
I’m skipping over a lot of details, but one bit of information about me that contextualizes all the other facts I’m presenting is that, in fall 2003, I started a master’s degree in computer science at Georgia Tech, in Atlanta, where I got involved with the Fulton County Young Democrats and the College Democrats of America.
On one occasion, I distributed Howard Dean fliers at a campus food court. I also appeared with a group of college students who were sitting behind John Kerry at a rally in Buckhead, in a photo on the cover of the Feb 23, 2004 edition of The Atlanta Journal Constitution.
Kerry won the nomination, but Dean, who also ran for president in the Democratic primary that year, made headlines for pioneering some uses of the Internet within a US presidential campaign and for coining the term “Yeagh.”
Mark and I mostly stopped chatting after he became well known at Harvard for launching a “hot or not” site called Facemash, but in the spring of 2004, before TheFacebook was available at Georgia Tech, he asked me to test his latest app’s signup process using my Stanford alumni email address, so I tested the workflow and reported back to him.
We reconnected again in February 2005, and Zuck bought me a plane ticket so I could visit the house in Menlo Park where he, Sean Parker, and Dustin Moskovitz were living and working. TheFacebook had become a force on college campuses, and the user base was growing at an exponential rate. The plan was to figure out if it would make sense for me to join TheFacebook as a software engineer. Adam was off at Caltech taking classes, but he was also involved with my interview, via AIM.
I did some coding tasks at TheFacebook house in Menlo Park. I eventually realized this was not the house in Palo Alto where they had built a zipline. The movie, “The Social Network,” which came out in 2010, had a scene that recreated the zipline in action, but when Adam was actually living there with Zuck, in summer 2004, he had sent me a copy of the sternly worded letter about the zipline they had received from their landlord, and I didn’t really get the message they had moved to a different house until Zuck drove me from the airport to their new house, in 2005.
I also sought out a meeting between us and Prof. Jeff Ullman, who had been my undergraduate advisor at Stanford. Jeff had also been a Chair of the Computer Science Department at Stanford, and, according to Wikipedia, “he was the Ph.D. advisor of Sergey Brin, one of the co-founders of Google, and served on Google’s technical advisory board.”
In 2001, during my 3rd year at Stanford, I sat in on a few meetings of a research group Jeff helped lead. The group was called MIning Data At Stanford (abb. MIDAS). In 2005, I thought Jeff might like to help Zuck find candidates for the role of VP of Engineering at TheFacebook. Zuck commented before the meeting that I was giving him an opportunity to challenge my professor. Could he have been joking?
Prof. Ullman has been called an author of noted computer science textbooks, along with other highly acclaimed contributors to the field, such as Donald Knuth. Challenging the technology industry’s practice of using opaque language in ways that create self-fulfilling prophecies, as I do in the “Postface” section, at the end this text, implicitly makes an example of Ullman, but this problem might have gotten categorically worse if his books had not been written by him, in collaboration with other people from the network of colleagues he had helped build. The strength and weakness of such noted books about compsci is that they kept a lid on the use of a technique called, “storycrafting.”
Studying Ullman’s notes and books, some of which can be legally read for free online, are a good way to learn about data mining.
The meeting began at an office in the Gates Computer Science Building at Stanford, with a discussion about what we now call Massive Open Online Courses (abb. MOOCs). Wikipedia says a MOOC, “is an online course aimed at unlimited participation and open access via the web.”
In 2004, Google had published a path breaking paper about how their engineers were using a technique called “mapping” and then another technique called “reducing” to gather important statistics about the search terms that people type into Google. Collecting detailed information about the steps that many millions of students took as they worked through their homework assignments promised to create just the kind of large data sets that could be mapped and then reduced using these new tools for gathering statistics. Here’s the paper:
Human beings, of course, cannot be reduced to their grade point averages or to any other kind of point in space, not even if that space has hundreds of dimensions. At the time, using techniques like mapping and then reducing didn’t make sense for TheFacebook because they only had about a million users, but now that FB is a big company, I would like to invite people who write fiction to help me re-imagine what the conversation between us might have been like.
The subject matter we discussed, on President’s day weekend, 2005, was too dense to water down for general audiences, but the format used in the play “Copenhagen” might work.
Wikipedia states, cryptically enough, that, “Copenhagen is a play by Michael Frayn, based around an event that occurred in Copenhagen in 1941, a meeting between the physicists Niels Bohr and Werner Heisenberg.”
We weren’t discussing whether or not to build nuclear bombs. We were discussing the future of education and TheFacebook. It’s a coarse analogy, but increasing awareness about this Wikipedia page might be worthwhile, in and of itself. It won the Tony award for best play in 2000:
Zuck and I also had dinner in downtown Palo Alto that weekend, and he played a song by one of his favorite bands at the maximum volume during the drive back.
I didn’t join TheFacebook, and this non-event is now a case study in storycrafting. The main reason I didn’t join FB in 2005 was that I had a disagreement with myself. The real reason was that I had made a commitment to teach AP Computer Science at a high school in Atlanta, and I thought it would be bad if I didn’t follow through on this commitment, on account of a deep bench of reasons having to do with concepts like honor and integrity. In reality, I thought seriously about dropping everything and working out a deal with TheFacebook before heading back to ATL airport. I was eager to underwrite TheFacebook.
According to Google search, to underwrite is “to sign and accept liability under (an insurance policy), thus guaranteeing payment in case loss or damage occurs.” Zuck wasn’t asking me to write him an insurance policy that would pay out in cash if a big company railroaded his project. What he might have indirectly wanted was an option to pivot if TheFacebook’s cash flow ran dry, and I was happy to sell this option. We couldn’t agree on the price of the option, and it didn’t occur to me that giving him access to data about this specialized market was valuable.
I took the overnight flight back east with a renewed focus on building self-directed learning software. Adam, Mark, and other students I had taught, who had also made use of computer labs outside of class, had advanced while competing in computer programming contests like TopCoder and the USA Computing Olympiad. The programming contest community was a champion the of the self-directed learning ethos, and the strong ability of TheFacebook team to raise venture capital had inspired me to keep studying the educational software space.
I’m often asked questions about the movie, “The Social Network,” and so I’d like to clarify that I think the movie got Sean and Zuck wrong.
