HNHacker News
TopNewBestAskShowJobs

bhntr3

897 karma · joined May 28, 2012

submissionscomments
bhntr3··on 2+1=4, by quinoa
If you're curious enough to read a 900 page book containing very accurate information about the specific processes that occur when preparing most foods then On Food and Cooking by Harold McGee[0] is the book for you. It's not for everyone but I give you my guarantee as an internet stranger that you'll love it.

[0] https://www.amazon.com/Food-Cooking-Science-Lore-Kitchen/dp/...

bhntr3··on Colleges are reopening in fall, but many professors won’t be present
Jeez. Little Leland Stanford Jr. must have been a real terror before he died of typhoid at 15
bhntr3··on Ask HN: Did you take any fresh MOOCs as of 2020?
That's a great story! Thanks for taking the time to answer my question. You make good points.

I've experienced links not working. In fact, Berkeley's AI course (https://berkeleyai.github.io/cs188-website/) includes 3 contests but the server no longer accepts submissions for testing let alone actually competing against other students.

There have to be a lot of other people like you and I out there reading textbooks and looking at OCW but never talking to each other. A simple discord server for MIT OCW would be something. I'm currently frustrated because I'm working through the text, lecture notes and problem sets for MIT 6.045J: Automata, Computability and Complexity[0]. But there are no solutions to the problem sets! It's a proof based course and I just want to sanity check if my logic is actually valid.

[0]https://ocw.mit.edu/courses/electrical-engineering-and-compu...

bhntr3··on Ask HN: Recommend a maths book for a teenager?
I'm neither bright nor a teenager but I have been enjoying working through Spivak's Calculus. I picked it up because it was recommended as a good intro to pure mathematical thinking for someone who knows calculus. I've found it challenging but it has delivered on that promise.

There are many good recommendations here but I do think it will be good for them to gain some exposure to pure mathematics. It's different than what's typically taught in high school so they can start to get an idea whether they actually want to be a mathematician or instead focus on applied math in an engineering discipline.

Also you're probably going to get a computing bias here. I found the threads on physicsforums.com helpful so you might ask there as well if you want a different bias. (https://www.physicsforums.com/forums/science-and-math-textbo...)

bhntr3··on Is This a Turing Machine?
I agree with that. Would you agree that the Church-Turing Thesis is different from the equivalence between the lambda calculus and Turing machines? I was trying to argue that even if Church-Turing were somehow (magically) invalidated that the equivalence of the lambda calculus and Turing machines would still hold.

Thanks for bringing up strong Church-Turing. It was interesting to read about it here (https://plato.stanford.edu/entries/church-turing/#StroWeakFo...) I think I had a broader interpretation of Church-Turing than I should. I was thinking that if we somehow discovered a working oracle machine on Mars, it would invalidate Church-Turing. But that article says:

> the thesis concerns what a human being can achieve when working by rote, with paper and pencil (ignoring contingencies such as boredom, death, or insufficiency of paper)

bhntr3··on Is This a Turing Machine?
I think you're commenting on this line?

> can't be computed by a Turing machine or the lambda calculus

I agree they're equivalent. That's been proven. Maybe that would be clearer if it said "both a turing machine and the lambda calculus"? That's how I meant it.

Anyway I was just trying to add to the discussion by pointing out that what makes the Church-Turing Thesis a thesis or conjecture is that it hasn't been proven that "any computable problem can be solved using the device". That's generally accepted but not proven.

The equivalence of turing machines and the lambda calculus is actually a theorem though because it has been proven.

bhntr3··on Ask HN: Did you take any fresh MOOCs as of 2020?
In terms of mind-expanding or intellectually stimulating, I find it's easier to get that from Youtube. I recommend:

3blue1brown for math (https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw)

Everyday Astronaut if you are casually interested in rockets (https://www.youtube.com/channel/UC6uKrU_WqJ1R2HMTY3LIx5Q)

Bret Victor (https://www.youtube.com/watch?v=PUv66718DII&list=PLS4RYH2Xfp...) or Rich Hickey (https://www.youtube.com/watch?v=f84n5oFoZBc&list=PLXsqD83He-...) for programming and design.

