Technology does not automatically make the world better. Most tools can be used for both good and bad. The solution to bad outcomes due to technology is not more of the same technology.
A discussion of how we consume information, and the cyclical nature of how we share things online. Evans’ comparison of the number of items in your news feed to the number of invitees to a party is a clever analogy (kinda like Big O!) but the essay is fraught with uncharachteristically bad assumptions. Stories are less so about “units of content” than about market segmentation. Stories are a news feed, even if in a different format, and don’t solve the oversharing problem - you still get N*M pieces of content.
This problem has existed for ages. Technologists and scientists work on things because they are cool, and because they want to push science forward, regardless of the implications. The debate about whether technology is inherently good or not will keep going.
This column asks questions about the future of technology, and which technologies are worth supporting. As you’d expect, it brings in some “marginal cost of 0” ideas, along with questions about what the structure of society should be, and what should be the role of corporations in a financialized world. Ultimately, it pushes us to remember human agency. Being a cog in one of these machines, I can tell you how easy it is to forget we have that.
This one is a rollercoaster. Madrigal decides to buy a cheap coat off of an Instagram ad, and upon receiving the item starts wondering how it got to his home. He digs further, only to fall in the rabbit hole of modern e-commerce. Beacuse the front-ends are shiny squares on Instagram and Shopify, we usually don’t think of the cross country supply chains fed by Chinese entrepreneurs on Alibaba, or who is even doing that Instagram marketing in the first place. This is a good peek at the bowels of the machine.
As a heavy Twitter user, this is not at all surprising. The platform does not try to hide its spam. Notifications for automated likes, follow/unfollow schemes, and unrelated replies from random bot accounts are daily staples. Twitter staying away from politics has nothing to do with them not addressing the bot accounts. The reason this has not been fixed is that the yardstick Twitter gets measured with is the number of active accounts, and being able to say there are millions of accounts pumping content into the network is valuable to them - it props up their stock price, and keeps investors interested. Don’t get me wrong, I think some bots are valuable to the platform, and have coded up a few myself, but bots that merely amplify the reach of someone else’s account don’t improve the experience of any Twitter user. Of course all these bots disappered right after the NYT published this piece.
Men and women have substantially different experiences as they interact with the world, off and online. We’re addressed in different ways, we’re seen in different ways, and we’re marketed to in different ways. Hallden’s argument is that this last one is especially exacerbated on platforms like Instagram, where she is constantly barraged with hyper sexualized advertisements. So much for hyper-targeted ads from companies that know everything about us. This is why you need diverse teams. How did no one at Instagram catch this and say “maybe there’s a problem here”?
If you have been following along lately, I’ve been talking a lot about tech and politics. Here, Ben Thompson summarizes last week’s hearings. He describes the controversy as “largely centered within the coastal tech-media bubble,” and I could not agree more. The real question is what average Jane and Joe in the rest of the country think about this, and those are exactly the voices that these politicians (in theory) represent. I’m convinced the hearings were useless in the grand scheme of things, but the arguments presented will be tested in other arenas in the months to come - I can’t wait to see where this ends up…
The internet has changed a lot in the last 20 years. Ford does a little recounting of his time online, and the rise of walled gardens versus the open web.
More on the social cost of Twitter, and the kind of dynamics enabled by having social networks without strong principled moderation. It is always intriguing to read opinions that have morphed over time, and to go back to the origin story that most observers can’t tell first hand.
Hearing Benedict Evans and Tim O’Reilly discuss O’Reilly’s new book was good, but a lot of it was a rehash from the previously shared EconTalk episode. About halfway through there’s an interesting discussion on optimization. We’ve created institutions that optimize for certain metrics at all cost. At the micro level, we have companies building machine learning models to drive engagement, but at the macro level we expect companies to maximize shareholder value. This is a human decision, codified into law to maximize welfare - at least in theory. Perhaps trusting the market mechanisms and the individual search for arbitrage opportunities is no longer enough. There might be other trade-offs to consider in how companies, and the market at large, are run. In a way, the market acquires a life of its own, not too differentt from a paperclip maximizer.
I absolutely hate Buzzfeed’s super optimized methods to grab my attention, and knowing that’s the goal I try to avoid their non-investigative content. However, Tasty is a great idea, and it is very well executed. I probably have burned hours of my life looking at their cooking videos, and yet not once have I tried one of their recipes (even though I cook almost every day!). There is just something about melting cheese oozing out or chocolate drizzling that makes you want to keep watching.
The ethics of journalism, and the history of how modern journalism itself came about are interesting topics that I don’t know much about. I should work to change that. In the current historical context, it is important to understand how and why the content we consume is created. This post was a little too consparicy theory heavy for me, and yet I thought it was a worthwhile read. As I mentioned to my friend this weekend, I worry about the future of journalism. Laurene Powell Jobs buying a majority stake in the Atlantic or Bezos buying the Post kind of works in the short term, because their ideologies align with mine, and I kind of trust their intent, but tell me that the Koch brothers are buying the WSJ, and my reaction would be different. Creating incentives to keep the editorial integrity while maintaining a viable business is a tough 21st century problem.
Who do we trust in a world where everything is falsifiable? I am really curious about how cryptography and provable mathematics could change this. Companies like Keybase are already working on it, but it is far far from mainstream. Even technical people like myself have trouble wrapping our heads around this issue.
The first of a two part show. Alex gets a scam phone call, and he follows it to the source.
Holding the keys to the content is not as important as it used to be. Partially, I think this has to do with the fact that the market for content has been totally flooded, an aspect that Evans does not touch at all in his article.
