], but also how it may (and may not) change business strategy.
(thread 👇)
Today's dominant internet platforms are built on aggregating users (and their data!) to harness network effects – a key source of competitive advantage. [hbr.org/2019/01/why-so…]
BUT...
... how does competitive advantage work in the more open, user-focused ethos of web3 – where, for instance, users can own their digital assets, which are typically created according to interoperable standards on public blockchains (instead of on a company's private servers)?
Because a web3 platform's users own their digital assets, they can take them with them when they leave an incumbent platform in favor of a competitor.
In web3, new entrants can incentivize power users to move over to them through a strategy called a "vampire attack" that uses an incentive to "suck" users out of another platform (more on how this works, and views of whether it has worked or not at blog.coinbase.com/vampire-attack…).
Plus web3 platforms run on shared infrastructure layers – typically, public blockchains – which again makes it easier for new entrants to start out: they can innovate on top of existing software, and plug into existing content networks.
So while portability of digital assets increases the potential for competition, it also makes the overall dynamics of Web3 less zero-sum – and that means a platform's value creation opportunity can be bigger for all. hbr.org/2022/05/why-bu…
The ability to integrate pre-existing digital assets can help resolve the classic "cold-start problem" for new platforms; help existing platforms bootstrap off existing assets (vs. uploading directly) [medium.com/etherscan-blog…]...
Sharing infrastructure means that web3 apps can focus on building great experiences – driving towards a greater emphasis on platform design as a source of competitive advantage. Even in Web3, design and user insight will continue to differentiate apps and marketplaces.
All of this suggests an exciting new era of business strategy is being born, even while drawing on classic strategy frameworks.
Shoutouts (again!) to my incredible co-author @Jad_AE; to our fantastic editors @tom_stackpole and @GretchenMarg (and @smc90); and to @lukeabc for the extraordinary header art. Special shoutouts here to @ccatalini, who's shaped my thinking around web3 and competition for years.
QED, and gm.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
(It's not clear that the boost to downstream outcomes like flyouts could hold in long-run equilibrium, or that added visibility really is the mechanism driving it. But still, pretty striking, QED!)
Link:
Disclosure: as noted in the paper's acknowledgments, I was one of the people who participated in the experiment as a Twitter poster.papers.ssrn.com/sol3/papers.cf…
Misinformation is on a massive rise, and social media platforms are scrambling to figure out what to do about it.
In @TheAtlantic today, my @Harvard colleague Jesse Shapiro and I offer a surprising answer: 👇👀
Platforms should give up.
𝑊ℎ𝑎𝑡!?, you're thinking – but before you throw tomatoes 🍅 or quote-Xweet this post to dunk on it, read on. 🧵👇
We're not saying that misinformation isn't a problem, but rather it's one that platforms aren't well-positioned to solve – and we have a suggestion for what they should do instead. theatlantic.com/technology/arc…
Using percentage-of-deaths-in-region as a metric for thinking about a preventable illness is absurd even leaving aside all of the risks from "mild" Covid cases.
Suppose drunk driving deaths doubled. According to @CDCgov, that means we should care less about preventing Covid!?
Apropos of nothing in particular, I've been thinking about my advisor Al Roth's keen observation that "rules that shouldn't need to be rules" are often a sign of market failure.... 👇
... For example, before it switched to a centralized algorithm, the US medical residency match apparently used to have a rule that in effect said that offers could only be made during the official offer period. 👇
... The fact that such a rule needed to be instituted highlighted something going on under the surface, which is that more and more offers were being made early, locking in candidates before the offer period had started, and thus "unravelling" the market. 👇
Shanah tova, QED! May your 5783 be a prime year in as many ways as possible!
(5783 is prime, and is also a concatenation of three primes. And that's not all..!)
🧵👇
The last prime that was a concatenation of 5, 7, and a prime was four years ago (5779). The next one is five decamillennia away (57107). Even the next prime year ending in "83" isn't for over a millennium – it's 6883!
Plus 5783^10 + 5783^9 + 5783^8 + 5783^7 + 5783^6 + 5783^5 + 5783^4 + 5783^3 + 5783^2 + 5783 + 1 = 41841479054317853653109050661035088713 is prime, as are 545783 +/- 1 (= 312283 and 312281).
Imagine this trend extrapolated out to the limit. Do you think the @WhiteHouse would be proudly reporting "329 million people received Paxlovid this month"?
That would mean nearly the entire American population had gotten covid within the month!!
Plus as @wsbgnl points out, currently Paxlovid is earmarked for high-risk individuals; we shouldn't be celebrating the fact that the number of high-risk individuals getting infected is going up, even if we are successfully getting them treatment.