Doing a full college course is such a time investment and it usually takes a few weeks to start proving the interesting stuff. I find it more fun to look to shorter content for inspiration. I might realize I don't understand the fundamentals of the topic and then seek out a way to study it.

bhntr3··on Is This a Turing Machine?
Here's an interesting paper on the topic. It seems that only some NN architectures are turing complete: https://arxiv.org/pdf/1901.03429.pdf

To the essence of what I think the original question is asking: It seems possible (due to the turing completeness of some NN architectures) that a language model could learn a sub-network that was a turing complete computational framework + problem solver and then use that to solve problems in natural language. But it also seems highly unlikely that it would actually play out that way.

The linked paper has a nice summary of attempts to build neural nets that learn algorithms. It seems this more direct approach would be likely to yield a result faster than waiting for such a capability to emerge in GPT-12 or something.

Of course most ML models execute from within a turing complete programming language/execution environment anyway. If we want to solve math problems, it might be easier to train a model to translate natural language into a more formal definition (afaik this is still very hard) and then hand off the formal definition of the problem to an automated theorem prover. After all, being turing complete doesn't make all problems solvable or tractable from a run time / memory perspective and that's probably the bigger bottleneck when it comes to solving hard math problems.

bhntr3··on Is This a Turing Machine?
From the article:

> It's not so amazing, the λ calculus is equivalent to a Turing machine.

It seems meta3 is a very basic implementation of the lambda calculus. But I believe the equivalence of Turing machines and the lambda calculus is proven. The church Turing thesis is a bit more. My understanding is that it conjectures that there is nothing that can be computed that can't be computed by a Turing machine or the lambda calculus. So nothing more powerful than the lambda calculus exists. I'm not sure this article shows that.

bhntr3··on Ask HN: Did you take any fresh MOOCs as of 2020?
I agree with your assessment. Most of the MOOC websites have shifted to being much less academic, more applied and are pushing paid credentials. It doesn't seem like most of the top universities are creating much new content for Coursera or Edx compared to 2013.

I mostly use MIT open courseware for self learning. A lot of course websites are also online and if I hear about an interesting course at some University I'll just google the course number. Stanford online and MIT ocw have a lot of lectures on their YouTube channels and I've found some interesting lectures just searching YouTube for a topic. When I wanted assignments and lectures for Russell and Norvig's AI textbook, I googled "Russell AI course" and stumbled on Berkeley's AI course. I've been enjoying it.

How has the social aspect + taking the course with others helped you in the past? I never found it that important. But I have always preferred self learning even in college.

bhntr3··on Microfinance Is Mostly a Scam (2015)
Here's a better article from 2015: https://fivethirtyeight.com/features/microloans-dont-solve-p...

The primary research it's based on: https://www.aeaweb.org/articles?id=10.1257/app.20140287

I think micro finance is largely undercut by the claims made about it in the 2000s. But if it's an important issue for you maybe it's worth reading the linked book (https://www.amazon.com/exec/obidos/ASIN/1609945182/)? The reviews seem fairly positive.

bhntr3··on Santa Cruz, California bans predictive policing in U.S. first
I don't know enough to advocate a solution in the "defund the police" debate and I agree that I don't have enough personal experience from my aeron chair (which, as you know, I was actually born in.)

But, excluding the less police argument, how is the prevalence of violent crime in low income African American neighborhoods a justification for predictive policing? Given the risk of propagating bias and a history of racist policing messing up the data, wouldn't it just be better to keep humans in charge? They've been disappointing recently but I still think they have potential.

Maybe it would be helpful if I put better links here. I didn't expect this comment to get so many responses and my link was a bit flippant. This research probably does a better job of explaining the position than I do.

Here's a Stanford survey from 2016 in Oakland that showed that African Americans are ~4.5x as likely to be stopped as whites despite representing a smaller percentage of the population, that they are 4x as likely to be handcuffed in a stop, 4x as likely to be searched in a stop and 2.5x as likely to be arrested in a stop (which is, as mentioned, 4.5x as likely to happen.): https://news.stanford.edu/2016/06/15/stanford-big-data-study...

I got interested in racist inference due to this Harvard paper that showed that google search ads were much more likely to suggest arrest records for African American names than white names: https://arxiv.org/ftp/arxiv/papers/1301/1301.6822.pdf

I accepted that implicit bias is intrinsic and difficult to weed out when I heard about studies showing that resumes with African American names received much lower response rates than white names: https://cos.gatech.edu/facultyres/Diversity_Studies/Bertrand...

I remember reading somewhere that this effect even persisted with African American studies professors selecting grad students. I can't find that paper but this HBS article describes the phenomenon and discusses how the response rate effect is even present when the company claims to promote diversity within the application: https://hbswk.hbs.edu/item/minorities-who-whiten-job-resumes...

I also advocate for learned epistemic helplessness (https://scienceforsustainability.org/wiki/Epistemic_Learned_...) even if I don't always practice it well. I'm only claiming to have knowledge in the small area of predicting bad behavior using machine learning. It's probably best to find a person of color with personal experience and a better knowledge of the literature for anything else. Of course, you already know that. I hope you're doing well!

bhntr3··on Santa Cruz, California bans predictive policing in U.S. first
Here's the wikipedia article for the fallacious argument I used and which you're critiquing: https://en.wikipedia.org/wiki/Argumentum_ad_populum

Here is some statistical evidence for systemic racism: https://www.businessinsider.com/us-systemic-racism-in-charts... (Shockingly this is actually pretty good despite being from Business Insider)

Honestly, I'd just edit the link at this point if I could since it seems to have triggered so many people. My assumption, again, was that anyone who doesn't believe in systemic racism at this point is not going to be swayed by statistical evidence. I am not advocating that any critical thinker be persuaded by majority opinion.

Luckily we're all critical thinkers here and we crave statistical evidence. I assume that means we also all believe systemic racism is a problem in the US. But in case you're an outlier I hope that link helps.

That said, everything in that link has been true literally forever (for the US.) But opinions are only changing now. Maybe 30% of the United States didn't see those stats before now but I find that hard to believe. I think social proof, despite not being actual proof, is more powerful than you folks would like to believe it is.

bhntr3··on Santa Cruz, California bans predictive policing in U.S. first
Of course that's true. I wasn't so much trying to provide evidence as to point out that people who felt comfortable saying "I'm not racist but I don't think this country is racist either" should not feel comfortable socially expressing that position anymore. Social proof may not be evidence but it is powerful in changing people's minds who might not be as receptive to facts and figures. I don't mind if people believe the right thing for the wrong reasons (at least these days with how anti-science a segment of the population seems to be.)
bhntr3··on Santa Cruz, California bans predictive policing in U.S. first
The problem with predictive policing is in the name. Inference (ML) predicts the future from the past. If the past is racist, then inference will create a racist future. Since racism is systemic[1], especially when it comes to policing, predictive policing is actively working against an anti-racist future.

There may be statistical ways to factor out systemic racism. There are two reasons I don't think that works:

1. I don't see how one evaluates the correctness of the process that controls for racism. What is ground truth for anti-racist policing?

2. These systems are likely snake oil and the vendors of these systems are (possibly inadvertently) profiting off racist policing. If cops arrest more black people per capita, then send the cops to black neighborhoods and have them follow black parolees. The system works (according to an objective function which maximizes arrests.) Remove racism and send the cops to white neighborhoods. Now the cops don't arrest as many people. The system fails. So I think it's likely that if you remove racist policing from predictive policing, you get the null hypothesis.

I'd be happy to hear a counterargument from someone who has actual statistics on this though.

[1] If you don't believe this, you're in the minority now: https://www.vox.com/2020/6/11/21286642/george-floyd-protests...

bhntr3··on Mobilewalla used cellphone data to estimate the demographics of protesters
> I learned recently that Airbnb (YC W09) actually sells your chat logs[2] and stay history(!) to data brokers, which can be used for stalking or kidnapping in the wrong hands.

That's a pretty big misrepresentation of the NY Times article. From the article:

> Sift doesn’t sell or share any of the data it has with third parties.

Sift is a third party risk analysis provider. One of its big selling points is that it can correlate fraud signals across multiple data sets. So, essentially, companies that DON'T want to share data with each other can share data with sift and rely on it to provide aggregate results back. In this way it's the opposite of a data broker. Companies trust it to hold sensitive data they don't want to share.

There are a lot of reasons why any centralized fraud prevention service is concerning and I know that many companies, including Airbnb, limit their use of these signals for that reason. Truthfully, I think the dangers of such a service far outweigh their value and it simply shouldn't exist at all.

Regardless, it's quite different than selling your data to a data broker. If there were any indication that Sift were selling or sharing this data I guarantee they would get dropped instantly. If you believe that a company can't really care about its users' privacy (and you might be justified in that) just consider that sharing this data would be the same as sharing it with the fraudsters it ostensibly tries to stop. That would defeat the entire point of the service.

To a sibling poster, the premise of the NYT article is that Sift allowed end users to download their own data because of GDPR not in violation of it.

There are often threads on here criticizing Airbnb for allowing fraud on the platform and saying they do nothing to prevent it. Here we're criticizing them for sharing data with a centralized fraud system to gain signal on potential fraud.

I think this is the crux of the discussion about the surveillance state vs "defund the police". Algorithmic fraud detection is a frequently racist invasion of privacy. But fraud and crime are also both bad and tend to have a more significant impact on a company's brand / reputation than overreaching in terms of data collection and surveillance. So it's very important to hold them accountable for their use of data but let's not muddy the argument by misrepresenting what's happening.

bhntr3··on MIT 6.001 Structure and Interpretation (1986) [video]
I don't think you really need to "learn a lisp" but I recommend that curious developers read SICP at some point.

John McCarthy famously wrote a lisp interpreter in a half page of lisp in the lisp 1.5 manual. The simplicity of the interpreter makes it possible for SICP to really focus in on the essence of what programming languages are and how features can be incrementally added. Imagine trying to learn programming by building a java compiler. Lisp is a much simpler language for learning how languages work.

The lisp interpreter is so simple because the code of lisp is written in the data structures of the languages: a bunch of nested lists. So the parsed syntax tree is very similar to the code. This means that meta-programming on it is much easier to reason about. You can write macros in other languages to change the functioning of the language itself but it's much more complicated for complex syntax. This makes lisps languages for creating your own language. Some people probably love lisp because they're inspired by it but people who consider its power indispensable (pg) probably customize it a lot with macros.

Someday you may find yourself implementing a lisp unintentionally (https://en.wikipedia.org/wiki/Greenspun%27s_tenth_rule) . IMO this is because it's natural to start with some text based data structure format (JSON?) for some config file or something and then add features incrementally. Suddenly you're making a lisp. Understanding lisp can help you realize you're building a bad one. This has happened to me before.

Lisp is also very interesting if you want to write code that writes itself. For instance, if you were toying with genetic algorithms the fact that every piece of lisp code within matched parens is itself a valid lisp program is a nice feature to have.

In the end though, I don't think actually using lisp is as important as understanding lisp. I can see the insane power of meta-programming for creating higher abstractions for myself or a small team of like minded individuals. But whether it's lisp macros, scala implicits, annotation based byte code rewriting in java or method_missing in ruby, I think that meta-programming becomes a form of obfuscation for really large teams. If you take away meta-programming from lisp you still have some really cool languages. But I think many of those cool things about lisp can be achieved in more mainstream languages if you understand what you're trying to achieve.

After all, Brendan Eich was inspired by scheme when he created Javascript in ten days. But he was told to make the syntax "like Java". That's a bit ironic given that Guy Steele (one of the creators of Scheme) helped write the original Java specification. Neither language is very scheme-like in the end. But both authors contend that the languages would have been a lot worse if they hadn't been around to put a bit of lispiness into them.

So the last reason to learn lisp is that if you ever get asked to make a programming language for a large company with lots of stupid restrictions in a very short time frame you might be able to inject enough lisp into it that people will still use it under duress 25-30 years later.

bhntr3··on Microservices Considered Harmful (2019)
Microservices are the actor model (erlang or akka) except they require lots of devops work, being on call for x services every night, and a container management system like kubernetes to be manageable.