If you haven’t heard about Reddit’s April’s Fools experiment from earlier this year, you need to read this. Emergent behavior is awesome.
Nothing about Farhad’s argument seems controversial to me. There are still unsolved problems in the vein of “finding a needle in a haystack” where separating signal from noise has become increasingly difficult, but the expansion of content produced by humans can only be a good thing. If we trust that willingness to pay will somehow sort out the good vs. bad content problem in the long run, we’re headed in the right direction.
It has been interesting to see Ben apply aggregation theory to politics more and more. I agree with the views presented in this article about centralization (or lack thereof), regulation (or lack thereof), and market solutions (or lack thereof).
Since Snap’s S1 came out a couple of weeks ago, everyone has been discussing whether moats exists or not. The fact that their whole thesis revolves around the disintegration of sustained competitive advantages is fascinating. Evans’ index fund analogy adds an interesting idea to the mix: Facebook, Instagram, and Google must reflect reality and serve billions, while Snapchat will aim to create N things, each worthwhile to M million people, such that N*M becomes significant while not overtaking the role of the index.
I have had an Echo for several months now, and I still see it as a gimmick, but I understand why the strategy behind the device has so much going for it. Amazon is building a platform that makes a lot of sense, but the technology isn’t quite there yet. It’ll be interesting to see this pan out.
A fascinating story about figurative walls coming down. It made me think a lot about the inadequacy of the label “third world.” Coming from Costa Rica, which has the highest internet access rate in Latin America, the experiences told in this story are shocking. Widespread definitions and categorizations sometimes don’t make sense.
We are nowhere near a transparent general AI, and all the companies buidling voice interfaces know that. By now, Siri, Alexa, and Google Assistant have all been positioned away from such an all-encompassing solution, pushing for ever narrower use cases. Like the AIs out of Facebook’s bot fiasco of 2016, voice interfaces seem to be adding more friction than they take away.
I am sharing this, even though I honestly did not read the whole thing. An 11 part epic on how the internet works. A good production, even if a bit overwhelming.
As usual with Maciej, there are many layers to this essay. The comparisons between libraries and the internet are not new, and his railing against large companies aiding online surveilance are more than expected. Much more interesting are the questions brought up about archiving the modern web - where content is selected, joined, and rendered dynamically per user at load time, with large portions behind walls: What is the point of building a community you don’t own? What should be kept for posterity? What is a the point of a site’s snapshot without the code that makes it work? What happens when a company dies, or misses, and we go beyond simple link-rot? The conclusion is hand wavey, but the future of the internet is, as Maciej put it, contingent.
There are clear tensions regarding how information is stored and accessed on the internet. In the OSS world, there is a loud group that constantly complains about the IRC => Slack trend, for example. Whether the fringe is becoming more or less accessible, I don’t know, I have not tried to hang out there, but there is an overwhelming feeling of the walls closing in on themselves.
Another piece about the perils of living attached to our screens, and taking a break from the addiction. These have become more and more common, but somehow Sullivan gives a refreshing view.
Starting a career in software engineering during the days of AWS and Heroku gives me a strange vantage point. The story of how Netflix switched their whole infrastructure would not be half as impressive if I didn’t understand the role of culture in organizational change. The fact is that “this is how we do things around here” can make or break you. This episode talks about the architecture that underlie the modern web stack.
A follow-up on last week’s post on Docker, and the state of distributed systems on the web. This one being the non-satirical version.
As usual, Evans gives us a lot to think about. Our phones aren’t really just phones, and our cameras aren’t really just cameras.
“Serverless” is the Next Big Thing. Serve static stuff on S3, make everything stateless, and plug a bunch of functions in Lambda. AWS once again changing the paradigm.
I like to think of myself as a much more fox-ey than hedgehog-ey person, and I think success comes from getting a fox to surround themselves with hedgehogs. This is true in the AI context, as well as many others. “To think about tech now is to think about many things.”
The internet: where nothing is really private.
Related, and somewhat more in depth, is the a16z podcast on emoji. Why do these icons carry so much weight?
Leadership is hard, and requires personal sacrifice. People appreciate that, and can tell when a person in power doesn’t put in the required effort. According to Woehr, Marissa did.
In a recent post, I shared Charlie Warzel’s analysis on the future of payments. This episode of the a16z podcast takes a deeper dive.
HTML, Flash, Video, etc, are only a medium. Corporations today are working hard to exploit these new means of distribution.
Yes, two posts by Evans today. It’s that good.
Orwell would be proud.
Extreme clarity on the future of journalism, media, and strategies for companies in the space to respond to change. TL;DR: create better content or disappear. The arguments fit perfectly with Aggregation Theory, and while the article is a bit too focused on politics, the analysis could apply to any other news covered by the media, from the Tech Bubble, to ISIS, or Millenials. Long, but worthwhile.</br>I have been reading Baekdal for years. I can’t even remember how I ran into his blog, but it must have been 7 or 8 years ago, and I am glad I did.
Possibly more interesting than the opt-in model championed by Blendle.
Ben sounds more bullish in this article than in the past few, especially Exponent.
Yet another bear case for Apple pinned on the cult of personality for Steve Jobs. While I disagree with the overall message, the writing is really good, and Lefsetz does have a point on the strategy of innovation, viz. Christensen’s disruptive innovation.
I don’t buy the bot craze. The technology is not there yet, and as the author well describes, the user experience feels just like calling a bank, or a telco, and being greeted by a distorted digital voice asking how one can be helped. Some day.
Sometimes, free is a problem.
A little bit hyperbolic, I use Feedly and it works more than fine, but the internet definitely changed the day Google shut down Reader.