Make Me Do Things, I Done Learnin
“I’ll never get stuck in a musical time-period like my parents; I’ll always stay hip to the newest cool music”
Maybe you didn’t say this about MUSIC when you were frustrated with your elders’ dated sense of style, but you probably said it about something. Maybe it was watching your old uncle brag about how he still has a flip-phone? Perhaps a crusty teacher rebelled against the “New Math”. Did you have anyone in your neighborhood who refused to use power-tools, or refused to use portable power tools, or still went to the bank teller despite ATMs being around since the 1980s? Chances are, you’ve witnessed conservatism and refusal to change in some form. I’m here to tell you, however, it doesn’t happen like you think it does. What happens is people clutch to familiarity and avoid change because it has an upfront cost which is not tasty. The meek don’t inherit the earth, the young do.
At age 48 I said to my wife…
“Honey, maybe our #1 (our oldest son, 9) should just skip the half-million-dollar colleges and go to Lambda School and learn how to code, learn more efficiently, and get to doing meaningful work much quicker than his peers?”
As a keen study of what new concepts would change the world and become great long-term investments, I couldn’t afford to continue looking at things differently, or lose touch with the newest things. While 4-man bands of my youth dropped to just an “emcee” plus a “DJ” with Hip-Hop, and then furthermore to just the DJ of EDM, and I lost touch, I couldn’t afford to lose touch with world-beating products. I had been searching for how education would be fixed in the future and who would make the money from fixing it, and had come across Lambda School in my twitter feed. Youtube and Kahn Academy were shots across the bow, but what about formal education? What was truly going to knock un-economic college tuitions off their perch and how?
Her reply: “Maybe that’s what YOU should do, and set an example?”
This is why I married her; my wife is a DO-er. She learns by doing, and I hoped it would rub-off someday on my overly-academic style of learning.
Immediately and excitedly I applied for Lambda School to learn how to code — code like a teenager looking to figure out how to make mods on Minecraft or flippant Lenses on Snapchat Lens Studio. It’s actually something I always vowed to do at some point. I had even had a business idea in my head which could revolutionize finance but knew from the experience of watching the best CEOs and tech leaders that I needed to lead by example or I’d fail. I needed to DO to be respected enough to lead others into a startup-battle. Like in music, I also knew I needed lot of practise.
Consider my career arc to understand why I stopped everything to go full time to Lambda School:
About 1978 I was introduced to pong in Minnesota during a long road trip. My cousins seemed rich because their Dad bought them the game which turned a TV set into user-controlled screen. In the 1970s this was unfathomable. TV content was set by giant corporations like ABC, NBC, CBS and PBS (the big 3 plus that grainy little guy who was always begging for money twice a year in between Doctor Who rerun marathons). A home TV watcher didn’t simply interject his OWN content onto the sacred TV screen. Or did he?
After jaw dropping at the incredibly boring pong, I was then introduced to computer programming in grade school at a very young age for that time — a Tandy all-in-one with a magnetic tape drive if I remember correctly. The screen came in one color, green, the number “8” maybe had 10 pixels resolution, and the only language that worked was BASIC. “Goto 10” is my fondest memory of BASIC. Yeah, you had LINE NUMBERS in BASIC, so when you looped you used a “GOTO” statement and told the program what line to go next. I didn’t know it then, but that was actually a lot “closer to the metal” than modern languages which have gotten rid of all that fuss. I also learned that BASIC was already a far cry from what preceded it, which was two-stack languages like COBOL and FORTH, or early object-oriented langauges like smalltalk which began the next-level of langauges in the 1980s.