Actors are a simple solution to the same problems microservices solve and have existed since the 1970s. Actor implementations address the problem foundationally by making hot deployment, fault tolerance, message passing, and scaling fundamental to both the language and VM. This is the layer at which the problem should be solved but it rules out a lot of languages or tools we are used to.

So, in my opinion, microservices are a symptom of an abusive relationship with languages and tools that don't love us, grow with us or care about what we care about.

But I also think they're pretty much the same thing as EJBs which makes Kubernetes Google JBoss.

bhntr3··on Introduction to Apache Airflow
When Max started building airflow Airbnb's engineering team was probably 80 or so people. The data engineering and data infrastructure teams couldn't have been more than ten combined. I'm not sure that's massive but it is the point where the previous solution (Chronos: https://medium.com/airbnb-engineering/chronos-a-replacement-...) was starting to strain.

Airflow may not be necessary for a ten person team but it could be if that team has complicated data orchestration needs and doesn't want to incrementally replace their infrastructure every couple years.

bhntr3··on Go + Services = One Goliath Project
> If we moved from Python to a language that is an order of magnitude faster, we can both improve how responsive our site is and decrease our server costs dramatically.

I see people say things like this a lot but my experience is that while other languages are 10x or more faster than python in some benchmarks it's very rare that computation time dominates server latency or that servers are running at 60%+ cpu across all cores.

If 90% of your service latency is not directly on the cpu and/or you haven't profiled to see that the performance bottleneck is evenly distributed across all tasks, then it's super dangerous to migrate to a new language thinking that will fix it.

I hope people inside Khan Academy know this and it's just a clickbait blog. If they really think "go is 10x faster than python so we'll only need 1 server for every 10 when we migrate" then I think they'll be disappointed.

bhntr3··on Interactive Linear Algebra
So far I think I prefer the visualizations in http://immersivemath.com/ila/index.html

For pure visual intuition, I prefer 3blue1brown's Essence of Linear Algebra on Youtube (https://www.youtube.com/playlist?list=PLZHQObOWTQDPD3MizzM2x...). Even though they're not interactive, the visualizations themselves are the most compelling and insightful.

Has anyone used both interactive texts? My initial impression is that the GA Tech text starts with matrices and applications to systems of equations (like Strang) while the immersive math one seems to start more focused on vectors and geometry with computer graphics applications.

Neither seems to have supporting exercises yet which I think really limits their use as primary texts.

bhntr3··on The Mind of a Mathematician
I became a "fan" of Tao after reading this piece in the NYT: https://www.nytimes.com/2015/07/26/magazine/the-singular-min.... It talks about how normal and well liked Tao is. How much he collaborates and assists others in his work.

I think I saw it linked along with this article about the warped portrayal of genius Hollywood feeds us (especially in the Imitation Game): https://www.huffpost.com/entry/turing-the-myth-of-the-heroic...

So Tao's celebrity status as a genius does have that positive aspect. I think he disproves many stereotypes of genius and helps people believe that it's possible to brilliant, decent, humble, social and likable. I think that's useful.

He talked about this eloquently in his own blog where he argues against the "cult of genius" and says one needn't be a genius to be a mathematician: https://terrytao.wordpress.com/career-advice/does-one-have-t...

So, by lionizing him as a genius I think media actually undermines its own narrative about what genius is. In that sense, maybe the cult of genius is bad but if we have to put a genius on a pedestal, it's great to have it be one who seems so normal.

With respect to his work, I always think of what Richard Feynman said in the start of this book "What do you care what other people think?" (https://www.amazon.com/gp/product/0393355640) He tells a story about an artist who holds up a flower and says "I, as an artist, can see how beautiful a flower is. But you, as a scientist, take it all apart and it becomes dull." Feynman thinks he's nutty. He can see the beauty of the flower and the beauty of the science.

People who don't understand Tao's work probably think it's dull. I can't understand Tao's mathematical work. I'm studying math and I really hope someday I could. But math and science have gone so far now that work at the cutting edge really is incomprehensible to people outside the field. So, I'm inspired by the person who does the work. It's just the surface but it's what I have access to.

Feynman talks about this at the end of the chapter that starts with the flower. His father helped him love science and taught him how to think. Then he sent him to university to learn all the things he hadn't. But when Feynman came home, he couldn't explain quantum mechanics simply. So his father never learned what he didn't know. I hope he still loved his son for understanding it.

bhntr3··on Why XHamster Is So Much Better at Content Moderation Than Facebook
Not only that, the founder of xHamster doesn't claim they manually review everything: "We don’t allow just anything to be uploaded and immediately appear on the site. Our A.I. reviews the content for some violations or refers it to our team for additional review. We have had a legion of volunteers... who review uploads in exchange for in-site rewards, as well as the health of the community."

Also, this line from the article is blatantly false and shows they know nothing about how content moderation actually works at Facebook or other social media sites: "Although some links and words do trigger a basic moderation algorithm that prevents the update from being posted, most moderation is done post hoc, often after problematic content is reported by users."

Last I heard a lot of the haskell core team worked at facebook on Sigma (https://engineering.fb.com/security/fighting-spam-with-haske...). Everything goes through Sigma and while Sigma is primarily a rules engine there are also many sophisticated ML algorithms in the background as well. It's ludicrous to suggest that xHamster's "AI" is better than the haskell team's rules and ML from the people who created PyTorch.

Content can be sequestered for manual review. It doesn't even sound like xHamster pays anyone to do this. I guarantee Facebook pays hundreds if not thousands of people to review content already.

Facebook is just as culpable for underage images or video as xHamster. It processes payments so it's responsible for anti-money laundering. I'm sure there's a list of things it does regulate.

The issue is not with Facebook's technical ability or willingness to identify and sequester problematic content. It's all about what they consider problematic.

Making them legally responsible for bullying or hate speech is probably going to happen and may be necessary. But it's silly to make believe they don't already deploy extremely sophisticated technology, sequester content, and perform human reviews for all the things they are held legally responsible for.

bhntr3··on Still Manufacturing Consent: An Interview with Noam Chomsky
This article is from June. I've generally been a bit skeptical of Noam Chomsky's politics but this interaction seems prophetic given the current situation in Syria and makes me think that I should give "Manufacturing Consent" some attention:

>But as Patrick Cockburn pointed out in the Independent, what is happening in Afrin is about the same.

> AM: Happening where, sorry?

> NC: Afrin. Turkish forces and their allies are carrying out the attack in a mostly Kurdish area. Patrick Cockburn has covered it, but almost nobody else.

Doing some research on it, the articles I read in the Washington Post (https://www.washingtonpost.com/world/2019/10/11/who-are-kurd...) said that the Turkish offensive was "days old" while the Independent article Chomsky references (https://www.independent.co.uk/voices/syria-afrin-crisis-turk...) attributes a death toll of 220 to Turkey in March 2018.

bhntr3··on Habits of Expert Software Designers
I especially liked the image of the guy roasting marshmallows on his burning computer.

Error might be opportunity but it's best to triage and fix the immediate impacts of the error, document the root causes, then explore any potential opportunities later.

Similarly, you should put out a burning computer with an extinguisher. It might be neat to know that computers can start fires but I recommend matches or lighters if you want to roast marshmallows.

So, I agree, and I think these articles are actually harmful. The biggest skill is knowing what the most important thing to do is, how much of it to do, and why you're doing it. Doing something because you read it's a best practice done by experts is almost the direct opposite of that because it does nothing to help you prioritize, scale or understand.

Don't be marshmallow guy.

bhntr3··on What are the happiest jobs in tech?
What I found interesting about this article was that data scientist/machine learning specialist was in the top 3 happiest jobs whereas business/data analyst was dead last. As the article notes there is a lot of overlap. Working with data scientists, I found that many thought they would be doing applied ml and ended up doing mostly data analysis.

Since the role of a data scientist varies dramatically based on the organization and even the needs of a team, it seems like a job with a very broad spectrum of experiences and satisfaction.

bhntr3··on Arcs of Seniority
Yes. The article is a bit tautological. It's not defining seniority in terms of time spent. He actually says this at the end. Instead he's defining a "senior developer" as someone who has a necessary mix of the qualities he talks about. It would be more accurate to say "These are the traits your manager will tell you to develop to get promoted."

