Bits and Paradoxes - #24
Mimetic Traps
Disclaimer: If you find this iteration long, do save it for reading later. Also, I am putting “#” to the iteration number now. Seems like it’s an “unspoken” rule for newsletter. :/
Hey folks,
How often do you get bombarded by "what the fuck am I doing" feeling? Do you like (love?) whatever you are doing?
How often do you find yourself doing things for the sake of doing, all the meaningless rut that you have no control over?
[[Mimetic Traps]]
If we think really hard about "Is this really what I want to do?”, it's obvious that the answers aren't that hopeful. What we do is probably imposed by a lot of conventional flow of life and society.
We pass the high school phase and yet we don't seem to have any wishes to follow. And end up “following the crowd”. Coming from an engineering background, it might sound egoistic, but I never had any idea of joining whatever the freaking things were out there. I used to love mathematics (overestimation) but my parents wanted me to go into biology. Probably "the sciency kid" narrative was into play.
During the first month of my grade 12, I had taken both biology and mathematics with the hope that I could nail both the domains. However, biology didn't seem to interest me much because of all the rote-learning thing. [1] So, without informing anyone I quickly changed my department to physics. Obviously, I used to love physics (still do, but not that knowledgeable). Up to that point in my life, I used to love Quantum Mechanics and aerodynamics (yet another overestimation). Yet, I didn't have any tangible plan to continue whatever I was to study. There was always this “what's next” element that didn't have any answers. I didn't even have thoughts about engineering.
But then, engineering happened.
I have no idea why I chose computer engineering. Probably, it's because of my “likeness” to solve theoretical problems. During my childhood, when I got my hands on QBASIC around grade 6-7, I used to do simple programming. Things like making quizzes — as hilarious as it sounds — "Ko bancha roadpati”, some weird window animations, and mathematical sequences like Fibonacci might have fascinated me. Also, a whole lot of other things like cracking up software and tweaking (twerking? :D) the OS might have planted a seed I guess.
I didn't have any definite plans during the engineering days. Initial years were more about game programming and API development. And I think that was also a trap because everyone was doing it. So, why wouldn't I? Another default mode. Plus a sense of competition for scholarships and "passing" the semester were also into play. There weren't any absolute goals. [2]
Even entering the machine learning domain was partly by chance. I got intrigued by clustering algorithms from the Data Mining curriculum which might have catalyzed my interests, which further paved a way for projects in a similar domain. The minor year project was related to Nepali-English text translation and the major year project was a “query by humming” system that could help search any songs by just humming.
After graduation, I wasn't sure what the fuck I would do. I was naive. I lacked knowledge.
I entered my first corporate job hoping to get into AI and ML. But for a year, I couldn't. I just couldn't because of well…competition… There were a lot of people who were much much better in terms of math/coding/research. I had discarded the idea of AI and just focused on that competition, did a lot of side projects, tinkered and showed-off those projects so that people could really see me as the “guy who could do it”. Trust me, those weren't related to the absolute goals of getting in AI. It was for the sake of competition. For a year or so, I was in the meaningless work-rut that I didn't love. Finally, I quit hoping to start my own startup. And that was it.
The startup didn't do well for me either. It was also one of those things revolving around "everyone else is doing, so why shouldn't I?". AI was a hype train. One year into my own startup, and I realized it wasn't working for me no matter what I tried to do. I quit again. It has more to do with a lot of context-switching between different tasks than product vision.
Eventually, I started to focus on pursuing grad schools. (But getting nowhere :/ )
Finally, here I am talking about all these competitions, hype-trains, and traps.
You might be wondering why I am telling this to you? One lesson from this introspection is [[Mimetic Traps]]. Most of our lives, we do things because of competition, not for the sake of achieving absolute goals. We imitate (see: Mimetic Desire) what other people are doing, what the hype is providing. I was illusioned that whatever I was doing mattered to me.
