It's not a tech reason: how many Ponzi's schemes you have heard on PGP/GNUPG?
That's why blockchains "are popular", it's not a matter of what they are technically, they can be even just a mythical entity, a person, a particles, they are popular because someone have advertise them as "a tool to make money" and than countless humans have immediately responded. Try to talk about some crypto-enthusiast: I'm pretty sure 99% of the will prove to have no idea of what technically the blockchain is, BUT they surely tell you many strategies and judgments on this and that investment opportunity with this and that crypto, normally materialized for them in some remote web-services, of course.
It's the same model that push "anti-radiation, anti-5G magic stickers&co" on the market. People have no idea on them, they have a fear, a desire, a need, so an emotion and they hear that such tools might satisfy such need/desire/emotion, if the entry barrier is cheap they'll try "just in case". It's nothing about the nature or technicality of the tool.
This very business is the old classic Eduard Bernays way to push things to the public: he push cigarettes crafting a new scientific journal who publish easily new real papers, than he spread it a bit in most famous doctors' offices, MDs read real studies on them and like the journal. They also read dummy studies about how healthy is smoking and they blindly trust them. At the same time on the public he manage to get some famous people to smoke in public, he say that smoking was about feminine emancipation etc something that any relevant cohort like to hear. A small summary of his campaigns is here: https://www.apa.org/monitor/2009/12/consumer
Crypto start in the same way: spread to some FLOSS cohort as a new project by a secret hacker, something to pique interest in those who represent the crème of the IT world, a bit later Ponzi's schemes for the general public, in the middle some spread in the business sector depicting how anything can be solved in a snap with the new exiting tech. A cohort see the interest of the others and global interest grow blind. We do not even know if the legend of some aborigines of I don't know which island have invented large rock to mark they financial transaction on is real or just a fabricated legend.
Try another point of view: what's wrong with the web? IMVHO:
- dependence on search engines, witch are so far a big&powerful only game, witch means that those big&powerful decide what we see or not on average;
- no easy entry barrier for those who want to produce contents, not just consume it. We have had and still have various "free hosted platforms", from the classic Geocities to Substack, but nothing individual and easy to access. We can easily publish on ZeroNet but how many know ZeroNet exists?
- lost of focus and software complexity, the web was about valuable contents who do not change much and many ask for, a usenet companion to avoid keep answering same questions all the time; now is mostly a software platform and nothing in the middle clearly define the right place in IT panorama.
Does web3 address any of these? IMVHO no.
What can solve these?
Well, first of all surely on the internet layer IPv6 with a static global per device for all, so we do not have NAT and dynamic IP issues, those who want can get a domain name and being easy to reach, others can still access their devices with pre-recorded static addresses. IP2IP calls are straightforward like filesharing especially with modern connections for those who have them etc. For the web part... A newborn Plan 9, witch means a new universal 9p modern protocol to mount remote resources with proper access level, the same for share files of any kind, so the same to see a streamed video, the same to send messages (just mount the target mailbox and save a file, the email was sent, namespacing ensure you can't read anything) or to read a classic web website, like one with pure CSS3+html5 and no js.
Do you want a newspaper? No need for a specific protocol and a specific server, the newspaper offer various access (free or paid) to relevant documents, you can even read them with your visual style easily, just overwrite the default CSS. No need separate feeds, they just grab the html without the CSS. The entire software stack for a modern web 1.0 will became so simple that we do not need modern WebVM no-one-but-some-giants can create/maintain.
Internet banking? No need from the bank side a crappy porcal with it's costs, just a common file format to talk to the client, relevant auth is part of the underlying system protocol, transactions are just signed text, like part of an XML stream. Customers can have their own favorite banking app with all their banks in one place, with all their history, notes etc integrated. Again a hyper-simplified software stack for much more results than now.
Nothing of the above proposals are in Web3. Nothing else in the end, just a smoky promise of being so powerful without details...