3 ms·

All your examples have fundamentally new stuff at the core though? E-commerce lets you compete with hundreds of brick-and-mortar locations using a single wareho

All your examples have fundamentally new stuff at the core though? E-commerce lets you compete with hundreds of brick-and-mortar locations using a single warehouse. Crypto enabled global 24/7 markets. CDOs were derivatives with built-in hedges: the houses. Enron is an accounting fraud case, not a bubble—energy trading is still super profitable.

acdha 3y ago
E-commerce isn’t on that list for the same reason AI isn’t: it had real value. Cryptocurrency had a ton of money poured into it but never found something it did better than the alternatives - PayPal didn’t even feel the need to pretend to treat customers better! – and magical thinking about security and customer service not mattering didn’t help. Pets.com is harder to understand if you weren’t there: it looks like a not bad idea, which it was, but the business was setup as if it was a lucrative high-margin industry. There was no way it could succeed losing money on every sale, and they locked those costs in so well there was no real way to lose less money unless you could figure out how to make shipping far cheaper (most of what people buy at pet stores is heavy food). The hype was off the charts, though, and people poured money into the stock anyway. I worked with a major competitor and was on a few calls with their ostensible technical staff, and it was palpable how they had hired anyone who could spell HTML and everyone was killing time waiting for their options to vest rather than building anything real. AI is like the early web in that regard: there are plenty of outright fraudulent companies and a bunch of doomed ones who just don’t realize it yet, but there is also a real capability that wiser people will use. As with the early web, I’m expecting a lot of startups to get a ton of attention compared to boring old companies actually doing useful things.
fancyfredbot 3y ago
There was definitely some level of novelty in each case I mention, but a very different level than AI offers. Without going into massive detail I would point out that CDOs were not the first example of securitisation and crypto was not the first example of a 24/7 market.

acdha 3y ago

E-commerce isn’t on that list for the same reason AI isn’t: it had real value. Cryptocurrency had a ton of money poured into it but never found something it did better than the alternatives - PayPal didn’t even feel the need to pretend to treat customers better! – and magical thinking about security and customer service not mattering didn’t help. Pets.com is harder to understand if you weren’t there: it looks like a not bad idea, which it was, but the business was setup as if it was a lucrative high-margin industry. There was no way it could succeed losing money on every sale, and they locked those costs in so well there was no real way to lose less money unless you could figure out how to make shipping far cheaper (most of what people buy at pet stores is heavy food). The hype was off the charts, though, and people poured money into the stock anyway. I worked with a major competitor and was on a few calls with their ostensible technical staff, and it was palpable how they had hired anyone who could spell HTML and everyone was killing time waiting for their options to vest rather than building anything real. AI is like the early web in that regard: there are plenty of outright fraudulent companies and a bunch of doomed ones who just don’t realize it yet, but there is also a real capability that wiser people will use. As with the early web, I’m expecting a lot of startups to get a ton of attention compared to boring old companies actually doing useful things.

fancyfredbot 3y ago

There was definitely some level of novelty in each case I mention, but a very different level than AI offers. Without going into massive detail I would point out that CDOs were not the first example of securitisation and crypto was not the first example of a 24/7 market.