Also before the 1980s, towards the end, I encountered the Apple II machine — at the time the highest selling personal computer — a hit that launched Apple into mainstream consciousness. Sure, the game Breakout to me was addictive and garnered all the attention, but in my Mom’s community college Math Lab which she ran, the Apple II in a closet (yep, computers were put in the closet as strange toys for academia back then) only had less than a dozen programs on it and I ran them all. Most were beyond awful, even then, and typically text-based which is why graphic Breakout was so cool. It was there I met the work of my favorite coder of all time: John Walker. I wouldn’t learn his name for about three decades, until the internet became SO filled with information that John Walker claimed “ANIMAL” as a viral game that he created. John Walker read an article around the early zeroes talking about how Animal seemed to be a viral game on some of the earliest computers, but the writer of the article didn’t know who created the game. Why? The game was likely the first computer virus — it copied itself in memory and threw itself onto floppy drives to infect the next computer it met. Back then, the most common way to get cool “programs” was to basically steal them. If you read “Hard Drive” about the early days of Microsoft you’ll understand how much Gates & Allen dealt with piracy back then. Animal was glorified 21 questions, but with a twist I’d never forget.
Animal when run from the command line (which WAS the computer, this was loooong before Steve Jobs decided to make “windows” mainstream) would simply ask the user to think of an animal which the program would then try to guess. It was a clever little program, and guessed rather well for a 1970s algorithm running on 64 KB of RAM (little more than enough to store all half-dozen of the Medium articles I’ve written). It had one quality that changed a part of my life, the end. When Animal failed, it failed well.
“Is it a Zebra?”
No, no it wasn’t Animal the Game. I win. But Animal wasn’t beat, not in the long term. You see, Animal instead of giving up asked you the user to help improve the algorithm. Whaaat? By the time I ran into Animal, I’d already run some pretty basic programs which allowed the program to ask the user questions, but this was different. Even at age 8 I was floored by this. This Animal program was going to teach itself. Prior to guessing “Zebra” Animal had asked me if my animal had stripes. When I said yes, Animal jumped to the ONLY conclusion its algorithm had for that answer: Zebra. So Animal asked me to write a yes/no question to clarify the difference between my animal, Tiger, and their guess Zebra. So after typing my animal was Tiger, I wrote “Is your animal in the cat family?”. Animal was so smart, it could figure out that if your animal was bigger than an armadillo and had stripes, it was probably a Tiger. Or, at least NOW it did, now that I “taught” it.
Thirty years later when I searched the internet hoping to find out who the brilliant man was behind Animal, I read John Walker’s reply to an article about his game. However, the article wasn’t focused on MY favorite aspect of the game, Articificial Intelligence; it was focused on Animal as the first computer virus. Animal wasn’t malware tho, it was really the first “viral” game. At the same time I found out John Walker didn’t just make Animal but also started a tiny little program called AutoCad that just so happened to make him rich beyond anyone’s wildest imagination. Like Bruce Brown the one-and-done independent movie king and lifelong surf bum of “Endless Summer” fame, John Walker after selling AutoCad retired into what I’d label a software-bum-genius. To me, his legend will live forever, as his program taught me everything I wanted to know about what “A.I.” truly is.
Aside: Want to learn a LOT of diverse interesting topics in one place, go to… https://www.fourmilab.ch/fourmilog/ which is John Walker’s excellent blog he’s blessed us all with writing since 2014 and earlier. You can even read his Animal story there and if you’re interested in viruses or artificial intelligence I insist you read that story.
It’s indeed possible computers can learn a little bit on their own, but can they? While a computer can learn what a dog is using many pictures, doesn’t it need human input to describe dog-like features? To distinguish between dogs and foxes or woodchucks? Computer algorithms are just like people, they learn from taking consensus and matching it to new data. When you learned what a dog was, or what “no” meant you as a small child with very little memory simply repeated the data many times over. How many dogs you had to see to know what made up a dog is likely variable. One of my kids called every woman she saw a dog for a few months, before realizing she was on the wrong track — tripped up likely by some old crusty uncle at a picnic or an errant YouTube video made by disagreeable teenagers. The point is that artificial intelligence is pretty obtuse most of the time, until it learns beyond a shadow of a doubt what’s what. Just like Animal, almost all, if not ALL, artificial intelligence software must start from a basis of humans telling it what is right and what is wrong. Humans must go thru the process of growing and strengthening neurons and brain cells, whereas computers have a small advantage in the fact theirs are created FOR them in a pre-quantized state. All computers have to do is get programmed. What some people call “AI” is computers learning without human interaction. This just models a child learning that heights can be dangerous, or steam can be painfully hot — without the aid of a parent.