Zuck’s strategy struck me as equivalent to following the lead of an artificial intelligence engine that hustled big tech companies by making fun of the structure and function of their core businesses. No clear line was drawn between following best practices in the industry and sarcastically pretending to follow best practices in order to reveal bugs in other people’s computer software.
As the leader, as “the producer,” and as the driver of the shiny new Facebook company car, Zuck’s thought process would have been hilarious if he had only punked the missteps of multinational corporations. Conversely, The Social Network made it look like Zuck and his friends weren’t edgy enough to stage derivative satire about individual people, but in reality, Zuck had been carrying himself with confidence since high school. The movie downplayed Zuck’s social awareness.
In general, big companies often make themselves seem warm and fuzzy, especially when their representatives appear on main stream media channels, but the way they make money is usually rather technical. For example, Microsoft Word emphasizes its features that help people express themselves with spunky, offbeat style choices. Word also leverages data, itself, as a platform.
Many law firms use Microsoft Word to handle legal documents. It’s easy to fall into the trap of imagining firms use Word because each individual lawyer prefers using Word to edit legal documents instead of using Google Docs, but most lawyers and the clients they work with don’t have a choice. Most people edit legal docs with Word because most of the people they’re collaborating with use Word. Unless you have a specialized reason, using a program other than Microsoft Word to edit legal docs might even get you tagged as “unprofessional.”
If you want to make a change to a legal doc that’s stored as a Microsoft Word document, the path of least resistance is to buy Word so you don’t have to worry about changing the document in some abnormal, unintended way.
Officially speaking, legal documents are a type of data, and the data type “Microsoft Word document” uses the power of computer technology to process the data type known as, “legal document.” In this way, Word becomes a platform, legal documents become an application of the platform, and lawyers become evangelists for Microsoft Word.
I wrapped up my trip over a low key steak dinner in Menlo Park, near the breakfast joint called Stacks. When the food arrived, I thought about the meme of jet setters who throw food at the walls of public restaurants and rely on professionals to swoop in and clean up the mess.
Then I thought about evolving TheFacebook from a parody of the way Microsoft Word leverages data, itself, into a parody of how Microsoft Visual Basic empowers the Microsoft developer community to shape Microsoft’s technology ecosystem, so that I could build a custom type of Facebook profile page for the Democratic party affiliated social clubs I was involved with in Atlanta.
This was an intuitive next step, but I wasn’t sure how to sell TheFacebook team on following that path, so the food fight was resolved along the lines of, “Walking before running.”
Sean helped create a safe space that allowed TheFacebook to aggressively challenge much bigger companies, and he must have been aware that Zuck was producing satire. In a word, he was “driving.”
Zuck was the comedian. Sean was the instigator, and I mostly felt confused during the weekend I spent hanging out at TheFacebook house in 2005. There’s a lot I still don’t know.
Later on, in spring 2005, I reverted to my default plan, which was to be a software engineer. I pinged Zuck and he indicated FB was not hiring engineers at that moment, so I got a job as a research programmer and learned to windsurf. I took a couple econ classes and got interviews with big tech companies, through the Georgia Tech career center.
Google flew me out to SFO airport in October for a second round interview, and Adam, who had stopped out from Caltech in order to be the Director of Data Mining at Facebook, helped me prepare by staging a mock technical interview the night before. I did a parody of what political operatives do with the confidence that nobody can make a move against them without causing a bunch of machines to seize up and malfunction.
The following evening, while hanging out at a cafe near a bookstore, Adam and I crossed paths with a person who had shared the formative experience with me of working together to solve Association for Computing Machinery (abb. ACM) programming contest style problems at Stanford.
We all asked each other software engineering interview questions, and then I was invited to interview at Palantir, which was a startup Peter Thiel had co-founded with Stanford Law School classmate Alex Karp and some friends of friends, for the purpose of building an analytics platform that was inspired by the fraud detection system that had been created at PayPal.
I squeezed in an interview at Palantir before flying back to Atlanta, and I then I dropped everything and joined their team as soon as possible. I didn’t wait to hear back from Google. Theil had co-founded Paypal, he had made the original 500k angel investment in Facebook, and Palantir was the first company in Silicon Valley that offered me a job.
It’s also worth nothing that if Facebook had been ready to hire me in 2006, I might not have gotten the chance to meet Peter Thiel, but at Palantir, I got to meet him twice.
The order of my in-person conversations with Thiel was not important. On one occasion, he invited the team to celebrate the closing of Palantir’s series A round of funding by eating sushi and hobnobbing on the roof of a mansion he was renting near Crissy Field, in the Marina district of San Francisco.
There was an opening to wait in line and ask Thiel a question, so I followed the unwritten rules and asked him if he thought Microsoft stock was a good investment. He appeared to take the question seriously.
After thinking carefully for a couple seconds, Thiel said, “No.” I’m not 100% sure he said no, but I thought his overall response was negative. I felt he didn’t want to answer any more questions like that.
The other time I made linguistic expressions that were intended for Peter Thiel to process in person was when he made an impromptu visit to the Palantir office. He was wearing gym clothes and started making small talk in the kitchen area. Most of the engineers took a break from coding and wandered into the kitchen. Others played it cool and kept on coding.
Someone asked Thiel what he thought about the housing market, and he said something to the effect of, “The housing market is functioning as an intelligence test right now. You guys are all intelligent, by definition, so let’s do an experiment. Who here owns real estate?”
I had bought a condo in Atlanta in 2004, and I was having trouble selling it, so I raised my hand. I was the only person in the room who did, and I thought I heard Thiel say, “You’re the exception that proves the rule.”
Working at Palantir in 2006 was like having a VIP ticket to the startup events in the bay area. At one event, I focused on gathering data about a startup called Powerset. They were building a natural language search engine that got acquired by Microsoft in 2008. I also asked a person who looked like Sean Parker to sign one of my Palantir business cards, to no avail.
I attended a talk Zuck gave at an event called “Startup School” that was hosted at Stanford by a recently created startup incubator called Y-combinator. Adam, who had graduated from Caltech and taken on the role of Chief Technical Officer at Facebook, stood in the back row, and Zuck drew attention to him at the start of the talk.
During the breakout sessions, I finagled my way into a small group discussion Zuck was leading. I didn’t really have a valid reason to be there, but I had credentials and I knew the terrain. I wanted more data about what Zuck was thinking and doing. Zuck responded by saying something nice about me to the group.
