Neil Postman 的《娛樂至死》

娛樂至死》這本書雖然討論的議題(媒體生態學、媒體科技對文化的影響等)和當代非常有關,但書本身是一本很硬的哲學書,很慚愧的我也看不太完。

新版介紹有一段印象深刻、切題的摘句,Postman 比較了歐維爾的《1984》描述反烏托邦,相對於赫胥黎的《美麗新世界》,下譯:

歐維爾恐懼禁書的政權。赫胥黎所害怕的,則是再也沒有任何禁書的理由,因為沒有任何人會想要看書。歐維爾害怕資訊被剝奪,赫胥黎則害怕過多海量的資訊導致我們更被動,更本位主義。歐維爾害怕真相被蒙蔽,赫胥黎則害怕真相被無關的事物所淹沒。歐維爾害怕文化被當權者掌握,赫胥黎則害怕文化被瑣事佔據。赫胥黎評論,時時警醒反對暴政的公民學者與理性主義者「未能考量到人類對於分散注意力的近乎無限的渴望」。

Tristan Harris(Center for Humane Technologies;《智能社會:進退兩難(The Social Dilemma)》)的 Your Undivided Attention podcast 也剛好講到了這句

The Google Web

I’ve been closely watching coverage on Google’s two antitrust trials on Big Tech on Trial.

I am not naïve — as a public-traded company that is only and rightfully accountable to the shareholders, I have always known Google is not simply benevolent when it comes to funding the Web. But not only until hearing about the details revealed by the States in court, I realized how successful Google has been (and illegally, alleged by the States) at capturing both ends of the Web, effectively capturing its values in a way that is as profitable as closed platforms.

Google has been paying billions of dollars to all browser vendors to allow Google to be the default search engine1 — something the court deemed to be illegal just recently — to capture all traffic that goes into the Web.

In the ad marketplace, Google has captured the entire ad revenue stream of the content Web, making it impossible for any content provider to generate ad revenue — the primary business model of content — without Google. This is something the States allege to be anti-competitive, and the court is going to weigh in pretty soon.

As it turned out, the Web being a ubiquitous technical success is a chess piece of Google’s money-making machinery. People who claimed to advocate for Open Web without talking about the nuances are just doing community evangelism for Google without getting paid by them.


Where do we go from here?

The browser engines need new business models and revenue streams. People are surely talking. I agree that engines are implementations of the one and only viable mostly non-proprietary application platform. They deserve to exist as public goods. Regretfully, unlike rails and tunnels, their public benefits are as abstract as any other software infrastructures (obligatory XKCD reference). We are often blind by our passion for technologies to think of them, but it will take some persuasion to be entitled to public money — skipping the persuasion part sometimes sounds arrogant.

The ad revenue dependency of the content web is a much harder problem. The Sherman Act can only ensure the market is fair; it can’t create new markets. Donations/subscriptions have been successful in a few content segments, but they have not yet become mainstream. Lots of YouTubers have been talking about that — I have no more insights to offer than to provide my support occasionally as a consumer2.


Do I still believe in better access to the Web and better application capabilities offered on the Web? Honestly, it is a MAYBE now.

I see great use cases of powerful APIs like Web Bluetooth and agreeable arguments made for them, but I also value the security sandbox promised by the Web. I wanted to see features that prioritize the decentralized root of the Web than re-enforce it, but nothing has truly happened yet other than questionable offerings from cryptobros.

Again, looking at the grip Google has on the Web right now, I would decline to be their useful idiots.


  1. Which is the primary revenue of my previous employer, and I benefited. I am certainly also benefited indirectly by working in tech and as an index fund investor. ↩︎
  2. Yes, my account credential was leaked on Patreon. ↩︎

The Software Crisis

We developed methods of building nested layers of abstractions, hiding information at multiple levels. We took the problem of constructing software and morphed it into towering layers. We integrated these layers into the software required to use our computers, and the software that drives our lives.

From The Software Crisis.


One pivotal distinction between senior and junior engineers is how they manage abstractions, both in building them and understanding them.

After all, “We can solve any problem by introducing an extra level of indirection.” The temptation to solve problem at hand with another wrapper object or adding an if block is irresistible. The abstraction layer could also be a comfort zoom that many may unintentionally trapped themselves in; it is more rewarding for them to build cathedrals in their layers than reaching down and out (yes, this has been a critique of Web engineering and myself in the beginning of my career.)


How would you like to survive in this era of crisis?