Data Addict

Startups, Self Quantification, and Internet Culture


Archive for April, 2011

Another Way to View the “Decline” of HN

Recently I was on a public bus in the dark traveling from Dominical, which is a small surfing town on the Pacific side of Costa Rica, to the capital, San José. Across the isle in the dark, my friend and his girlfriend had her notebook open and they were watching a movie. Rather, they were watching a Galifianakis flick, the sort film that was written to showcase the actor’s bizarre non-sequitur style that is fast becoming indispensable in quirky comedies. The bus stopped a few times as it climbed the mountains towards the capital, and during one such break I took the opportunity to ask them how they were enjoying the film. I had been watching bits and pieces (without sound of course) from across the isle and wasn’t impressed.

My jaded film ego wafted across the isle and they immediately called me on it. Within a few minute’s discussion we had concluded that the film wasn’t bad, but that it was typical of a genre and Galifianakis had been typecast. What I also quickly realized was that movies of that type had not become worse, my pretentious movie tastes had.

So what does a trip to San Jose on a dark bus with a bearded funny man have to do with HN? I believe that much of the recent outcry against the decline of Hacker News can be understood as the same process which brought me to decry the quality of a film simply because I was used to, and bored with, the style.

As a long time reader of Hacker News [1], I’m concerned that HN is getting trashed too much without considering the underlying reasons. Clearly, online communities change and grow. With growth comes a dilution of the original intent, and this caused churn. You can choose to believe or disbelieve in evaporative cooling in online communities [2], but I think it captures quite nicely the effect at work on HN.

The problem is that evaporative cooling doesn’t describe the whole picture; there’s also personal change. As regular readers and contributors to Hacker News, we are the most qualified to discuss the community and at the greatest risk of internalizing bias. This gets back to the opening anecdote: people will continue making silly films about bachelor parties just as people will continue making posts about why Techcrunch’s reign is over or Twitter will never find a revenue model. There are a finite number of topics which bear repeating and a limited supply of new information. Yet there is a nearly infinite human capacity to consume and argue about information on the internet. So it stands to reason that the more time you spend on Hacker News, the less satisfied you will be with the experience. You gradually file more and more information as noise, not signal. You came for the learning, but those novel articles about how to bootstrap a company or including game mechanics into your app only enrich you the first 6 times around. Responding with grandparent-speak [3] to the changes isn’t productive.

If you want to drop out of the community in search of different pastures, you probably should, but it seems more personally productive to recognize that a large factor in leaving an online community is personal growth. I don’t like Galiafanakis films, but it’s better for my mood (and those around me) if instead of saying something like “I hated the Hangover, it was so plebeian” to instead say “I’m no longer interested in that kind of movie.”

The take home here is that it isn’t just about Hacker News, this effect is at work all throughout our lives. The music we listen to changes as we gain more exposure to music. We change our tastes in movies, plays, discussion patterns, and living conditions. It’s too easy to blame external factors, “[band] has gotten too mainstream,” “[playwright]‘s works are too determined,” “she always wants to discuss the same things,” “the Bay Area is just too insular.” In many cases external factors have not been radically altered, you have just grown. In the case of Hacker News, I’ll admit that things have changed, but from my perspective it seems that I have changed 80% and HN has changed 20%. I love the idea that my tastes are evolving and changing, and if that means Hacker News is no longer interesting to me in six months, that’s more worthy of celebration than gloom.

I do not intend to stop reading or participating in Hacker News, although I have already cut down substantially on my daily reading binges (productivity was being harmed). I’ve learned a ton from the community, so less of the content now holds my interest. I’m proud that less of the articles now interest me because it means I’ve learned the lessons contained therein [4]. I’m also grateful to the HN community for giving me such a useful education in how to run my company.


Special thanks to Nick Winter, my trusty editorial and bullshit detector, for helping me to write this.

[1] I have been regularly reading Hacker News for about 4 years. I can vividly remember the switch from Reddit to HN and just being blown away by the quality of discussion and filtering of posts. Just as I had done when I discovered Google for the first time, I immediately bookmarked the site, and what followed can only be described as a love affair.

[2] Evaporative cooling is best described by Hang at Bumblebee Labs in this post: “[Evaporative Cooling] occurs when the most high value contributors to a community realize that the community is no longer serving their needs any more and so therefore, leave. When that happens, it drops the general quality of the community down such that the next most high value contributors now find the community underwhelming. Each layer of disappearances slowly reduces the average quality of the group until such a point that you reach the people who are so unskilled-and-unaware of it that they’re unable to tell that they’re part of a mediocre group.”

[3] Grandparent-speak can often be identified by the presence of phrases like “they don’t make ‘em like they used to” or “it’s not like it was in my day.” Not only does such rhetoric lack critical hindsight, but it also perpetuates ageism: young kids learn to dislike their grandparents for denigrating their world and grandparents learn to further dislike their descendants when they respond with irritation to their nostalgia.

[4] At least, I think I have learned the lessons, certainly this could be arrogance on my part, and some knowledge genuinely bears repeating, which is why I wrote this blog post a while back.

