Three Steps to the Future
ben-evans.com
ben-evans.com
But, amusingly re: your point that it could be done in a centralized way: recently my wife's (sigh) Helium miner stopped working for a few days because... centralized stuff failed!
So as far as I can tell, it's a bit of huckster speculation on top of a centralized protocol. Which, theoretically, might one day compensate random people for putting up 20 watt transmitters to spread ultra-low-bandwidth internet access around (right now we get compensated in VC-funded scrip aimed at expanding the network).
Not terrible, but the crypto bits do seem only lightly glued on top of the basic business proposition.
Right now, I am pretty sure there is nothing to fit that bill. If there is in the future, I definitely want this, if only to be able to play games while lying in bed or to have several virtual computer screens for coding and work.
IMHO however, movement and tactile feedback are still lacking so far behind that I don't even know if they can be solved in a non-invasive, average user friendly manner. I mean sure, you can strap yourself to a treadmill contraption and use controllers or gloves to track your hands and maybe even wear some suit which vibrates here and there. But you won't feel real collisions, let alone texture, realistic force and acceleration. And somehow when I tried VR I felt like this was missing so much that it limits my enjoyment big time.
Sure it is, my MIL doesn't do anything on the internet other than entertainment (TV and music) and maybe email or texting. If she wanted to, she could cut it out completely and get along fine with landline phone calls and mailing bill payments and such.
also, you seem to think projecting more information onto the world is a Good Thing, this is not obvious to me.
Now-a-days I question my immediate reaction to anything.
Parallels with early computing industry? The very nature of technology adoption is such that it always appeals to a niche first aka "no one but a handful want it". Then, depending on incentives, an ecosystem crops up that continues to add value as it grows. Computers had internet to fallback on for network effects, with web and email help add value. For VR, that could be gaming, crypto-currencies, social networks... who knows what else.
I haven’t been able to tell what it is?
From what I've read, the main idea is decentralization, either of website databases using blockchain (which seems wildly impractical and ludicrously expensive), or of payment of advertising revenue to website users and contributors (which seems like an impossible business model)
The main issue, I think, is that "decentralization" means a different thing to every person, therefore "web3" can mean a different thing to everyone. This makes it hard to define it or to argue for it or against it.
I read a blog post linked on HN recently on "the architecture of a web3 application" [1] and it was so long and complicated that I honestly thought it was satire. I don't think the methods described in the post were at all practical.
Ultimately, web3 might simply be a rebranding of blockchain and crypto technologies in order to make them seem more democratic and transformational than perhaps they are.
[1] https://www.preethikasireddy.com/post/the-architecture-of-a-...
Tell you what, as full of scams and cons as the world of crypto is, i still sympathize with it much more than I do with any random argument in favor of a real monstrosity of dishonesty and prying scumminess like Facebook or today's Google. Those and so many of their modern digital cousins are the real enemies of so many good things.
Web 3.0 didn’t exist, so it was necessary to invent it to give cryptocurrencies value again.
Web3::web 2.0 IPFS::filesystems
The most interesting part I find is not the 3D VR element, that's like saying the most compelling part of Web 2 was CSS and HTML5.
It's the distributed nature of data and data ownership. Where Web2 was all about user created content -> corporate owned content, in web3 you own the content you create.
The benefit of this is that the platform that you use your data on, is just a platform. You can easily move your data to another platform for other purposes.
The example I use to compare to "real world" stuff is your fitness. You pay a monthly membership to go to the gym and stay fit. When you want to go on a ski holiday, you don't have to go to your gym and tell them you want to go on a ski holiday and ask if you can use your fitness at the resort, and they can say "oh, we don't have an agreement with that ski field, or any ski fields, so sorry, your fitness stays with us. But we've been selling your data to a bike tour operator, so they know all about you, and you can go use your fitness with them...but only with the bike tour operators we've partnered with and have a commercial agreement with...if it isn't them, then...sorry, you can't use your fitness.
In a Web3 world, your data lives in your wallet, and you should be able to decide who gets access to it and how they use it, and revoke it from those you don't want access to. We're looking at adding this capability to Ayvri. We did an integration with Strava, but our users want an integration with Garmin, Suunto, every little paragliding device you've never heard of, etc.
A friend that runs a fintech company was describing the process people go through when looking for a mortgage. They go to a banks website, give their details to the bank, see the offers on mortgages, then go to the next bank or lender, and do the same again and again. If your data is in your wallet, you should be able to just go - here's my data, do your thing.
