At least I see a path for this to be true.
There is an interesting argument about micro-transactions. The practical challenge there is getting the price to be enough below the cost of credit card transactions to spur adoption, especially since the processors have plenty of margin to cut since their overhead is so much lower.
The bigger problem is one of market: beyond the questions about restricting sites to people affluent enough to pay for those micro transactions, it’s unclear how many people will actually pay as much as advertisers for most sites. This has been tried a number of times over the years but there have been relatively few successes.
I think urbit’s approach to spam makes sense. Non-zero cost IDs (cheap) in the form of NFTs that grant network access and a pseudonym. This makes spam no longer economical because spammers can be trivially blocked. Pseudonyms also accrue reputation.
It basically moves auth and network complexity into the OS layer under an abstraction which makes application deployment way simpler. No distinction between a user and a server also fix a lot of the other broken things that cause federated systems to centralize or end up DOA to be avoided.
In the ETH world, decentralized state and the ability to incentivize network nodes and development via tokens and programmatic money make it possible to push back against the things that advantage centralization.
The issue is the protocols have to be good enough to solve a lot of this stuff, the old web didn’t do this well for the reasons you suggest and administering Linux servers is way too hard.
I understand the dismissal of technical people because crypto is filled with know-nothing scammers and gamblers, but it’s worth taking a closer look. There’s lots of interesting stuff happening in the space. A lot of people dismissed the web in the 90s too. There can be a bubble and a crash, but still be underlying tech there that will have a big impact.
Yes, I always think of Google in that way. They just powered through the dot com crash by being really useful.
I'm sure there are examples of this in crypto.
It's true that those articles existed but I think there were two key differences: they were written by [generally] old people who weren't at all conversant with the technology and were recognized as such by many people at the time. It was widely recognized that the dotcom stocks were overheated and some of the companies had no chance but at the same time most people were buying PCs and internet connectivity because there were tons of things they _did_ want to do.
That situation is generally reversed today: the people who understand technology are highly skeptical and the people most bullish about cryptocurrency do not show much sign of understanding either the technology or the markets where they're looking for buyers.
The strongest parallel I see now are the cryptocurrency bulls sound exactly like this sales guy I knew 1999-2000 who was convinced that the primary reason to buy a stock was how much the share price was increasing. Sadly, he neglected to cash out before the crash and ended up substantially in the red.
Regarding decentralization, I think it's important to consider how the internet has become less decentralized over time. The situation in the 90s was far more distributed but the centralized solutions have won out in many spaces, and it's not just well-funded rent-seekers but also factors like cost, reliability, support, and often even performance. I'm philosophically inclined towards open standards and distributed protocols but I have to say the experience has generally not lived up to the dream.