April 30, 2019 (1346 words)
::
A summary of the last 30 posts.
Tags: recap
This post is day 120 of a personal challenge to write every day in 2019. See the other fragments, or sign up for my weekly newsletter.
This month’s recap
(These no longer have new descriptions by default. I’m sorry. I got lazy.)
- 91: If tech job postings were honest. For execs: money, more money, and protection from getting #MeToo’d. For contractors: tepid pay and working conditions, but if you complain you get fired.
- 92: When your startup turns sour (on npm). It’s time to make money, which means: cost-cutting, layoffs, and a desperate rush to turn the product into a revenue engine, no matter how much that damages reputation. (A fictionalised postmortem of what went wrong at npm, Inc, a for-profit company that maintains the node.js package manager.)
- 93: On YouTube’s engagement-driven recommendations. Reflections on today’s Bloomberg piece about YouTube’s unwillingness to stop recommending toxic videos in order to preserve high ‘engagement’.
- 94: When gatekeeping is legitimate. Some amount of gatekeeping in tech is valid, because not everyone is suited for every role. The problem is when the incentives are distorted by economic factors.
- 95: When the seas rise. I have still not come to terms with the fact that things could be catastrophic, and I’m not sure how I ever could. (Reflections on an article about the rise of sunny-day flooding in Miami.)
- 96: The billionaire who wants to reform capitalism. Hedge fund manager Ray Dalio thinks that capitalism needs to be reformed. Who cares. (More broadly, I think the political views of anyone with an outsize degree of power should be treated with suspicion.)
- 97: Workplace surveillance. Workplace surveillance is more than merely an issue of privacy. It’s about power, and control, and deepening a relation of class domination.
- 98: Eero’s fire sale. The seemingly singular story of a distressed startup’s acquisition which inadvertently reveals the con embedded within capitalism.
- 99: Why Peter Thiel praises monopolies. Thoughts on Peter Thiel’s political belief system and the unstated assumptions buried within it. (The gist of it is that he thinks monopolies are good when they are run by people like himself, because people like himself are intrinsically good.)
- 100: Gratitude is a trap. How dare you criticise a system that provided you with the means of criticising it in the first place? You should be grateful. (A follow-up to fragment 51, which challenges the idea that it’s hypocritical to criticise capitalism if you own an iPhone.)
- 101: If the market doesn’t give us affordable housing …. An elegant way to fix the housing crisis would be to take housing out of the market sphere. Here’s one way we could do that. (tl;dr: reduce the amount of profit that can be made through providing housing, and if the market fails to deliver on housing as a result, use that as an opportunity to evade the market entirely by building more social housing.)
- 102: There are experts, and there are experts. Liberals love to scorn conservatives’ dismissal of so-called ‘experts’. But there is no ultimate technocratic authority to appeal to. (In other words: being recognised as an “expert” doesn’t mean you are free from ideology.)
- 103: Insulin as a public service. The public health crisis created by rising insulin costs is a strong argument for public provision of important drugs. This isn’t as impractical as it sounds.
- 104: Unbundling progress. How can we criticise capitalism if it’s brought us so much progress? (A follow-up to fragment 100, on gratitude being a trap, and a rejoinder to the Pinker-esque line which dismisses today’s high levels of inequality on the grounds that things are better now than they were.)
- 105: Why we can’t have free online tax filing. Powerful corporations view politics as an arena in which to ensure their continued existence, at the expense of everybody else. (On Intuit lobbying to prevent the US government from offering free tax filing. A few days after this fragment came out, ProPublica published an article about TurboTax deliberately hiding its free product from search engines, in order to steer customers towards paid products.)
- 106: Building a better billboard. Advertising technology solutions tend to optimise for the world as it is, not the world as it should be. (We don’t need more granularly-targeted billboards; we need fewer billboards.)
- 107: Finding the breadcrumbs. I found the left at a time of my life when I was forced to admit that I didn’t actually know as much about the world as I thought I did.
- 108: Locally optimal, globally absurd. Housing is an arena where the choices available to an individual are constrained within a narrow set of possibilities, each of which is globally suboptimal. (On the renting vs dichotomy for individuals, and why those aren’t the only options on a larger scale.)
- 109: When is a company a ‘tech’ company?. And why does it matter? (Tech companies’ high margins tend to come from 1) underpaying or simply misclassifying employees, and 2) hoarding intellectual property.)
- 110: Borrowing an ebook from the library. All copies of this ebook are currently checked out, so you’ll have to wait until one becomes available. (I strongly believe that digital content like ebooks should be available to everyone, at any time, for free, and I would like to transform the economy so that such a thing is feasible.)
- 111: Why do people work?. Under capitalism, the way to encourage someone to work is to offer them money. But the effects of offering more money are neither guaranteed nor uniform. (Socially-important work is often underpaid, whereas some of the most pointless or even destructive work is overpaid. This is a highly inefficient way to run an economic system.)
- 112: On internal migration restrictions. When China restricts rural-urban migration, it’s bad. When the same sort of restrictions are created by capitalism, that’s totally fine. (n.b. I do not actually think this is fine.)
- 113: What’s wrong with inequality?. Surely inequality is just a natural consequence of innovation. And anyway, it’s not that bad.
- 114: We can have entrepreneurship without extreme wealth. (continued from day 113) Contesting the idea that you need the promise of massive financial reward to spur socially-useful entrepreneurship.
- 115: Economic inequality is almost never harmless. (continued from day 113) The reason economic inequality is a problem is because it grants some undue power over others.
- 116: When the sorting function is broken. Our current ‘meritocratic’ sorting function is a terrible way to sort people into a system with highly variable economic rewards.
- 117: 996.ICU and the rewards of hard work. Some Chinese tech workers are being asked to work 9am-9pm, 6 days a week. Will this serve as a wake-up call for Silicon Valley to start working harder? (Some Silicon Valley investors would like that very much.)
- 118: Customer happiness isn’t enough. The typical goal of antitrust enforcement is to ensure better outcomes for consumers. But customer happiness is not the most important factor when it comes to tech companies. (The customer is not always right.)
- 119: You are not your code. Why software engineers need to worry about the social, economic, political implications of the products they build.
Major themes
- Intellectual property is bad; same deal with inequality
- Criticising capitalism, on the other hand, is good
- No one should have to work at a company whose mission they don’t believe in simply because they need money
- Be suspicious of those with power over you (what do they have to gain from making you believe something?)
- Me finishing these fragments later and later
What’s coming
I always think my posts will get shorter. I don’t think they’ve really gotten shorter yet. Next month, though, they will get shorter for real.
After writing these (nearly) every day for four whole months, I think I now have a pretty good sense of what I’m going to cover in my book, which I will be working on furiously for at least the next month. Please do not commission me for any writing or events until it’s done, I beg you.