Sure there's also all the stuff about de-platforming, and there are going to be issues with propaganda, scams, etc. But will it be any worse than what we have today?
Web3 is normal web development where you use some APIs to interact with these protocols and condition access to {functionality,resources,..} in your webapp on the current state of the multiparty computation. You can also participate in coordinating the computation.. idk maybe I'm taking the 2007 constraint a bit too far.
1/ New wave of applications - imagine 3d, AR, VR - new interfaces and possibilities.
2/ When you have these new wave of applications, these applications are going to do something (transactions). What if the transactions did not have to go through a bank or Google or Amazon.
3/ Today, you and your data are the product. What if you could control who gets access to your data.
i.e. makes web sites harder to switch off/censor.
p.s. Freenet has been doing this since Satoshi was in diapers.
Ugh. Is that all we get? Not even self-driving cars?
The big trend in cryptocurrencies is crackdowns. In June 2021, China dropped the hammer on crypto mining. They started with "please stop doing that" in 2018-2019, and progressed to turning off the power to hundreds of Bitcoin mining operations and arresting people. This stopped the big use case of Bitcoin in China - evading exchange controls on capital. In the US, we're seeing much more involvement by the IRS, SEC, FBI, and New York State authorities. The US government seized US$3.3 billion in cryptocurrency in 2020. Cryptocurrency operations are being forced into either bank-like regulation (UK, Canada) or pushed to the usual regulatory havens (the Cayman Islands is popular). The SEC stopped most ICOs back in 2018. Currently being looked at hard, stablecoins. Next, NFTs. Most legit countries are now requiring Know Your Customer identification, and many of those countries share that data between tax authorities. The legit use cases of cryptocurrencies being somewhat absent, this doesn't look like a transformative technology.
2019 was the year of VR. Like 3D TV, it works, but it's just not that interesting. AR, though... For that, see the "Hyperreality" video I've mentioned before. That's where Facebook probably wants to go.
Much as I like virtual worlds, I don't see them as being a big thing. 3D social worlds go all the way back to Quantum Link, which predates AOL, and continue through to Second Life, which is sluggish but has all the features people claim metaverses should have. They are only useful to people with too much free time.
The technology most likely to change the world in the next decade is vaccine engineering. Not just COVID. Vaccines for AIDS, tuberculosis, all variants of influenza, and Alzheimer's are in development. Some of them will probably work. Creating new vaccines is now done by sequencing and then synthesizing a candidate vaccine. There's much less trial and error involved.
Batteries are going to get cheaper from production economies alone. This will quietly change the whole energy landscape.
And we're going to get self-driving cars. Waymo and Cruise test in San Francisco routinely now, and they have few accidents. Usually they get rear-ended at slow speed because they're cautious about left turns. That will get better as most manual cars get auto-braking.
Indeed, and even better, I think this will finally be the decade of grid storage. There are something like a dozen pilot projects for all kinds of theoretically scalable grid storage solutions being built and tested right now. This will make or break renewables as a viable source for baseload power.
It will also see the revival nuclear because grid storage can't be built cast enough and many governments don't want to invest in the infrastructure improvements that will be needed.
A 3x overbuild of solar+wind in an optimal mix, along with a continental grid and 3 hours worth of batteries is all you need for 99.99% reliability: https://www.nature.com/articles/s41467-021-26355-z
> It will also see the revival nuclear because grid storage can't be built cast enough and many governments don't want to invest in the infrastructure improvements that will be needed.
I don't understand this point -- if governments don't want to invest in infrastructure, why would they invest in nuclear which is an order of magnitude more expensive?
Cheap enough at current production levels is not necessarily cheap enough to meet the capacity needed for widespread grid storage. Lithium extraction is pretty dirty right now too.
> A 3x overbuild of solar+wind in an optimal mix, along with a continental grid and 3 hours worth of batteries is all you need for 99.99% reliability
I think that's overstating those optimistic results.
If you also account for the end to end/total carbon footprint, some calculations show nuclear emits less carbon than wind and solar. Consider that a wind farm producing as much energy as a nuclear plant requires an order of magnitude more concrete, and concrete is a large carbon emitter.
> if governments don't want to invest in infrastructure, why would they invest in nuclear which is an order of magnitude more expensive?
It's not an order of magnitude more expensive than restructuring the whole grid. Arguably nuclear is more cost efficient overall: