Bayes’ Theorem? More like Based Theorem.
dkirmani
Karma: 317
On an hour-by-hour basis my time at UIUC was about as enjoyable as my time in jail, but that’s admittedly a question of four years vs. twenty hours. My best friend went to CMU, and he concurs.
Wow. That bad, huh?
Don’t hesitate to ask me anything.
I’ll probably update significantly on most things you have to say on mentoring and evaluation, simply because it directly impacts my future, and you seem both capable and credible. What I really want to know, though, is this:
In your value estimation, a UIUC CS dropout-turned-founder has about the same EV as a UIUC CS grad, but with significantly higher variance. To what extent is this view the consensus among those responsible for evaluating software talent? Would many of your colleagues instead see the dropout-turned-founder as lower-EV (or maybe even higher-EV), but don’t give much consideration to variance?
To someone who knows that universities operate upon selection effects, getting into a selective college is almost as impressive as graduating from it, but most people subscribe to the mythos that the degree is what matters (as illustrated by the sheepskin effect), which might mean that talent evaluators that read LW favor the dropout more than those who don’t.
As someone who evaluates and mentors young software engineers as part of his living and has for about thirty years now, “dropped out of UIUC to found an interesting startup” would strike me as being good a resume’ on average as “fresh CS grad from UIUC”. What I would be worried about is the variance.
This information is absolutely priceless to me. Thank you so much! One way or another, your comment will be instrumental to my career decision(s).
In my experience, UIUC is a uniquely horrible environment, so much so that flight from there should always be a consideration.
This rings true to me. I know a guy majoring in data science at Carnegie Mellon that says it’s pretty glum there as well.
More importantly, you’re using far-mode arguments (general, long-term, non-specific) for pro and near-mode (specific scenarios that you have to actually do) for contra, which is going to be a painful awakening when you have to actually act.
This is a very, very important point. Thank you. I intended to talk about concrete non-college plans under “opportunity costs”, but never got there.
No matter what, my cofounders and I will spend at least the summer developing Poetic. If Poetic takes off, wonderful, but the default outcome is that it will die.
If Poetic does die, I can use its corpse for cover during the job hunt. To quote lsusr, who’s quoting Paul Graham:
When I was 24 I had a hard time getting a job as a software developer. As an self-taught engineer, I had no credentials.
To make sure I asked some friends who work for big companies. I asked managers at Yahoo, Google, Amazon, Cisco and Microsoft how they’d feel about two candidates, both 24, with equal ability, one who’d tried to start a startup that tanked, and another who’d spent the two years since college working as a developer at a big company. Every one responded that they’d prefer the guy who’d tried to start his own company. So there you have it. Want to get hired by Yahoo? Start your own company.
“Hey,” I thought, “I’m 24. I can game the system! If I start a startup with the deliberate intention to fail after a few months then I can get hired as a software developer.”
-
I am 90% certain I can get hired at Introspective Systems, as my internship there was pretty transparently part of a talent recruitment pipeline, and the CTO told me college was a scam and tried to talk me out of going.
The uncertainty arises from the fact that that CTO has since passed away. However, the current CEO was her wife, and she was nice to me too.
I could resolve this uncertainty by calling them and asking for a job, but that would imply that I wouldn’t be working on Poetic full-time, and I don’t want to cut off my exposure to right-tail risk and/or throw my cofounders under the bus.
-
I also have a pipe-dream where I convince an EA institution to pay me money to feed mice lamivudine in my backyard, and record its effect upon the aging process. Worth pursuing, but not a secure career path by any means.
Yup. Rationality is orthogonal to values.
Nah. Epistemic rationality is instrumentally convergent.
Though zero-day exploits themselves are precious secrets, “how to find” zero-days is public knowledge.
I’ll bite. How does one reliably find zero-days in a given system?
From Neal Stephenson’s Diamond Age:
“So there’s no way to trace the path taken by a message.”
“Right. And the real situation is even more complicated. The media net was designed from the ground up to provide privacy and security, so that people could use it to transfer money. That’s one reason the nationstates collapsed–as soon as the media grid was up and running, financial transactions could no longer be monitored by governments, and the tax collection system got fubared.”
When I think about where rationalists came from, my answer is 1) EY writing the original sequences, and 2) EY writing HPMOR. It feels like those things happened, tons of people joined, and then they stopped happening, and people stopped joining.
Is this really the case? I’m eighteen, and I know a few people here in my age group.
I also didn’t take either of the membership routes you listed. When I was 16.75 (±0.33) years old, I happened upon a Slate Star Codex post (might’ve been this one). I thought “Wow, this blog is great!” so I then proceeded to read all of the SSC backcatalog[1]. Once I ran out of posts, I saw “LessWrong” under SSC’s “blogroll” header, and the rest is history. I didn’t systematically read the sequences[2], but instead just read whatever looked interesting on the frontpage. I had previously read Superintelligence, Thinking Fast and Slow, and Surely You’re Joking, Mr. Feynman, so I already had some kind of background exposure to the ideas discussed here.
I was trying to curb my Reddit addiction, so I used SSC, Hacker News (and later, LW) as substitutes. Still do. ↩︎
And I never have. Did read HPMoR, though. ↩︎
Maybe prefix notation feels weird because English (and Chinese, and Spanish, and Russian...) follow the subject-verb-object word order. “
list.append(3)”, for example, is in SVO order, while “(append list 3)” is in VSO order.Most languages primarily use either SOV or SVO word-orderings, with VSO trailing at a distant third. Funnily enough, both Classical Arabic and Biblical Hebrew are VSO languages. Looks like God does speak Lisp after all.
I don’t expect LW to outperform the market short-term. I would expect LWers to be better at anticipating plausible Black Swans such as the crypto boom and AI takeoff.
Personal anecdote: Before I discovered LW/SSC, when I was 14, I convinced my parents to invest a substantial sum into Bitcoin back when it was trading at $5.2k. Despite my exhortations, they haven’t sold any of it yet.
This is golden, thank you! I wonder if healthy adults should take ARVs...
Maybe you’re conflating an increase in DNA damage with an increase in the rate of DNA damage. The data is consistent with:
One-time exposure to lots of DNA damage (nuclear bomb) → Increased cancer risk
Increase in background rate of DNA damage (aging, working near ionizing radiation) → Increased dementia risk
I grew up in Pakistan. We had more days with blackouts than without, and there were often multiple per day. Everything in this post checks out (except we had a diesel generator, not gasoline, which is easier to improvise fuel for). Incidentally, Pakistan has been in a Cold War-ish situation with India for over twenty years, so we had lots of flour in the basement, and we had on-site water tanks (as opposed to tap water).
I’m curious what people here think the second most important problem is.
Any recommendations for works of fiction where the protagonist is sufficiently strategic and agenty? Limitless, Death Note, and HPMoR come to mind.
I think you dismiss fear #1, “Addiction”, too quickly. When I was fifteen, I realized that games were hijacking the productive impulse to better my life. Some part of my brain had conflated being good at CS:GO with status/achievement in the real world, and as a result, I sank hundreds of hours of otherwise useful time into the game. I made a concerted effort to quit, and I don’t regret it in the slightest.
Better yet, if you could measure both mental strength and tendency to obedience in a fast, Goodhart-resistant way, you could create a credential that could compete with, and undercut, college.
What would “incoherent decisions” look like for an agent that has a utility function defined on action-observation histories?
More good stuff, thank you again.
Ahh, so that’s why companies want you to be “aligned with their mission” so badly.
I like making positive impact, and I like making money. Both would be optimal, but either one is leagues better than neither, so under my value system, I would count your move as a very positive one.
I’ve heard this sentiment on Hacker News often, and always chalked it up to sour grapes. Glad to hear that there’s some truth to it. Also, I’d be a one-year dropout, not a two-year one, if that changes your evaluation significantly. However, I’m technically a junior, since I took a bunch of classes at UMaine during high school (including DS&A, Linear Algebra, and “ML”).
That’s perfectly fine by me. If you see me as a potential source of alpha, I’d be willing to help you test that hypothesis :)