5 ms·

Toys, Secrets, and Cycles: Lessons from the 2000s

33 points by revorad 4y ago

Juliate 4y ago
Overall good observation, but... > You just need to know where to look. My belief is that the best place to look is crypto and web 3. I've been around the web2 emergence and seen the meetups/evening/weekend reunions around the excitement. It felt and looked like toys, and the tech was dead easy to comprehend and extend at the time. Neither crypto or web3 seem to fit in this picture though (overly complex to explain, to implement, no fun factor, and obnoxious people promoting it). On the other hand, Dall-e kind of AI tools (recreation from textual prompts and/or visual/audio hints) do fit: progress is apparent, people play with it for fun and discovery. Ad-blocking and privacy techs, also, are still in a toy-like/diy stage but seem to become more crucial. Human-scale centered stuff (better housing, cities, transporation, care, slow-everything, energy-effiency, etc.) also seem to be in the toy-becoming-crucial stage.
- eternityforest 4y ago
  As far as I can tell, what the smartest people do on weekends now is pure analog/nontech, or toy/throwaway/education stuff, learning new programming languages, or very small personal use software(I know someone who made their own VCS!). Most engineers thing "Unnecessary privacy risk" or "Boring bloatware" when they see the stuff that would have been big in 2003. In 10 years, I doubt I'll be using anything that the smartest people are playing with now, the smartest people seem to be more interested in ideas and logic than products and projects. If there's no real novel idea in it(Or at least new to them), they'll get bored. A lot of stuff is possible and users would love it, but... engineers have no interest, because their idea of tech progress is less software not more. Human centered stuff is basically impossible to change on a DIY scale, so the transition from toy to practical is a lot harder. A city is inherently almost unchangeable except via million dollar contracts, and any kind of commune/ecovillage/etc has the overhead of trying to do its mission, while also existing within the current economy, so everyone seems to do a lot of work and make a lot of sacrifices as far as I can see from the outside. There's still smart people playing with stuff that will be everywhere(Like IoT), but not as many. Or, perhaps it still works that way, and in 10 years the world will find a deep respect for Kackzynski, and my next computer purchase will be a black market laptop when everyone votes to ban them and be done with it!
- gwern 4y ago
  I agree. There was a time when cryptocurrency gave me the vibes which Dixon describes: weirdo cutting-edge hackers on the weekend screwing with this strange new toy - it was 2013 or so for Bitcoin, then 3 years later for Ethereum. That was an era where '$30M' exits were shocking and flattering. Today in 2022, $30m isn't even the losses in bankruptcy to a crypto thing you've never heard of up to its gills in suits (both human and law-). Somewhere around the $100,000m marketcap, the cute-and-fluffy aspect of it all kinda wore off. Meanwhile, there is another area today where $30m is a surprising exit for a weekend project taking off, where hackers eagerly debate on Twitter and show off the latest baffling thing they've just coaxed a new tech into doing and wonder what it could do next, and draw lines on log graphs... And that is indeed AI.
kkielhofner 4y ago
I’ve seen these takes and I can never get a simple answer to a simple question: “Where are the users?” By the 2001 timeframe he’s describing the internet had roughly 500 million users worldwide. I think his descriptions of how most people used and viewed the internet are revisionist history and self serving for criticisms such as the one I’m starting on but the fact remains - half a billion people with a daily average of 30 min. Let’s look at what it took to get the first half billion on line (starting from 1988): - You had to buy a computer or device that was capable of accessing the internet. Let’s be very conservative and say this is $1000 (non inflation adjusted). You spent this money knowing the computer would be obsolete and nearly worthless in 1-2 years. - In many cases dialup was long distance and pay by the minute by the telephone company and ISP. I know it was for me and my family until about 1996 or so. We used it 30 minutes a day because it was double billed by the minute and ridiculously expensive. - Dial up providers had to deploy massive hardware infrastructure - hundreds of thousands if not millions of modem banks to telephone exchanges throughout the entire country. - “Backbone providers” had to dig trenches, lay fiber, and dispatch ships to lay fiber across oceans. - Telephone companies and a variety of other infrastructure providers had to upgrade, repair, etc untold miles of existing copper infrastructure, switches, etc, etc The first transatlantic telephone cable to use fiber optics was activated in 1988. 1988 to the 2001 period he describes is 13 years. Bitcoin was released 13 years ago. From an exhaustive analysis I’ve done over the top 20 Coindesk blockchains at best there are MAYBE 100 million users of anything blockchain related worldwide. As cdixon mentions in the article in 2009 we had the iPhone and App Store. We had Android. The internet had roughly 2 billion users. Ubiquitous broadband, social media, cellular data, etc, etc. Getting a user on to the internet (and supporting them) in the 13 year period between 1988 and 2001 was a tremendous challenge. Getting a bitcoin/blockchain user online (and supporting them) for its entire existence is downloading an application on your phone or computer. For the bast majority of blockchains history I could make someone a bitcoin user in five minutes or less in a bar. Or even simpler - they could click a link on a social media post from one of their 500 connections. Remember the $1000 guaranteed to depreciate computer to get on the internet from the 90s? With bitcoin and other cryptocurrencies the other half of the so called value prop is that it’s virtually guaranteed to “moon” and make you 100x (or whatever). Yet blockchain user adoption has grown at 1/5 the rate of the web/internet - and this is being very generous to blockchain. People saw value and utility in the internet and the rate of user adoption reflects that. I can’t get anyone to answer my simple question: “Where are the users?” As we’ve seen with every technology invention of all time if it provides value, utility, quality of life improvements, etc people figure it out and will adopt it - even if it involves jumping through the kinds of hoops, expense, etc we did in the 90s for the internet. That clearly and plainly isn’t happening with anything related to blockchain. Why?
ramesh31 4y ago
It always seems like people are writing these pieces to convince themselves of something. You never see one from an uninvested 3rd party.
- thesausageking 4y ago
  Every VC blogpost is done for marketing. They sell a commodity product (money) so, just like Coke or Tide detergent, they need to differentiate based on brand.
thesausageking 4y ago
It's interesting to read Chris Dixon on Hackernews. In the Bitcoin world, he and a16z are regarded as so toxic that another investor was recently cancelled for just co-investing with them. After Filecoin, ICP, BitClout, Basis, ... and so many other scams that blew up, how can you take anything they say seriously? One example: Chris wrote blog posts for[0] and promoted the "Internet Computer" project everywhere. It launched at >$400, insiders sold, and then it quickly collapsed to under $50 and is now at $5.[1] This was promoted by Chris and other investors, bought by retail (ie, normal people) who thought they were getting in on the ground floor of a big new exciting technology, and then lost almost their whole investment. Web3/crypto should not be about billionaires making money by pumping and dumping useless tokens on retail investors. [0] https://a16z.com/2018/08/29/dfinity/ https://a16z.com/2018/08/29/dfinity/ [1] https://www.coingecko.com/en/coins/internet-computer https://www.coingecko.com/en/coins/internet-computer
- ilrwbwrkhv 4y ago
  A16z has lost all their credibility with the massive crypto pump and dump.
  - funstuff007 4y ago
    I guess, but without crypto I think their overall returns may be close to flat or even negative after fees. And now crypto...