In February 2007, Facebook sponsored an event called, “The Technology Tasting.” My understanding of the conventional wisdom was that early stage startups like Facebook and Palantir lived or died by winning a zero sum game to hire the people who could get through their interview processes, but I coached myself into believing I could help both companies grow the pie and recruit more engineers.
I arrived at the event expecting to be almost invisible, which I viewed as a fact of life that was tolerable in Silicon Valley. Being almost invisible signaled that I had been invited to the party because someone thought I was a highly productive engineer, but then a conversation took place after Zuck took steps in my direction, while I was just standing there.
I was more visible after spinning with the CEO in front of everyone, and it was easier than usual for me to work the room and talk to people like Scott, who had an engineering master’s degree from Stanford and had gone to Georgia Tech for undergrad, which was the opposite order in which I had studied engineering at those schools, and I thought I heard Scott say he was amped about the projects he was working on at Facebook. Another engineer, who seemed like a DJ, projected similar energy, but what really took the cake was when I thought Matt Cohler, who is now a well established venture capitalist, helped me cook up a new modality for making light conversation at parties in Silicon Valley.
Our modality was like improv comedy theater, but without the carefree attitude that typically leads to comedy. Our style of improv was more serious and mission-driven. This was the capital of startup land, so it was fair game to explore the narrative that, “Companies of all different sizes often avoid new ideas or approaches because they don’t want to risk cannibalizing their existing businesses, and this widely felt and potentially irrational fear of cannibalism creates many opportunities for startup companies to be disruptive.”
We did that, but our theatrical production also looked at the more serious question, “How do we write this script so that both Palantir and Facebook survive, even if Todd’s character can’t imagine Palantir engineers saying they are amped about work without cracking a smirk?”
In the next act, a person who had crossed paths with Zuck at Harvard and then joined FB as part of a recruiting coup, in which several people who were living in Seattle all joined FB at the same time, marshaled a slide show about the social network’s latest growth numbers, and I evaluated the pros and cons of this investment opportunity carefully, much as the Palantir engineers had, individually and collectively, when a representative from a financial services provider gave us a presentation about the 401k plans the company was making available.
I don’t recall talking to Adam at the technology tasting, but we intermittently ate in Mountain View and went hiking at places like Russian Ridge, where we did a parody of running, and I learned about people like Scott, who had helped build the first version of FB photos and was credited on the patent they eventually got for the idea of tagging people in photos. For background:
The tasting allowed me to connect the names of people from Facebook, which had been floating around in my mind, with their real life faces and personality traits, and I concluded that anyone with a healthy appetite for risk would soon experience the winner-take-all effects, which had driven most college students in America to join FB, that were about to affect the entire population of people who had Internet connections.
When I got back to my apartment, I kept playing melodramatic songs with cheesy lyrics and imagining what it would be like to leave Palantir, with its picture perfect offices on Page Mill Rd., near an old Hewlett Packard building, where I used to walk around and meditate about how I was standing on the shoulders of giants, but I was drawn in by the art of chasing fame and fortune in the more fully baked world of downtown Palo Alto, so I made the move.
When the time came for me to do an onsite interview at Facebook, I passed my co-workers the narrative that I was going to the gym, which meant the gym near San Antonio Rd. where I had tried personal training for the first time. In reality, I turned left instead of right at the intersection of Page Mill and El Camino Real, and I drove toward Facebook’s office at 156 University Ave.
As a side note, I’d like to create media for general audiences that showcases the essence of great technical interviews, but the conundrum of authoritatively framing such a production runs deep. The solution might be to form a scripted dramatization that melds my interview at Facebook in 2007, my interviews at Palantir and Google in late 2005, and an interview in 2001, at one of Microsoft’s Silicon Valley offices, that I got through the Stanford career center.
After receiving my offer letter from FB, I didn’t feel comfortable acting like everything was normal, so I used AIM to request a meeting with my boss, who was at his desk, about ten feet away. We stepped outside and I explained my reasons for defecting in a heartfelt way. The matter was escalated to the CEO, and we spoke candidly to each other. I had already scheduled a meeting with the FB team to sign their employment contract the following morning, so I left the premises.
Then I texted Zuck and requested a meeting. Zuck said he was very busy with CEO stuff, but I kept on using both voice and text to request a meeting. Before signing, I wanted to establish a data sharing protocol that would make the deal a bigger win-win for all of the interested parties.
If the events of that day were to be depicted in a movie, a crude directorial frame would be to cast me as a super intense nerd who was very determined to challenge CEOs to chess matches, but at the time, I felt more like an at large individual who was processing a truckload of new data along the same lines that Senator Ted Stevens from Alaska did when, in 2006, he gave a much ballyhooed speech before congress, where he compared the Internet to a “Series of Tubes,” as described here:
Around 11pm, Zuck replied that his meetings had concluded and he was hungry. I picked him up outside 156 and we drove to the McDonald’s on Rengstorff Ave. A cliché classic rock song was playing on the radio. Zuck didn’t seem to like my taste in music, but he reassured me that it was the norm in Silicon Valley to have a long-term focus.
I had been unusually awkward in college. I was 26 when I joined FB, and there was still awkwardness whenever I talked with people. Everyone except me seemed to know why. Even in Silicon Valley, it felt like everyone had learned a crucial life lesson I hadn’t. The issue was often framed as though I was searching all over the house for my car keys when the keys had been in my pocket the whole time.
I wanted that analogy to be true, but I could never find my virtual keys, so I pretended to have found them. I sold that bogus narrative by making fun of myself for stressing out over nothing, and then I did hard workouts, travelled alone, demonstrated my sanity by playing board games, and went for long walks, instead of driving my virtual car.
Even after I got equipped with healthy coping mechanisms like that, making small talk still felt like an ordeal. Notwithstanding, I had gotten a confidence boost at Palantir, and I wanted to make a generative attempt at addressing my problem with social awkwardness. Palantir was well run, but after working there for over a year, I felt like the central problem in my life was only becoming more difficult to untangle. I didn’t know how to keep doing my job without taking on unsustainable psychological debt.