“Absolute goal” here was to channel my interests into machine learning research. However due to the very mimetic trap of competition and proving my self-worth, I became competitive and strayed away from what I wanted to do. That competition could be seen from my involvement in our AI community where we taught students for 14 weeks, helping them to get into the AI hype train. It was very naive (without knowing much) and, I guess one of the mimetic (imitative) traps. Because there were a lot of communities in different parts of the world that were doing it, so why shouldn't we?
This mimetic trap is the compensation we provide to our real goals (vaguely).
Over the years, I have realized my interests in ML stems from two things:
Fascination with mathematics (although I don’t know much).
Solving problems
To its deepest roots, it's simply about solving problems I wanted to solve. This can be seen in many of my naive, unwanted side projects; most of them are simple utility scripts that solve my personal problems. So, it seems I treat “domains” as “tools”. The same goes for AI/ML/DL/DS thing. These are tools.
Algorithms are tools to solve problems. If a problem fascinates me, I spend an absurd amount of time solving it irresepective of its level of impacts. If not, you are sure to find me trapped in boredom.
Oh well! There are other parts of the story.
Why am I doing many side projects? I have no idea. There's this idea me and my friend is working on. It’s about a private search engine; indexing contents we read daily. But that might also be one of the mimetic traps. Personal Knowledge Management (PKM) tools are picking up the hype market more than ever. As a result, PKM-rut is haunting me. Personally, that side project will solve our own private problems of semantically searching the information we consume. But from the “product level” perspective, I guess there are already tools out there.
Now if I talk about this newsletter, it can also be a mimetic trap.
I am imitating all the newsletters I get in my own inbox because I find them fascinating. This newsletter won't even earn me anything and yet I do it for the sake of doing. An absolute goal was to share whatever I consume on weekly basis, to connect many dots. But then, consuming a lot of weekly newsletters, I feel it's also a competition; competition for both the attention and knowledge. I want this newsletter to be "valued” which can further create “value” for whoever is consuming it.
What if my user-base doesn't grow much? On one side it's the very mimetic trap that can force me to put out low-quality letters every week, overshadowing the original absolute goal. On the other side, it's a bet that something will eventually sprout from whatever the fuck I am writing in these fragments.
As I look back at all the previous letters, I feel I haven't even applied 10% of the shit I talk about. And probably I should and will. I must. But one thing is sure: I know about this mimetic trap and that might help me to introspect more and more.
Of course, the world is competitive. But healthy competitions build quality. That's the only coping mechanism to mimetic traps. Instead of mindlessly and meaninglessly imitating the norms, we can at least try to build quality. And just enjoy the process.
[Figure: A stick figure trapped in its own thought bubble of competition]
Mimetic Traps
Reading | Brian Timar | 7 min
This is the essay that got me thinking about the whole mimetic trap thing. It's probably one of those "why didn't I discover it earlier" reads. Everyone should read it. Seriously. Stop. Click the link and read the whole thing!
Brian talks about how competition trapped him to indulge in a meaningless rut of studying physics for the sake of excelling academically and how he couldn't enjoy the overall process.
When multiple people are striving towards a shared goal, they often rank themselves by progress within their peer group.
I swapped an absolute goal (figuring out how bits of nature work) with a relative one (scoring higher on tests than my classmates).
I was wrong in thinking that my work mattered to me, and I avoided asking myself this question because I knew the answer would be painful.
Whenever we are involved in works, projects, or whatever we are doing, we don't stop enough to ask ourselves the question: “why does this matter to me”. Probably, we know the answer is sour, less hopeful. So, we try to avoid such monologues whenever possible.
In some sense, when we are trap in that “life rut”, we don't see any negative aspect of it. For example, when someone is working for years doing mundane and boring tasks, they don't have the guts to quit because they wouldn't know any other way to earn money. It hurts to leave. And the only possible trade-off is to be in the competition to ascend the meaningless hierarchy of powers and internal politics of the organization. Why should they quit when they are earning enough, despite not enjoying the work? A big big big dilemma.
Mimetic Traps in a nutshell: it hurts to leave, and there’s nowhere to go. It decouples the social reward signal from the rest of objective reality — you can spend years ascending ranks in a hierarchy without producing anything that the rest of humanity finds valuable.
Mimetic environments are a serious problem only if you fall into one where you can’t enjoy the process. They’re a tool for amplifying ambition and diligence, and it’s up to you to apply this tool to yourself wisely.
Some of the important fragments to contemplate about this trap might be to:
Be ambitious to your own standards in morals and remind yourself about [[Absolute Goals]].
Maintain a diversity of pursuits.
Enjoy the process of whatever you're doing.
Make sure your job has clear price signals for success and failures.
I re-read this essay every morning for a few days and can't think enough of the emotions I have launched myself into. This is so relatable than anything else I have read about life and choices.
NO! It's not about regret. It has a proclivity towards the “Why does this matter” narrative! Sometimes I wonder if doing anything like side-projects, or maintaining a weekly newsletter...or looking out for grad schools is what I really want to do. This rabbit-hole of "overthinking" might go on indefinitely...for now I can only think hard...
What Does Any of This Have To Do with Physics?
Reading | Bob Henderson | 37 min
Bob Henderson studied physics, worked on Wall Street and is now an independent writer focused on science and finance. So, in this autobiographical essay, he contemplates his journey of pursuing research on Quantum Mechanics and the struggles he had in academia. He starts with a fresh, optimistic mindset with the "You can do whatever you want” narrative and eventually hits his mental limit leading him to quit the research.
Reading this is an emotional roller-coaster that, in its truest form, helps us to understand how stressful it is to do Ph.D. or similar research works that require unwavering attention. I feel this is also a [[Mimetic Trap]] where young graduate/students go into research works with hope and optimism to do something impactful, and later mutates into knowledge struggles. This struggle is also about keeping up with the competition, and in that competitive process, they indulge in unnecessarily nuanced works that don't seem to provide any “meaning” to them. Probably, that's also the reason why these types of postdoc research activities give rise to mental problems. That's scary. If you don't put up your thesis out there and don't bring out the results, there's a chance that you will fail. [0]
The gap between the other students and me didn’t seem to shrink much either.
I particularly saw the author's mental breakdown when his frequent "motto" of "you can do whatever you want" regresses to increasing self-doubts. He also talks about the tension between him and his supervisor.
The longer and harder I worked, the more I realized I didn’t know.
Dashing my dreamy ideas of what doing physics would be like. Fuck him, I thought to myself, more than once. Maybe I’d be better off without him. “You can do whatever you want” just didn’t want to die.
But there was also relief once I’d decided to do it, a lightness at the loss of a weight, as I finally and forever let go of “You can do whatever you want,” and accepted the idea of following someone else’s lead.
I think the very starting question and the ending question he pondered helps us think about the type of work we like to do and we are forcibly doing.
Have you ever been happy?
Was my failure to follow through on physics more about emotions than math?
The former one might relate to a certain sense of meaning in our life. It helps us introspect -- as cliched as it sounds -- our life, work, and happiness.
The latter tells us about the mimetic trap and academic struggles. Are you doing “this work” because every other person is doing? Or do you really want to work on “this”?
Related: Grad School and Public Health HN thread. There might be two different opinions. But it’s interesting to read the discussion.
Competitiveness and Mimetic Traps
Listening | Ali Abdaal | Taimur Abdaal | 59 min
Here Ali and his brother Taimur conversate about Brian’s essay on the mimetic trap. They contrast absolute goals with relative ones. The former might possibly be about passion and love for work. The latter might be about competitive behaviour.
There might not be a good or bad contrast between them. Perhaps, gamifying them might be better? Of course, there isn't any optimal final state. But it might be sub-optimal. Gamifying is entirely contextual. There are no solutions, just tradeoffs.
Say, if you are into grad school and want to get a RA/TA position, you should know that there's a tight competition. In this case, having competitive behaviour might be better since your financial incentive depends on that. You know competition sucks, but you have to. There's no other way. That, in the long run, might be able to help you move towards your absolute goal - - pursuing whatever you intend to.
The main point is the same — enjoy the process. If you hate whatever you’re doing, quitting is a possible solution.
#Listening
Good For You Podcast with Whitney Cummings | Lex Fridman - #50
I found this conversation really entertaining, insightful, and probably more thought-provoking than I had expected. I especially loved Lex's awkward side -- occasionally humorous, trying to connect to Whitney. The occasional healthy flirts in the conversation were also fun to watch (listen?). :D
While the conversation revolves around love and relationship most of the time, it also becomes philosophical and profoundly deep. I'd say Whitney and Lex had “Deep Meaningful Conversations” while trying to cope with the awkwardness of the dating-like elements. Haha.
Probably I felt connected because of Lex's natural behaviour; nerdy and serious about life with occasional humor. Probably, the topic of solitude, love, and relationship is something I can resonate with; from a perspective of a guy alone in his mind-cave, doing things for his own sake, not many social connections, and sucking at dating girls. Haha.
Few things that really struck me are:
“I am looking for a home now…something to return to.”
Dreams take sacrifices.
Relationships are about both doing what they are passionate about and cheering for each other's success.
A relationship gets deeper and deeper when you have a chance for a few moments which become magical. :)
Side Note: This is the rarest of the conversations to which I laughed so hard frequently. Probably because of Whitney’s comedic timings. Anyway, it’s mostly serious and deep. Highly recommended!
#Watching
Animation of how bridges were built in Central Europe in the middle-ages
It’s an amazing animation for remarkable engineering.
The Man Who Floated Away Into Space
This gave me goosebumps. One of the best stories from the Pursuit of Wonder.
Extraordinary Octopus Takes To Land | The Hunt | BBC Earth
David Attenborough’s voice can invoke any emotions. This one is a bit creepy while the octopus crawls on land, especially with the sound effect. Haha.
Anyway, octopuses are awesome. One of my favorite animals. **coughs** **Cthulhu**.
Do check out a very intimidating movie: My Octopus Teacher (trailer).
#Fascinating
The Super Duper NLP Repository
From Sebastian Ruder’s newsletter:
The Super Duper NLP Repo by Quantum stat has grandiose ambitions. It is the largest collection of NLP demos in colab notebooks (262 as of today) that I’m aware of—covering everything from table parsing to text-to-speech. A great way to get started on a new application! If you want more NLP goodness, then check out their NLP Model Forge and The Big Bad NLP Database for datasets.
This might be useful for tinkering.
#Fragments
Music: Choo Lo by The Local Train
On repeat for the whole week.
Music: Mo ko Lai Diu - Gauley Bhai - Music Mojo Season 6
This live session from Gauley Bhai is so beautiful. I had discovered them ~8 months back. And occasionally listen to them these days.
Ending Thoughts
This tweet by Patrick Collison sums it all about my obsession with Personal Knowledge Management and the current side project we are doing.
Signing off for now. Do share the letter if you find anything useful.
Love,
Nish
Footnotes
[1] - Biology fascinates me like any other thing in the universe. It’s just that I don’t like rote-learning. Probably, a lot of similar domain is like that because they are taught in a very non-interesting, boring ways. And probably the curriculum at the stage of grade 12 was also mundane; trying to fit in everything in a single book. Also, it might be possible that biology didn’t provide any way of side-research/exploration for me.
[0] - The “feeling of failure” might also be a side-effect of the competition in the domain.
[2] - I still don’t have any definite goals. Only a vague sense of direction now. Life’s fuzzy, but it’s okay this time. All these years have at least shaped meta-learning skills. No experience is wasted.