Why Recessions Suck More Than Expansions Rock

We technophiles are sitting pretty with regards to jobs and pay, even in this recession. According to The Wall Street Journal, Forbes, Careercast, and CNN Money software engineering is rated in the top 3 jobs in the nation for 2011. Other job descriptions in the tech industry consistently rank in the top 10. Despite this relatively sunny outlook, recessions in the tech industry still suck more than expansions rock. Here are the empirical reasons why:

1) In tech, raises don’t make us happier.

There is a body of empirical research that states that all things being equal, more money doesn’t make people any happier beyond a finite threshold. One of the most authoritative researchers on the topic of human happiness, Daniel Kahneman, made this clear in this recent TED talk [1]. In a representative sample of 600,000 Americans, Kahneman found that respondents who earned more than $60,000/yr showed no increase in experiential (moment-to-moment) happiness.

Salaries for experienced programmers easily exceed $60,000/yr, which means that in times of prosperity, when pay is higher and raises abound, the added income doesn’t tend to impact moment-to-moment measures of personal well-being.

During recessions on the other hand, salaries stagnate, jobs are harder to find, and many people find themselves making less than $60,000/yr, which means–strictly speaking–that making more money would make them happier. Recessions therefore suck more than recessions rock because less money does make us less happy whereas more money has diminishing positive effects.

2) Money aside, the underpaid suffer more than the well-paid prosper.

Studies [2] have found that those who are comparatively underpaid relative to their peers are significantly less happy than the well-paid are pleased with their success. In other words, when there is a significant (and transparent) pay gap, there is an aggregate loss of happiness. Although analyzing and breaking down unemployment numbers is a notoriously contentious topic (how do you define “discouraged workers” for instance?), it is not contentious to assert that during the current recession unemployment has risen. In the purest sense, complete unemployment is a wage of $0, and when large numbers of people are unemployed there is therefore a significant and transparent wage gap.

The effect of social comparison on well-being is implicitly understood in the professional working world: it isn’t polite to talk about it generally. This tactic of relying upon social conventions to keep inequity hidden works fairly well when most people are employed or not struggling financially. When that doesn’t hold, however, simply being polite can no longer obscure inequity, and people suffer. Times of high unemployment and under-employment provide people with a glaring sense of earning inequity that is both obvious and hurtful. So, even when absolute monetary compensation is not an issue, income inequality leads to greater unhappiness during recessions.

3) Humans [irrationally] suffer more from downturns.

If you offer someone a salary that increases from $35k-50k over a year as opposed to a salary that goes from $70k-50k, people will widely prefer the first, even though it earns them $35k less [3]. In studies in which subjects were given a choice over when an unpleasant experience would occur, people overwhelmingly chose to experience it earlier and then have a positive conclusion, i.e., an improving sequence of experiences.

The quirky thing about these preferences is that they have been proven to be inconsistent with general happiness optimization. Consider the study in which the researchers concluded that controlling for peak pain and end result, men would prefer a 60 minute colonoscopy that forced them to endure a declining amount of pain than a 9 minute colonoscopy in which they experienced a rising level of pain [4]. Even though a 60 minute operation induces much more pain total pain, subjects widely preferred it. In this way, sequences of outcomes cause us to act irrationally and select experiences that don’t give us the most happiness.

Sequences of declining goodness make us irrationally unhappy. Witnessing the negative effects of an economic downturn will cause us to overemphasize the negatives in a way we don’t naturally tend to do with upturns.


The takeaway here is that at least in aggregate, empirical studies suggest that we in the tech community should be significantly less happy with our lives during a big recession than we might otherwise be, but that a commensurate big expansion will have a smaller-than-expected happiness-enhancing affect.

The key here is the phrase “in aggregate.” Individuals are capable of learning from our cognitive biases and re-training how we perceive stimuli. With knowledge of how those biases work, we can make our own world a better place to inhabit. Personally, I have found that simply knowing the direction in which my biases tend gives me the ability to compensate. For example, when news about the stagnant economy starts getting me down (I think I read too much Econ news), I remember how declining sequences of events irrationally depress me and I can short-circuit the negativity before it has a big impact on my state of mind. It’s not scientific, and I don’t have citations to prove that method works for everyone, but I firmly believe that cognitive re-training is possible and beneficial.


[1] The statistic in question is brought up in the question and answer period, which begins 17 minutes into the talk.

[2] Studies concerned with this question deal with range frequency analysis, many of which have been co-authored by Ed Diener, but the specific study I am referring to is this one:

Hagerty, M. R. (2000). Social comparisons of income in one’s community: Evidence from national surveys of income and happiness. Journal of Personality and Social Psychology, 78(4), 764-771.

[3] This is covered in the Kahneman TED talk in greater depth.

[4] Cited in Thinking about Values in Prospect and Retrospect: Maximizing Experienced Utility, Joel Huber, John Lynch, Kim Corfman, Jack Feldman, Morris B. Holbrook, Donald Lehmann, Bertrand Mueir, David Schkade, Itamar Simonson