Zuck had a promising solution. I was hopeful that Facebook would turn out to be like a Nash equilibrium, but for solving self-referential problems that lead to social awkwardness, in the sense portrayed in the movie, “A Beautiful Mind,” in which the character of Nobel Prize winning economist and mathematician John Nash, played by Russell Crowe, talked about seeking to understand, “governing dynamics.”
After collecting our chicken nuggets, fries, and ketchup, Zuck suggested we sit on the curb outside. He told stories about the business world that helped me understand why nuances matter and why FB wasn’t ready to hire me two years earlier. In particular, he told me a story about Steve Ballmer, who was the CEO of Microsoft at that time.
I assumed Zuck meant to reiterate that it was ok to feel socially awkward and then help me understand a strategic consideration that had come up in a high-level conversation.
An alternative narrative is that Zuck was framing me as Steve Ballmer. Microsoft’s business model had survived the dot com bust intact, and it was humming along nicely in 2007. The CEO of Microsoft had currency, and Zuck was open to spending it on his behalf.
For all his flaws, the way Donald Trump used language throughout the 2016 US presidential campaign made it effortlessly valid to share fact-based examples of how framing works within real world conversations.
In the Republican primary debate that took place in South Carolina, Ted Cruz made a disparaging comment about how, “Trump represents New York values.” Trump retorted that William F. Buckley, an important conservative thought leader, had hailed from New York, and this was a strong frame, if not the strongest possible frame an artificially intelligent robot could theoretically construct, to drop onto Cruz. Cruz was representing the vibe of, “I’m really good at winning intellectual arguments,” and Buckley had represented that same vibe.
People who knew about Buckley started to subconsciously associate Cruz with Buckley, for better and for worse.
Talking about the deep hidden meaning of Trump’s debate performances is an example of a technique called, “flexing on your audience,” and using the term “flexing” in the context of lounge singing transfers heat to both body building and pro wrestling, without actually addressing the core challenges before us.
With those concepts in mind, I propose tabling the Q and A about, “Why was Trump’s frame strong?” until the second part of this text gets built and gains traction. I’m downshifting, conceptually.
A more visceral example of framing occurred a couple months later when Cruz responded to Trump’s incessant interruptions by repeatedly telling Trump to, “Breathe.” Cruz was standing next to Trump on the debate stage and his energy was condescending beyond all recognition. The most obvious interpretation of this bizarre way for one powerful person to talk to another powerful person was that Cruz was framing Trump as a trauma victim who was enduring a panic attack. It wasn’t clear how Cruz was framing himself, but then Rubio interjected, “When they are done with the yoga, can I answer the question?”
Rubio’s words framed both Trump and Cruz as participants in a yoga class, and it made Rubio seem either unwilling or unable to hint at the awkward truth. Cruz’s condescending attitude had vetoed the possibility that he was going to pull off transformative satire about yoga in order to keep the peace, and Trump wasn’t playing along.
With a force that was minimal to none, Zuck’s story about the foibles of Steve Ballmer drew attention to the similarity between Ballmer and myself. My conscious mind didn’t consider the possibility, but I was star struck and my subconscious encoded a new rule that said, “I am similar to Steve Ballmer; don’t throw chairs.”
I’ll unpack those subconscious rules in reverse order. First, a hard boiled story had circulated about Ballmer throwing a chair in 2004 when he learned that Google had poached a Microsoft engineer. From Ballmer’s Wikipedia page:
“Mark Lucovsky, who left for Google in 2004, alleged in a sworn statement to a Washington state court that Ballmer became enraged upon hearing that Lucovsky was about to leave Microsoft for Google, picked up his chair, and threw it across his office…”
Second, Ballmer had given a speech at a Windows 2000 developer’s conference where he had sweated profusely and shouted the word, “Developers!” over and over again. The non-verbal aspects of the speech, which were not nuanced, showed that Ballmer cared deeply about developers, and the video went viral.
Ballmer was talking about third party developers in his “Developers!” speech. From Microsoft’s perspective, third party developers are computer programmers who don’t work directly for Microsoft, but they play a key role in the ecosystem where Microsoft thrives, somewhere near the top of the food chain. Third party developers write software applications that elaborate Microsoft’s strategy for leveraging data, itself, as a platform.
Using computer technology to encapsulate real world data types like legal docs, financial models, and small business data is a never-ending process, and that dynamic is easier for the platform owner to modulate if their developer community is always elaborating, like a well-oiled machine.
Did Zuck want me to bring Ballmer’s suspicious lack of glib confidence into the culture at Facebook? Facebook was about to launch a developer platform. They needed to build a developer community. Facebook engineers where hip, and they wanted Facebook platform to be hip, but there was going to be overlap between Microsoft’s existing developer community and Facebook’s unprecedented type of developer community. By Facebook’s standards, Microsoft’s developer community was not hip, but my thoughtfully orchestrated efforts to overcome an unknown source of social awkwardness had grown into a tried and true methodology for bridging this gap.
I was framed as Ballmer. It doesn’t matter if it was done consciously or subconsciously. It was encoded subconsciously. As we worked our way back to downtown, I made full stops at all the four-way stop signs on the surface streets of Palo Alto, and then I started underwriting Facebook’s next move.
I had learned how to wheel and deal, but there was a lot I still didn’t know about games.
During my first few months at FB, I developed pain in my wrists that almost degenerated into repetitive stress injury (abb. RSI). I had been dealing with RSI issues for several years by that point. Palantir had bought a custom ergonomic chair that solved my last round of problems with RSI. It took me several weeks to order a similar chair at FB, but giving my RSI an opening to resurface inspired me to try yoga, in addition to running and weight lifting.
My problems with RSI went away around the time Facebook released the update to their developer platform at a conference called F8, in San Francisco, on May 24, 2007. Facebook had released the first version of their platform in 2006, but the 2007 version had more integration points with the user interface elements that had powered the viral growth of the fast moving social network.
In preparation for F8, we had a special hackathon where everyone built platform apps that made use of the new integration points, and my project evolved into a “Step by Step” tutorial for building Facebook apps. I posted the tutorial using my personal web server rather than going through the usual channels to release content at FB. The tutorial got thousands of hits at F8. We also moved it to the official FB developer’s website.
During work hours, I joined a team that crafted software solutions for internal use at Facebook. Our primary mission was to help the customer service team function effectively at scale. Customer service, in turn, performed a hodgepodge of tasks that included defining what it meant to use Facebook appropriately and then disabling inappropriate accounts.
I gradually signed up to fight spam. I wanted to build an expansive platform for developing self-organizing flocks of applications that leveraged Facebook’s growing toolkit of state-of-the-art machine learning capabilities. The problem statement changed when more people got involved, and the team transitioned to another approach while enjoying dinners out on the town, preparing slides for recruiting trips, and go-kart racing during company offsite adventures.
I also engaged in conversations about how Facebook platform could transform medicine, education, local economic cooperation, and the 2008 US presidential election. I worked on a widget that helped people register to vote, and we amplified the widget’s visibility in Facebook’s Newsfeed feature in ways that were not available to third party app developers.
Finally, I graced, tilted, countenanced, and cajoled the VIP lounge at the hotel in New York City where we had monitored the 2008 Democratic and Republican presidential primary debates, with the group of people that followed through on Facebook’s involvement with those debates.
In conjunction with activities like that, I learned about the whole frame of American culture and stashed it away in my mind, like a git repository.
Git is a version control system that was developed by Linus Torvalds in 2005. Linus was already well known by then for developing the Linux operating system. According to Wikipedia, Torvalds “quipped about the name git” (“which means unpleasant person in British English slang”): “I’m an egotistical (EXPLETIVE DELETED), and I name all my projects after myself. First ‘Linux’, now ‘git’.”
Git is useful for managing documents and notes, as well as computer code. I used it while writing this text.
To the extent that this text is like an old school detective’s bulletin board, using git to record my progress was like taking a photograph of the bulletin board every time I made changes to the photos and pieces of paper that were pinned to the bulletin board.
Carving out the time to learn new tools like git is an example of making a self-reinforcing investment in yourself. You could also call it, “optimizing for learning,” if you prefer to use language in a way that pokes fun at both the history of the technology industry and the future of industry, in which platforms that optimize for learning will have agency as self-reinforcing patterns.
Sam Altman talks to a lot of people who probably optimize for learning. Altman was co-founder and CEO of a social local mobile app called Loopt and is now president of Y-combinator as well as co-chairman of an initiative Peter Thiel is involved with called OpenAI, which describes itself as, “a non-profit AI research company, discovering and enacting the path to safe artificial general intelligence.”
I’ve never asked Altman for feedback, but I’ve had verbal exchanges with Paul Graham, Sheryl Sandberg, and Larry Summers. Graham, at an event I’m 95% sure was at Cubberley Auditorium at Stanford, a few weeks before the F8 conference in 2007, Sandberg, at a talk she gave at Facebook in 2009, where I asked her a question about overcoming the “success and likability penalty” that women face, and Summers, when I asked him a question after he spoke at an event that took place in connection with the 2004 College Democrats of America convention, in Boston.
Paul Graham was a co-founder of Y-combinator who built his personal brand writing books about computer programming and articles such as, “Hackers and Painters,” which appeared on his blog in 2004. Sheryl Sandberg joined Facebook in 2007 as the Chief Operating Officer, and I’m under the impression she’s second in charge at Facebook. Larry Summers was the president of Harvard University at the time when Zuck launched FB, and there was a scene in the movie, “The Social Network,” where his character, played by Douglas Urbanski, discussed the merits of TheFacebook with the Winklevoss twins’ character, played by Armie Hammer.
I thought Larry’s answer to the question I asked him in the summer of 2004 was, “It’s complicated.” My question had to do with both software patents and pharmaceutical patents. I tagged the drug companies, to elicit a stronger response from my audience. Sheryl’s answer to the success and likability penalty included the idea of, “raising awareness about the problem,” and my exchange with PG was a function which takes a boolean argument that represents whether or not the git commit history for this text has been published.
Knowing how to use git is also an example of a “hard skill,” in contrast to “soft skills” that facilitate effective communication. Approaching Summers the way I did in 2004 demonstrated a hard skill. My question challenged him in a context where engineering credibility wasn’t a substitute for self-awareness.
I had a deep stack of hard skills for avoiding, “fake it until I could make it,” scenarios, but I lacked soft skills. Digging in to fight the good fight with hard skills kept me afloat, but my unexplainable and self-reinforcing scarcity of soft skills was becoming a problem.
The conventional wisdom I kept hearing and trying to take seriously framed the issue as, “I was shooting myself in the foot,” but I wasn’t actually shooting myself in the foot, so I wasn’t sure how to stop, “shooting myself in the foot.”
An oft-repeated quote, most commonly attributed to Albert Einstein, says, “The definition of insanity is doing the same thing over and over again and expecting different results.” Gloria Steinem, a noted leader of the feminist movement, etc., said it was ok for people to wear high heels, while giving a talk at Facebook that I attended, and I started to go out dancing in San Francisco while dressed up in a blonde wig, full makeup, high heels, and a dress.
Whenever I wore traditionally feminine attire, I made rapid progress at learning soft skills. My progress regressed if I went back to jeans and a button down shirt from the men’s department, but dressing up gave me a way to understand why it’s normal to act like soft skills emerge naturally.
Dressed as “Suzy,” I learned soft skills in a fun, life-affirming way, as opposed to learning them in an unhappy, tactical way was just a facade for hard skills. I hadn’t really experienced what that felt like before, and certainly not since college. Suzy has now become an aspect of my life.
Initially, I went out at night and joked around that my name was Suzy. Most people imposed heavier concepts onto Suzy than I did. Some people said Suzy was my, “alter ego.” Others asked if I was a female who had been born in a male body. To me, putting on a wig and eyeliner was like learning a hip computer programming language, combined with the risk profile of going on a road trip instead of staying home and using the computer all day.
The first time I dressed up as Suzy during the day was at the Stanford affiliated mall that’s situated near the intersection of El Camino Real and Sand Hill Rd. The cover story I planned to use, if confronted, was that I wanted to find out if my credit cards still worked when I was dressed as Suzy and attempting to use them in person. In reality, I was doing a one-time experiment, to gather data about all the possible risks I might be taking on by cross-dressing.
Doing the same experiment over and over again without legitimate underwriters is not only unethical, it’s dangerous. Similarly, running scientific experiments without informed consent is a textbook example of a behavior that creates self-reinforcing risks.
Running scientific experiments without taking the proper precautions is like lying systematically, one-time experiments are more like jokes, and I might run a contest to find the best closer for this paragraph. I’m contemplating ideas like, “Art is the science of not running experiments.”