33 points by revorad 4y ago

Overall good observation, but... > You just need to know where to look. My belief is that the best place to look is crypto and web 3. I've been around the web2 emergence and seen the meetups/evening/weekend reunions around the excitement. It felt and looked like toys, and the tech was dead easy to comprehend and extend at the time. Neither crypto or web3 seem to fit in this picture though (overly complex to explain, to implement, no fun factor, and obnoxious people promoting it). On the other hand, Dall-e kind of AI tools (recreation from textual prompts and/or visual/audio hints) do fit: progress is apparent, people play with it for fun and discovery. Ad-blocking and privacy techs, also, are still in a toy-like/diy stage but seem to become more crucial. Human-scale centered stuff (better housing, cities, transporation, care, slow-everything, energy-effiency, etc.) also seem to be in the toy-becoming-crucial stage.

eternityforest 4y ago
As far as I can tell, what the smartest people do on weekends now is pure analog/nontech, or toy/throwaway/education stuff, learning new programming languages, or very small personal use software(I know someone who made their own VCS!). Most engineers thing "Unnecessary privacy risk" or "Boring bloatware" when they see the stuff that would have been big in 2003. In 10 years, I doubt I'll be using anything that the smartest people are playing with now, the smartest people seem to be more interested in ideas and logic than products and projects. If there's no real novel idea in it(Or at least new to them), they'll get bored. A lot of stuff is possible and users would love it, but... engineers have no interest, because their idea of tech progress is less software not more. Human centered stuff is basically impossible to change on a DIY scale, so the transition from toy to practical is a lot harder. A city is inherently almost unchangeable except via million dollar contracts, and any kind of commune/ecovillage/etc has the overhead of trying to do its mission, while also existing within the current economy, so everyone seems to do a lot of work and make a lot of sacrifices as far as I can see from the outside. There's still smart people playing with stuff that will be everywhere(Like IoT), but not as many. Or, perhaps it still works that way, and in 10 years the world will find a deep respect for Kackzynski, and my next computer purchase will be a black market laptop when everyone votes to ban them and be done with it!
gwern 4y ago
I agree. There was a time when cryptocurrency gave me the vibes which Dixon describes: weirdo cutting-edge hackers on the weekend screwing with this strange new toy - it was 2013 or so for Bitcoin, then 3 years later for Ethereum. That was an era where '$30M' exits were shocking and flattering. Today in 2022, $30m isn't even the losses in bankruptcy to a crypto thing you've never heard of up to its gills in suits (both human and law-). Somewhere around the $100,000m marketcap, the cute-and-fluffy aspect of it all kinda wore off. Meanwhile, there is another area today where $30m is a surprising exit for a weekend project taking off, where hackers eagerly debate on Twitter and show off the latest baffling thing they've just coaxed a new tech into doing and wonder what it could do next, and draw lines on log graphs... And that is indeed AI.

kkielhofner 4y ago

I’ve seen these takes and I can never get a simple answer to a simple question: “Where are the users?” By the 2001 timeframe he’s describing the internet had roughly 500 million users worldwide. I think his descriptions of how most people used and viewed the internet are revisionist history and self serving for criticisms such as the one I’m starting on but the fact remains - half a billion people with a daily average of 30 min. Let’s look at what it took to get the first half billion on line (starting from 1988): - You had to buy a computer or device that was capable of accessing the internet. Let’s be very conservative and say this is $1000 (non inflation adjusted). You spent this money knowing the computer would be obsolete and nearly worthless in 1-2 years. - In many cases dialup was long distance and pay by the minute by the telephone company and ISP. I know it was for me and my family until about 1996 or so. We used it 30 minutes a day because it was double billed by the minute and ridiculously expensive. - Dial up providers had to deploy massive hardware infrastructure - hundreds of thousands if not millions of modem banks to telephone exchanges throughout the entire country. - “Backbone providers” had to dig trenches, lay fiber, and dispatch ships to lay fiber across oceans. - Telephone companies and a variety of other infrastructure providers had to upgrade, repair, etc untold miles of existing copper infrastructure, switches, etc, etc The first transatlantic telephone cable to use fiber optics was activated in 1988. 1988 to the 2001 period he describes is 13 years. Bitcoin was released 13 years ago. From an exhaustive analysis I’ve done over the top 20 Coindesk blockchains at best there are MAYBE 100 million users of anything blockchain related worldwide. As cdixon mentions in the article in 2009 we had the iPhone and App Store. We had Android. The internet had roughly 2 billion users. Ubiquitous broadband, social media, cellular data, etc, etc. Getting a user on to the internet (and supporting them) in the 13 year period between 1988 and 2001 was a tremendous challenge. Getting a bitcoin/blockchain user online (and supporting them) for its entire existence is downloading an application on your phone or computer. For the bast majority of blockchains history I could make someone a bitcoin user in five minutes or less in a bar. Or even simpler - they could click a link on a social media post from one of their 500 connections. Remember the $1000 guaranteed to depreciate computer to get on the internet from the 90s? With bitcoin and other cryptocurrencies the other half of the so called value prop is that it’s virtually guaranteed to “moon” and make you 100x (or whatever). Yet blockchain user adoption has grown at 1/5 the rate of the web/internet - and this is being very generous to blockchain. People saw value and utility in the internet and the rate of user adoption reflects that. I can’t get anyone to answer my simple question: “Where are the users?” As we’ve seen with every technology invention of all time if it provides value, utility, quality of life improvements, etc people figure it out and will adopt it - even if it involves jumping through the kinds of hoops, expense, etc we did in the 90s for the internet. That clearly and plainly isn’t happening with anything related to blockchain. Why?

ramesh31 4y ago

It always seems like people are writing these pieces to convince themselves of something. You never see one from an uninvested 3rd party.

thesausageking 4y ago
Every VC blogpost is done for marketing. They sell a commodity product (money) so, just like Coke or Tide detergent, they need to differentiate based on brand.

thesausageking 4y ago

It's interesting to read Chris Dixon on Hackernews. In the Bitcoin world, he and a16z are regarded as so toxic that another investor was recently cancelled for just co-investing with them. After Filecoin, ICP, BitClout, Basis, ... and so many other scams that blew up, how can you take anything they say seriously? One example: Chris wrote blog posts for[0] and promoted the "Internet Computer" project everywhere. It launched at >$400, insiders sold, and then it quickly collapsed to under $50 and is now at $5.[1] This was promoted by Chris and other investors, bought by retail (ie, normal people) who thought they were getting in on the ground floor of a big new exciting technology, and then lost almost their whole investment. Web3/crypto should not be about billionaires making money by pumping and dumping useless tokens on retail investors. [0] https://a16z.com/2018/08/29/dfinity/ https://a16z.com/2018/08/29/dfinity/ [1] https://www.coingecko.com/en/coins/internet-computer https://www.coingecko.com/en/coins/internet-computer

ilrwbwrkhv 4y ago
A16z has lost all their credibility with the massive crypto pump and dump.
- funstuff007 4y ago
  I guess, but without crypto I think their overall returns may be close to flat or even negative after fees. And now crypto...