Whether programmed directly by a human or learned via Mother Nature-as-teacher, computers learn by INPUT. This is what I learned from Animal, and I began applying it in everything I did with respect to computing.
In sophomore year of high school my parents recommended I take Latin to do better on standardized tests for college acceptance. I survived 2 days. Latin was my first “elective” ever, until 10th grade I had never ONCE been asked what I WANTED to learn. When FINALLY presented a choice, between computer programming and Latin I had let expedience prevail rather than my heart. In my heart, I thought computers were pretty cool, and wanted to learn how to make them do stuff like Animal and Breakout. So after a disheartening 2 days I switched to computer programming and began a 3 year trek which ended with the Advanced Placement Computer AB exam. (I got a 4 and my college wouldn’t accept that). MUCH more important than getting college credits or latin derivations (altho the latter comes in to play now — at the end of this article) my senior year project for Mr. Russell was super important to me. 4 students and I (John Wall, Dave Wallace, myself, and someone else I can’t remember) were tasked with a senior year compu-sci project of our own picking. As has become common in my modern coding school, picking a project can seem daunting, but “I got a million of’em” in my head at all times — always have.
Whaaat? This was the response from my classmates. I explained. Our spell checker, like Animal started, would know exactly NOTHING when it was first run. But by verifying each word it saw with the USER (teaching it when a word was right) it would learn all the words. The goal was to feed the spell-checker a book, let it READ every word in the book and only stop when it didn’t have the word in it’s dictionary — a dictionary which started at zero. I hate to admit, but Wallace and Walls did most of the work for this program, as I quickly learned that working in a group can be difficult if coders have different styles. The functions I wrote back then had to be rewritten by them because they were writing the function calls and the parameters were incongruent. We were simply too young to know how good collaboration was done; I figured this out MUCH later. But the program got build to MVP (minimum viable product) and it was run on our high schools mini-computer which occupied a full room of its own in 1988. Our program broke the schools computer. Not because of an infinite-do-loop, but because after reading ONE paragraph of words our program had eaten up ALL the memory of the school’s computer with our data structure. Mr Russell gave us each A+ and told us NEVER to run our program on his computer again (he struggled for a half a day to restore the computer to working order after “grading” our project).
Alas, no one told me riches and fame could be had by learning how to code computers in 1989, in my small circles I didn’t even realize coding was a paid profession. In college I took math and physics instead, and only took one (again PASCAL, uggh) more computer course in my early student career. So I took my Navy ROTC scholarship and ran nuclear reactors and shot torpedoes for the US Navy on USS Houston for 5 years to pay for college.
While running 1960s-technology “green-screen” fire-control systems to calculate torpedo trajectories for enemy submarines kept me using computers, my knowledge in programming atrophied quickly. The world began using things like frameworks, classes, objects, and git. After getting immediately out of the Navy and switching careers to technology investing (20 years) for a hedge fund, I realized just how much software had changed by the early zeroes when dotcom millionaires became a dime-a-dozen. html looked like gibberish. I had to figure out what and IDE was, or a fibre-channel or application-server. These weren’t big topics in the 1980s; modern computer scientists didn’t even know what PASCAL was anymore, it had died between the efficient but overtly-simple langauges like COBOL and FORTRAN, and C and it’s many object-oriented derivatives like C++, C# (MSFT), and Objective-C (Apple). When I stopped my coding, implementing FUNCTIONS were kind’ve upper-level stuff. When analyzing tech companies during the 2000–2019 timeframe, a whole new language had been formed which was foriegn. Coders had learned how to re-use code, implementing standard libraries and frameworks instead of writing each and every for-loop or data-structure themselves.