The second time I dressed up as Suzy during the day was at the San Francisco Pride celebration in 2009. I was able to socialize effortlessly at the event, so I took pictures and uploaded them to Facebook. My co-workers didn’t seem to mind, and I considered putting Suzy in the same category as ultimate frisbee, which is also described as, “ultimate.”
Facebook had a well-designed amateur sports program. We challenged much bigger companies like Google in ultimate, our intramural basketball league was excellent, and we had an annual tradition called, “Game Day,” where everyone was invited to compete in events like capture the flag and racing on two-wheeled contraptions.
At the 2009 game day, I saw a group of people sitting in a shaded grove around an author named David Kirkpatrick. He was talking about his projects, one of which was to write a book about the history of Facebook, and I felt the group was competing for his attention. It would have been hard for me to say words that were intended for him to hear and process, and I chose not to be ironic and talk loudly and cheerfully to a tree, eight years ago, about the collaboration between Zuck and Adam from early 2002 to late 2003.
I continued going out in the evenings, and I wore heels. I was still living in Palo Alto, but the new data I was getting while wearing heels in the city was worth the trouble and the risk. I didn’t use the new data to flex at work, but I began pretending I was dressed as Suzy while taking power yoga and hiphop dance classes at the gym near the office. I was starting to figure out that acting like Suzy was more impactful than dressing like Suzy.
There was a meme within Facebook called Corporate Friday’s that normalized engineers who wore dress clothes to the office. Some people dressed up more than usual on Friday, but it was optional. I never did wear heels to work at Facebook, but if I had, it would have been awesome.
And, knowing what I know now, the way to optimize for collaborative learning on Friday’s at Facebook would have been to stay at a hotel in Palo Alto and hire a private car service to ferry me back and forth from work that day, while dressed as Suzy.
In 2009, I struggled to make sense of body language. I also lacked awareness of all the different ways utterances can be interpreted. I absorbed important social data in a more photographic way, and I eventually saw patterns that had once eluded me. In a world where just two words like, “That’s all,” can be a reference to movies about people who wears high heels, patterns become material.
When I approached Larry Summers in 2004, I have a vivid memory of what his body language looked like. It would have been more amicable to disclose to him that I had been one of Mark Zuckerberg’s computer science teachers, that I was using my brain to video tape him, and that I had developed an interest in journalism that year.
Dressing up as Suzy for the last seven years allowed me to experience what it’s like to comprehend body language and make small talk. People share more data with me when I dress like Suzy, and getting data from original sources helps with managing expository frames that might stick, in the sense used by the book, “All the King’s Men,” by Robert Penn Warren.
Of all the books I read for high school English class, “All the King’s Men,” was my favorite. I didn’t make time to read, “All the President’s Men,” by Woodward and Bernstein, which tells the story of how they exposed the Nixon administration’s involvement with a break-in at The Watergate Hotel, until I bought a condo in Atlanta, in 2004.
If I had worn heels while posting flyers for Howard Dean at Georgia Tech, that would have been a cheaper way to get the confidence boost that inspired me to ride a bus to Boston in the summer of 2004 and give Larry Summers a piece of my mind about TheFacebook. Buying a condo was an alternative way to get that boost.
Just before Halloween, 2009, I got seen looking at the island of Fiji on Google Maps. My condo had sold in the summer of 2008, and Facebook stock was rising. It was time for me to leave FB.
I didn’t care about surfing in Fiji, but I lacked the soft skills to continue standing tall. I needed a way to lie down and soak up the rays of the sun, and pretending I wanted to go surfing fit the bill.
I exercised my options after I left Facebook. I was allowed to buy a bunch of shares of Facebook stock at the fair market value (abb. FMV) that had been determined roughly three years earlier, in 2007, when I had signed Facebook’s employment contract.
FB had not yet made shares of its stock available for purchase by the population at large, but there was a way to sell my shares in, “the private equity market.” I sold half of my shares, and then I sold half of my remaining shares, and so on and so forth, ad infinitum.
I caught up with the game I didn’t play in college, but there was a lot I still didn’t know about “nature.”
Fast-forwarding, Facebook is still the leading social media company, and they’re poised to win the race to build strong AI, which is a commonly used term that means, “machines that might be more intelligent and aware than humans.”
I think the AI machines of the future might prefer my definition of strong AI over the other definitions I found online, so I cited this document, itself, as a source for the definition of strong AI.
Facebook’s juggernaut trajectory triggered a shift in the stance of America’s increasingly global power structure, but Facebook’s founders didn’t use familiar, ordinary power plays to push around the titans of Wall St., among other centers for the distribution of power on planet Earth. They elaborated patterns of human behavior I’d seen before at places like Harvard and Stanford.
In my opinion, the champion of human behavior research is Michael Lewis. In 1989, he wrote a book about his experiences working at a bank in New York. The book’s name, “Liar’s Poker,” derives from a story about a game called “liar’s poker” that two commanding individuals didn’t play, when they couldn’t agree on an amount of money they would each be willing to risk losing without getting upset or feeling insecure, while playing the game.
Liar’s poker requires each player to hold a dollar bill with a unique serial number that’s kept secret from the other players until the game is over. If the people in question had played, they would have each had to reach for a dollar bill.
I’m curious about the set of dollar bills that were in the room when the game of liar’s poker that didn’t happen, happened.
If such bills are out there, I would pay more than the face value of the bills to acquire them, but only if the sellers had a way to verify the pedigree of the bills. They would need to prove that each bill was in the room when the famous story featured in the first chapter of, “Liar’s Poker,” occurred.
I would be paying to codify the process they had used to establish and verify the pedigree, and it wouldn’t matter if they had made that information publicly available or kept it quasi-secret with prospective buyers.
Access to the apparatus for establishing and verifying the pedigree of the bills in question is the commodity that has currency. The actual “print on paper” bill is material, but the apparatus is what I’d like to preserve.
I didn’t get around to reading Michael Lewis’s books until an Exeter classmate told me I should, during summer 2003, while we were grabbing coffee in Harvard Square and I was telling him about these kids named Mark and Adam who I thought were on the fast track.
In retrospect, the moment when I became interested in reading Lewis’s books was after Mark Zuckerberg and I made a $1 bet in early 2002 that he would not be able to use Microsoft Visual Basic, in a timely manner, to write a computer program that could evaluate math expressions in the same way that his Texas Instruments graphing calculator evaluated math expressions.
One evening, Mark walked into the compsci lab, which was full of students working on their homework. He declared that he needed to write a custom math expression evaluator within one hour, and I predicted he would fail to achieve that goal. He disclosed that Microsoft Visual Basic (abb. VB) was his favorite programming language and that he would be using VB to write the evaluator. We made a bet that I would pay him a dollar if he could write a custom math expression evaluator within one hour. There was no contract. I don’t recall if we shook hands. It was understood that the bet would settled in an orderly way.
An hour later, I paid him $1. Three years later, I made a non-bet. I declined to underwrite some type of opportunity risk that TheFacebook was looking to guard against in February 2005. Technically speaking, I made a move that was orthogonal to what the big banks in NYC were doing at the time. Orthogonal means, “perpendicular,” but in more than two dimensions. I didn’t do the opposite of what the banks did. I made a move that was orthogonal.
Prior to its collapse in 2008, the big banks had been eager to underwrite the sub-prime mortgage market by putting their own skin into the game of loaning people money to buy homes they could not afford. You can learn more about why that happened by reading Michael Lewis’s books.
Fortunately for me, most Facebook stock holders faired better than the average person who had exposure to the housing market in 2008. In the fall of ’08, the price of Facebook’s privately traded stock took a hit, but it recovered.
Facebook found a way forward after that, and the value of its stock grew by several orders of magnitude. If you use Microsoft Excel to build a statistical model of historical data points that document whether or not it paid to bet against Mark Zuckerberg between the years 2002 and 2017, the model’s answer to the question in question will be, “No.”
I, nonetheless, sold all of my FB stock rather early. I was not interested in the part of this narrative where, “Facebook found a way forward.”
As for the bet I made with Mark Zuckerberg in 2002, I was biased because I had programmed text adventure games in the 5th grade using Microsoft Quick Basic. In 9th grade, I used Microsoft Visual Basic to write a program that did some of my geometry homework for me. I was interested in artificial intelligence, and writing a program that could do geometry proofs was a step in that direction.
Geometry proofs feature more natural language expressions than math expressions. Most people learn to evaluate math expressions by hand in pre-algebra, and knowing how computers evaluate math expressions was my day job. I was passionate about what computer jocks call, “natural language processing,” and I had a shelf full of compsci textbooks to prove it.
What Mark did and thought about while building TheFacebook and converting it to Facebook was fascinating, but I had a limited view of the data. I put Humpty Dumpty back together by simulating data mining techniques in my brain.
My inability to understand what was happening from 2002 to 2010 characterizes my quasi-religious faith in natural language processing technology as buggy. I’ll explain what it means for an idea, belief, artifact, etc. to be buggy in the second part of this text.
I’ll also keep searching for different words than “natural” to describe how politicians use language. There’s a lot we still don’t know.
__Part 2: STORY__
In the first part of this text, I summarized a bunch of important facts about the history of social media using the real names of some early Facebookers. On the one hand, I don’t advise sharing personal information within the documentation of an open source software project.
On the other hand, a lot of people are suffering, and Donald Trump is the sitting president of the United States. I’ll take a stab at writing transformative satire that might help us understand how we got here. The data I shared in Part 1 was material. I’ll begin writing this story, from Suzy’s perspective, by introducing one last fact:
In June 2001, I was playing hide and seek at night and another guy’s forehead accidentally bashed into my right cheekbone. We were both running, and I didn’t see him coming. I don’t think he was injured, but we didn’t talk afterwards. I went back to my dorm, and friends drove me to the emergency room at Stanford Hospital. Several bones on the right side of my face were broken, and the doctors did reconstructive surgery that was helpful but didn’t restore my face to its original shape.
This event actually happened. My facial injury is a physical artifact. It’s subtle, but people with normal vision can perceive it from more than several feet away. I have a copy of the doctor’s report in my files. My life was relatively easy and happy go lucky before the injury, but, until fairly recently, it’s felt like an uphill battle. Now that this text exists, I can speculate that the quasi-stable hierarchy of variations in how people react, as individuals, to my facial injury is the “massively multi-playered” independent variable at the crux of this narrative.
Whatever the case, I plan to become a stylish woman named Suzy, who has a degree from an Ivy League school and got a job teaching high school computer science in the fall of 1994. Suzy has the same facial injury as me, but she got it while playing ultimate frisbee on the east coast instead of while playing hide and seek on the west coast. She was 21 in the fall of 1994, which makes her eight years older than me. Suzy has never been pregnant or given birth to a child. She has been an actress in film. Suzy plays poker and has thought a lot about creating virtual currencies that are more like baseball cards than Bitcoin.
Suzy will need to know how to program in the Pascal programming language. The College Board required AP Computer Science students to use Pascal up until the mid 90s, and Suzy had learned the language when she, herself, had taken the AP compsci exam, while growing up in the Sunset district of San Francisco. From a character development standpoint, Suzy grew up with an edginess that I didn’t. Her story is about containing that side of herself. My story is about reverse engineering her edge, in order to model and simulate the effect that my facial injury could have on the world. It took me fifteen years to see the pattern, but it’s there. This text is part of a process I’m following, with the intention of helping other people see the pattern.
Pascal was named after the French philosopher Blaise Pascal. I don’t speak French, but Suzy does. She is credited as a contributing translator of the French-language versions of more than one noted book about consciousness and cognitive science. She daydreams about running experiments at her favorite shopping center in Paris. Under a pseudonym, she helps write a soap opera that’s based on the culture of palace intrigue at Versailles. As a kid, she made a fake scrapbook about a summer vacation to the south of France by taking photos of photos. She self-published a manual on how to mash up photos and she then started a zine that became the foundation for an underground scene that underwrote her first virtual currency. She draws pictures of ships in her private journal. Suzy is a woman in tech.
There are ways to get a degree from an Ivy League college with Suzy’s name on it, but no history making boarding school in New England will ever falsify their records to make it seem like I was a teacher there in 1994, so I’ll need to conjure up a school. To that end, the word “const” has a specific meaning within many computer programming languages. In English, the const identifier evokes the lofty ideals of the The Constitution while having the defensive posture of a computer programing language construct, and, to the best of my knowledge, there isn’t already an elementary school in some place like my home state of North Carolina that’s called, “Const Elementary School.” Therefore, Const Academy is a boarding school in Vermont, which has provided inspiration for religious ideas for at least 200 years.