Personally, my favorite book about old coders is Coders At Work (https://www.amazon.com/Coders-Work-Reflections-Craft-Program...). It's a bunch of interviews with programmers who have created programming languages and lots of the fundamental software we all rely on. It showed me what the journey to being a great programmer can look like.

To me, senior is a corporate term. Great programmers build great things. Senior developers get promoted. I sometimes ask young programmers this: Do you care about the craft or the career? I think being a great programmer will make a person money. There aren't that many truly great programmers. But if they're impatient or they don't think they can be great, they can probably be senior.

Becoming senior is easy: Just help your boss accomplish their goals. Pay attention and develop skills that will help you do this no matter where you are and what you're working on. If you over-specialize in a specific organization or person's need, you become the expert beginner and you can't leave or you will struggle.

Some of the things that help a person be senior can also make them great. But the path to being great is a very different and personal one (at least that's the impression I got from Coders At Work. I make no claims for myself.) Jeff Dean is undoubtedly a great programmer and was also the most senior developer at Google for a long time. They made new levels just for him. So they can overlap. If someone is lucky and their job is great, the things that make them senior can also make them great. If their job sucks and management is terrible, they'll have to choose every day between doing something great or doing something to get promoted (being senior.)

My favorite article on the journey of a software engineer is this one: https://medium.com/@webseanhickey/the-evolution-of-a-softwar... . To me it's the story of someone who started off trying to be senior but then started to become great.

bhntr3··on YouTube videos that have almost zero previous views
Awesome. I've been trotting this idea out for years as an interesting but silly website I should make: videos even the uploaders didn't watch. I'm so excited to see it at the top of HN. This might inspire me to execute on one of my other absurd ideas.
bhntr3··on Power Is Overrated
Great point. I'm not advocating for an anarchist utopia. I think Ursula Leguin's The Dispossessed lays out the problems with that pretty well.

We've probably all participated in organizations with ineffective leadership. I do think it's better to be a clear, decisive and consistent leader than a "cool" boss if you're participating in a power structure.

So it's not even a moral issue. It's just a personal one. I think it's rare for us to really ask if we enjoy holding power over others and whether those people benefit from it. Are there fruits of power (wealth, status) worth it? When does that change?

There is an imbalance of power across all demographics. If I step aside maybe the person who gets is traditionally more powerless. The question of whether I wield it better than them is hard to judge. There's a lot of nuance to when a good leader becomes a bad one. I think it's very difficult to judge our own working of power objectively. And of course power corrupts.

If there is a moral issue, it's just that I think a lot of our systemic power structures are archaic and ineffective at directing human potential. There will always be people craving and consolidating hierarchical power structures. I'd rather engage in organic, decentralized, emergent ones when I do participate.

bhntr3··on Power Is Overrated
I agree that the author's perspective is very privileged. The terms powerless and trapped are nearly synonyms. Yet he overvalues freedom and says power is overrated.

I think a more interesting statement might be that wielding power is overrated. It seems like you could have financial and personal freedom without holding power over another individual. This is something I've tried hard to do (and I think been successful at.) Perhaps it's an illusion.

But I think what the author and others are noting is that most people who gain power then feel compelled to use it. Once you start to wield power, it seems necessary to compound it. Because power is a hierarchy with a nearly limitless ceiling. There's always a bigger fish.

I've never felt comfortable wielding power. It comes from many things: I'm skeptical of central planning. I think most evil comes from misplaced certainty. I'm over privileged and I recognize the inescapibility of unconscious bias. It goes so deep that I prefer declarative programming to imperative. I'm not even comfortable dominating my computer.

Maybe it is just a privileged form of guilt only accessible to those who have attained sufficient power for their own comfort. But my choice to abdicate conventional power and refrain from participation in some conventional power structures (corporations, etc.) has confused a lot of friends and family. Regardless of being privileged, it doesn't seem common to actively choose to avoid wielding power.

← PreviousPage 2 of 6Next →