The reason most people stop listening to new music and get stuck in an era like “Oldies (50s)” or “Classic Rock” (1967–1979) or “80s Music” & “Grunge” is because of kids. When you have kids, there’s simply no time for keeping up with music anymore. Your KIDS stay current, but you start listening to talk radio if you can even HEAR the radio. Plus, the music of your youth becomes comfort music, a nostalgic way to remember all the FREEDOM you had when you were single. Sure, you can maybe hear about the really BIG hits of the times (Pumped Up Kicks, Stacy’s Mom, Gagnam Style) at the occasional wedding, but there’s no longer any time to find those cool deep cuts you prided yourself on uncovering in your 20s (18 Wheels of Love, Transdermal Celebration, Football Tonite, Three Days). BUT THIS IS JUST SAD EXCUSES FROM AN OLD MAN in a single topic he doesn’t care much about anymore.
DON’T STOP LEARNIN… EVER.
It’s 2020 and my FORMAL training is complete. Learned how to make working apps in Lambda School’s iOS school, then spent time in “Labs” DO-ing. Learned how to create the equivalent of Instagram’s mobile app inside of a couple months — or at least the hard parts — by myself but working with a team of back-end coders. After that I spent a couple months learning data structures, algorithms, and even how to write a primative but working blockchain complete with a proof of work. The instruction was fantastic the whole way, but then came the challenge — literally.
The Cambridge University Metanet Society Hackathon.
Ideas always come easy to me for some reason, so I didn’t just enter this 6-week hackathon — I entered it TWICE. One big lesson from Lambda School’s “Labs” program was that you can’t do it all yourself. So I took on 50/50 partners in each project and we divided up the labor according to what each of us did best. We worked hard and long hours, morning til night. There was always more to do, and a deadline to meet. This hackathon was “virtual” so the rules were a little different. While it was 6 weeks, projects of older age could enter as well.
Both projects had one thing in common: “Self Learning” John Walker-style. The projects are intended to be seeds, grown with patience, naturally, one curious, far-looking person at a time.
is one of my projects, coded by 25-year coding-vet & coFounder John DiFelice of Philadelphia, it’s an online dictionary that self-learns definitions and gets more efficient for users by using BitCoin SV — digital money which allows our algorithm to RANK definitions as they come in from “WordSmiths”. Anyone can write a definition, but it takes passion, and any definitiion entered must compete with any others which are written. Eventually the algorithms will prune-off bad definitions whether intentionally mischievous or just lower quality. The best rise. Just like Animal and a high-school spell checker, SLictionary is like a baby dictionary with no words in it, but it learns with use. It’s a labor of love, but one like a baby which can rise up and do great things. SLictionary was programmed in just the 6 weeks. Here’s a link, but you must use an email to sign up for a wallet called Money Button to search for words or even define words.
We make you pay to define a word, so we don’t get spammed or BOT-attacked. But also for a game-theory reason. If we make you pay a little money to definie a word, you’ll take it more seriously and do a better job. WordSmiths don’t just spit out definitions, they INVEST in them, with the hope of making a return over time as follow-on users can then vote on the definitions which helped them best. We have already purchased https://www.SLictionary.com and intend on making it a long-running commercial operation and continuing to improve it with our spare time. It has all the basics now, and you can tune in routinely to see improvements. The UI is a bit hacky, but that’s bc of the time constraints of the hackathon — we had to focus on back-end and getting the payments right before worrying too much how we look. UI will come soon, and we’re tapping Lambda School for students with fresh design ideas.
is “Your Comedy Forever” — a comedy app which specializes currently (v. 1.0) in jokes, coded up by fellow Lambda-Schooler Joshua Kaunert of Chicago & myself. Unlike all joke apps on Apple’s app store, Hilarist (currently on Test Flight) has a few groundbreaking features:
It’s a BitCoin wallet, which allows comedians to make money from their creativity. Anyone can submit jokes or comedic material, and fans can then save those pieces for what amounts to a transaction fee in BitCoin SV. Hilarist acts as middleman — a marketplace — but comedians receive the majority of economic attribution.
We offer “EtchRight” for comedians to effectively copyright their comedic routines — never having to fight with other comedians over who wrote what and WHEN. EtchRight we hope will solve those disputes.
For old jokes we’ve created a game-theory add-on called “JokEtymology”. This is where we borrow from BitCoin’s “Proof of Work” to put comedy fans to work attributing old jokes. Whoever finds the earliest documented use of an old joke, is rewarded and can earn tips from posting that joke and it’s oldest history. Oldest history wins. So not just comedians have monetary incentive to fill out Hilarist’s database, but fans do too. This part is practically more fun than the jokes themselves. In the end we hope Hilarist becomes the largest joke database in the world, and we’ll then expand to all other comedic formats.
Hilarist is undergoing a revamp, but you can demo it here if you have Test Flight app downloaded on your iPhone or iPad and you use that same device to click this link:
… and then open Test Flight and it should say “update” or “download”.
I have gained a new friend with one project, and strengthened another friendship with the other. Sometimes we fight and step on each others egoes, but all for the greater good of the project — the babies we hope to send off into the world and let grow into dragon-slayers.
It just so happens the two projects were judged last Sunday, and we were “in the money” on both. Despite ample competition from seasoned projects which began more than a year ago, both Hilarist and SLictionary tied for 4th place and received prize money for our efforts — prize money we will invest back into the business. We paid a designer a (VERY) small fee already, and used the proceeds to buy our website and pay for registering the business.
This recognition via a Hackathon competition might seem like a victory, but instead it just feels like the beginning of a long journey. Coding isn’t something I realistically thought would ever result in a product. Yet a year of dedicated full-time learning, and voila! Had you asked me a year ago if I could have learned to code apps for an iPhone and even place in a hackathon, i’d have said you were nuts. Certainly not in one year’s time.
I have to thank all the support I received along the way:
My Mom, my wife, my kids all of whom rarely see me anymore and have bent-over backwards to help me. Lambda School certainly, and I have a long list of names inside those virtual walls. But also John Walker, a man I’ll likely never meet, and Steve Jobs (iPhone) and Wosniak (Breakout, Apple II). I’ve since learned from some more recent coders such as Craig Wright (C++, Bitcoin a Peer to Peer Cash System) and several others within the BSV community — a small group dear to my heart at this point as I “moonlit” from my schooling as a BitCoin zealot for the last couple years. Let’s not forget Messrs Wall & Wallace either — who helped me get that A+ in CompuSci while I was probably spending more time cramming for exams.
What have I learnt from all this? You’re never done learnin. The TRUE beauty of diversifying seemingly unrelated fields such as math, submarines, nuclear reactors, torpedoes, free cash flow calculations, net present valuations, business organization structural analysis, “poker-reading” management teams’ body language, technology predictions, writing Swift for iOS devices, and 1970s-era Artificial Intelligence is when those things come together to push you past what single-field experts know. All the very BEST innovations come from those who not only don’t stop learnin, but learn from many different angles which may or may not be related.
Now excuse me while I talk to my wife about how to get my kids interested in playing a musical instrument, because performing live in a band is always something I’ve dreamed I’d DO.
BAEmail.me at firstname.lastname@example.org
ANIMAL: Read the story here https://www.fourmilab.ch/documents/univac/animal.html
Lambda School: https://lambdaschool.com/
SLictionary: A Self-Learning Dictionary using BitCoin SV. https://slictionary-fc2a0.web.app/
Cambridge University MetaNet Society Hackathon: https://phoenix-challenge.com/explore
Title Inspiration: https://www.youtube.com/watch?v=L1BDM1oBRJ8
Three Days — Jane’s Addiction: https://www.youtube.com/watch?v=Ds_mAJ5OKzE
Football Tonite — The Rugburns (w Steve Poltz-ie) https://www.youtube.com/watch?v=U30EFFe9FDo
18 Wheels of Love — Drive-By Truckers: https://www.youtube.com/watch?v=8AW14edk9w4
Transdermal Celebration fan-made video— Ween, New Hope PA