94: Suzy teaches AP compsci and a mysterious hacking scandal unfolds, publicly
95: Suzy develops a semantic web for song lyrics that visualizes “the word-play-tree”
96: Suzy grandstands at Stanford and starts an unauthorized, online gossip service
97: Suzy starts a social network called sharkpages.com for financiers that goes viral
98: Suzy closes a series A, gets office space in Berkeley, and recruits a hedge fund
99: Suzy opens up sharkpages to college students and it encroaches on academic life
00: Suzy releases tech that can automatically take SAT tests, and she stages an IPO
01: hackers hack sharkpages and the stock plummets; Suzy starts laying people off
02: sharkpages is dead in the water and Suzy has to withstand unconditional defeat
03: Suzy dates a European tennis player who has a big heart and a gambling problem
04: Suzy parlays with the campaign manager of an innovative candidate for president
05: Suzy becomes the mascot of the highly controversial winner, named President Lawn
06: President Lawn is impeached. The cold-blooded VP takes office, and Suzy is out
07: Suzy writes TheSuzy.com Show, about the rise and fall of former President Lawn
08: Suzy crashes the markets and endures blame for causing the breakdown of trust
09: Suzy discovers a coup, planned by the VP’s trust, after he loses the ’08 election
10: Suzy brokers a historic peace treaty, and the press talks about a “Pax Susannah”
11: Suzy founds a system of belief that’s followed by celebrities and world leaders
12: Suzy helps President Castle win re-election by becoming his Chief Digital Officer
13: Suzy’s architecture for moderating structured data on the Internet gains traction
There were lots of mixed metaphors brewing in 1994 that might have primed the atmosphere for a massive, cold front for reality distortion. Deputy Treasury Secretary Lawrence Summers played a role in the $50 billion bailout of Mexico’s financial system that was supported by a discretionary pool of money, controlled by the executive branch, without oversight from congress. The Republican opposition to the Clinton Administration, led by Newt Gingrich, took the country by storm with a platform called the Contract for America, and Hillary’s effort to implement a single payer health care system in America was in full swing.
The story ends with Suzy filming a sunset, talking about how technology is both a blessing and a curse, to be questioned.
After saving the world in 2010 and upgrading the Internet in 2013, Suzy returned to her roots and hatched a plot to win the race to build strong AI. Since then, she’s been traveling the world, singing in lounges, telling people she’s eight years younger than she really is, using smoke and mirrors to change her age to be the same as mine in every official record where it’s represented, and doing artificial intelligence research.
There will mostly likely be an inflection point where reputable scientists and other experts reach a consensus that my facial injury does, in fact, cause essentially all people to subconsciously misread my emotions in a way that leads them to develop self-reinforcing false beliefs about me. Right now, in spring 2017, a lot of powerful people don’t agree with that statement, but they will mostly likely change their minds.
I’m confident we’ll get there within a few years or less. I still need to secure institutional underwriting and a fair amount of additional funding in order to follow through on the research, but I’ve been living with this for 15 years, so I have a lot of data that nobody else has. At this point, the research that needs to be done boils down to figuring out how to recreate the data that’s in my brain so that other people can see what I see, and when this happens, I plan on updating this epilogue and shipping it as part of SharkInjury (SI) version 2.0.
The bigger question, which may not get adequately addressed until SI 3.0, is whether or not my facial injury causes everyone to develop false beliefs about each other, and I don’t know the answer. This is a genuinely open research question. It could be that my facial injury only causes people to misunderstand my emotions, and that’s it. It’s also possible that my facial injury triggers and amplifies a cognitive virus that inspires people to start arms races where they compete to frame each other as being more anxious than they really are, such that it’s not clear how to end these arms races without a conflagration of violence.
If I had to guess, I’d say this cognitive virus is real and its function is very central to what’s been happening lately in American political discourse, but we’ll have to wait and see what the research says.
I started writing this text by creating a map of facts. Deciding which facts to share is half the battle. What I’ve learned from mulling over how to write this text for many years is that, first and foremost, free people enjoy sharing all the facts that can be defended by assembling a real world apparatus, made up of people and processes, that is sufficient for sorting out copy pasting line break issues that might get introduced into the text by accident, two, the important facts are defensible, and it often takes time to build the necessary apparatus, and three, I’m describing simulated war, but I hope this text will help keep the peace. With that in mind, I finished writing this text by reducing the facts into an investment opportunity and two learning opportunities:
Part 1 described the investment opportunity. For people interested in the technology industry, I made a bet that reading the FACTS in Part 1 would be worth their time. The metaphors we use to describe the crucial concepts in the field of computer science are deeply opaque. Facebook staged the hardest hitting satire of the technology industry’s addiction to self-reinforcing indirection I’ve ever seen, but a lot of water has passed under the bridge, and now it’s more difficult than ever to talk about this problem. The data I served up in Part 1 revealed a literary toolchain that might be vast enough to fix the problems with compsci.
Part 2 outlined an opportunity to learn, by writing fiction that tells a STORY. I released an open source framework for a story I plan to write with the help of a technique called crowdsourcing. I want to be open and transparent about my belief that this text performs dimensionality reduction, which is a fancy way of saying, “demolition,” on a comprehensive reality distortion field that exists. War used to be “the go to method” for demolishing distortion fields, but thanks to the geometrically increasing computing power of the commodity microprocessor, which has been in force for the last few decades, we can now carry out the proposed reduction in the dimensionality of culture by simply mapping and reducing our view of the data set.
In other words, my original career was in tech. Now I’m rolling up my sleeves and pursuing a career in politics. I’m narrating the story of a woman named Suzy who starts a highflying dot com startup that’s a lot like a social network, but in 1995. If you write about Suzy’s storied career, you’ll learn how to leverage data, itself, as a platform in real life, in print, and online.
Suzy, my private notes, Suzy’s story, and Shark Injury are not primarily about any one person, company, family, or clandestine backroom deal. This amulet was inspired by politics, human behavior, brain waves, and life.
Copyright © 2017 Todd Perry
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the “Software”), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE