Don't conflate decentralization with blockchain.
I'm not convinced that Bitcoin or currencies are a practical application that results in net gain (economic or otherwise) to humanity. The internet was obvious from day 1, even if the actual end applications ended up quite a bit different.
> how can you not see the utility in an open firehose database/API without gate keepers?
This is what I don't see. The "gate keepers" are often a positive, not a negative. And it's hard for me to imagine what sort of database is being held back because of gatekeepers.
I want to build apps without having to worry about ever being restricted, limited or shut down.
For example the YT dislikes going away. What if you got people to install a chrome extension where it tracked urls with upvotes/downvotes (game system). The database is shared between installed apps no central db... would it work?
I could be far off with the intent of the application but another example is mesh networking where it is nice to not have a central pipe but is it going to be as good, particularly the case of cross continent.
So. Where's blockchain-powered Twitters and Facebooks then? For all the talk about gatekeeping and decentralization no actual apps appeared that would in any way support your words.
You seem to be insinuating something so please elaborate.
You're the one proclaiming and wanting things.
> No-one has built it yet? It's not technically possible? There's no demand for it?
Yes, yes, and yes, to varying degrees? You're so close to getting it.
My point holds for non crypto decentralization as well. Consumers choose against, no matter how much tech people like us want it.
I'm still not clear how this would work.
In a web 3 twitter clone that prevents one from ever being "restricted, limited or shut down", where does the crypto-tweet live?
Where is it stored?
How well does this scale?
What does it take to crypto-tweet (do you have to pay some crypto currency to have it be out there)?
What happens if you want to redact it later?
What is the energy cost of putting a crypto-tweet out there?
What would be the overall energy and storage costs be across all devices for a single crypto-tweet?
What would the costs be if this was to scale to 1% of Twitter (Twitter is 6k tweets/s so this would be 60 crypto-tweets/second)?
If the above doesn't reflect the proper design of a crypto-twitter, what would that look like? One of the classic system design questions is "how would you implement twitter" and so that's something that nearly every software developer has thought about at one point or another... and so if I've got the wrong questions fro crypto-twitter, how is it designed and what are the right questions to ask about such a design?
TBD
And yes, those people also make a distinction between themselves and the average grifter - something HN fails miserably at.
I am becoming convinced that there may be value somewhere in this open database, but I'm also convinced that 99% of the people advocating for web3 and making a ton of "money" on current token speculation do not know what this value will be.
The most convincing application I can think of is a PGP for the modern world that everyone can use without hours and hours of tech training. But I think there will be more.
Application data...
Let's say there's a Great Open Database of microblogging. This is somehow distributed across multiple systems, and sized so that it can handle the load, and provides a query language to make that database useful... Shoot, let's spin up a couple clusters of CouchDB and go from there and build a microblogging platform on top of that?
Does it solve the problem? It's trivial for anyone to spin up their own clone of a CouchDB - just replicate it. You want your own uncenersoable copy - its yours. And yes, there are numerous problems with this that as you pull the thread get to the point of "at some point, you're going to have to trust some central authority"
That might be where you're getting to the "PGP for the modern world" - though I'm going to contend that every solution has a similar problem of getting the average person to treat their keys securely... and that means having one copy and getting ransomeware or having key stealing malware has become much, much more problematic than it is in today's world. I believe that that social / computer literacy problem needs to be solved for the average person before one can reasonably move forward with PGP for the modern world.
I'm thinking broadly, but the idea of exchanging keys on a distributed system and then engaging in services using those keys peer-to-peer is appealing. You can audit all the software, and you use zero centralized third parties. I think it brings back a DIY attitude to the internet.
If you're tempted to lecture me on your thoughts about what 'normal people' want and how its not what I just said, I would pre-emptivly argue you do not consider that normal people often co-opt what the inside few value first, and that if a system like this remains niche I would still value it. Web3 on this site is too much about dunking on grifters and too little about focusing on what the grifters are stealing.
Why would I as a user prefer Twitter or prefer web3 decentralized microblogging?
With Twitter, even with minimal gate keeping, I still get a spam/scam message only once every month or so.
But minimal gatekeeping is soooooo much better than email, where I get so much spam email that it has made the service worthless without spam filtering, which has an error rate for finding spam.I literally only read my personal email messages once every week or so because it's such a chore to get through all the craft spam, much of it from places where I once bought a single item.
So as a programmer and creator, sure, you prefer the ability to not be locked out. But as a user of Twitter, I vastly prefer all those programmers being locked from access. It's the only thing that keeps it sane and almost-useful.
One application I recently heard of that might be relevant: using wallets as crypto for the average person, finally allowing more secure and better website login. If everybody has wallet, it might finally solve the problem of scattered passwords everywhere for every site, with many users reusing site password.
I'm even going to ask "why do you need web3 for decentralized microblogging?"
What problem does this future tech solve that Mastodon does not currently solve? What is the current solution lacking and is easier to solve in a hypothetical web3 world?
--
The "use wallet for login" I contend has an even bigger problem than "reuse the same password" for the average person.
Losing a device with a wallet (e.g. phone) becomes an even bigger issue.
Wallet stealing scripts/malware would compromise even more of a person's identity.
Additional difficulties in being able to recover an account if a person loses access to their wallet (e.g. drive failure).
I believe that existing password management systems do a better job of this for the average person and it would be better to have people use those than trying to get everyone to have a crypto wallet.
That has the added advantage that password management systems work wherever there's a password rather than a mishmash of "this site uses this wallet" and some sites will never get to the point of using a wallet for account management (compare how many have a 'login with apple' even though that's been around for a while and is pushed by Big Tech).
One of the difficulties that I have is that the decentralization proposals don't seem to offer anything over the existing solutions out there aside from the "I want to build it again."
Don't get me wrong - building it again is a legitimate reason. There are several services out there that I'd like to build again... though I also want to do it because I feel that there's a problem that the existing solution doesn't properly solve or has a less than optimal underlying model for my use case.
So, I return to the question of "what is this going to do better?" This can help lead to feature discovery of existing platforms (the "wait, it does that too?"), the addition of missing requirements to those existing platforms, or the beginnings of a paradigm shift for how one works with such services.
I want to know what's missing so that I can work on the making it better. I don't see what is currently missing in the current (distributed) solutions.
I'm not at all convinced that technology from "web3" or ethereum or whatever can actually do this.
Please don't tell me it's coming soon or just a matter of time. That's been said for ten years.
What is stopping you from storing HTML/CSS on torrent/IPFS/github/pastebin?
This stuff exists - right now. Your version of it just adds an unnecessary barrier to entry.
Well it’d be easier if crypto wasn’t every second word out of someone’s mouth when they talk about “web3”. The technology to do pretty much everything “web3” promises already exists, and works better than it’s crypto-equivalent. The Local-First software movement is a good example of this.
If the proponents of “web3”actually cared about decentralisation/federation/user controlled data/etc they’d be pushing *actual solutions” instead of recruiting for yet another crypto Ponzi scheme.
Worth noting, however, that I’m not saying there aren’t improvements to be made, but I have yet to see any demonstration that the crypto space can contribute anything actually meaningful to this space.
Let's start you off with the very basics; a low stakes example. What is a solution to email spam? You incur a small cost to honest senders and a large cost to mass senders. We'll call it Hashcash[2], and as a user you will only accept emails from senders with a header including your email and the date, which hashes to a sufficiently small value of your choosing, the smaller, the harder it was to send. Every sender gets a linear increase in amount of work to send, but for normal email users sending one email at a time its a small constant, while mass senders increase this value for each recipient. Because of the economics of spam email, where the profit is made off a few gullible recipients while the majority ignore it, Hashcash makes spam emails a losing strategy, at least in this simple example. This is an economic solution, and it is also exactly what the primary cryptocurrency does.
Bitcoin is primitive, and its data throughput reflects that. I won't be shilling or giving you tips, but imagine a blockchain system which ensures openess (no Infura Problem[3] as is common in Ethereum sized blockchains) and a scalable model for pricing and putting data on chain. It may be more expensive or slower than a federated system, but a system with the proper economic incentives and deterrents is now capable of doing something a centralized or federated system cannot: hosting data for which there is massive economic incentive to distort without a profitable way to actually do so, or more specifically, a predictable cost of attack that rises with subsequent block confirmations (just like Bitcoin's security model). Now you can price your data, and wait the amount of time that correlates to the cost of attack that would not be worth the money it takes to cheat.
[1] https://en.wikipedia.org/wiki/Dot-com_bubble
[2] https://en.wikipedia.org/wiki/Hashcash
[3] https://cryptobriefing.com/infura-outage-sparks-debate-over-...
At the end of the day, I guess I trust the goverment I vote for and the banks I decide on more than a distributed group of strangers. Do you have an example of a DB that was unfairly limited by a gate keeper?
"Just build your own Twitter!"
> Conservative Twitter gets kicked off hosting services.
"Just build your own hosting services!"
> People talk about the importance of Web3 for openness and censorship resistance.
"Web3 is a grift, and if its not its for illegal things then you're only getting marginal value as compared to using FB or Twitter! So just build your own Twitter!"
I'm getting dizzy.
This is a pretty ignorant take on how blockchains work. Feel free to have your opinions, but if you really think that billions of dollars is secured in Bitcoin simply because "a distributed group of strangers" are trustworthy, you've missed the entire point and you don't actually understand how it works. "Trustless" is a 'buzz word' you might want to put a lot more respect on - and maybe first learn what it actually means.
Putting that aside, let's deal with the original quoted issue. I am privileged enough to be in a place where I trust my government, and by extension agree with their rules. The "trustless" setup of Bitcoin has meant that CP can be hosted comfortably within it (https://www.bbc.com/news/technology-47130268). A system with a liable entity would be able to solve this, Bitcoin cannot. I don't feel ignorant to think this is an issue?
By the way, Bitcoin does not have that problem because the story you linked is about a wild fork of Bitcoin which is unsustainable anyways - not even counting the fact that BSV is not intended as a file hosting service. Perhaps see what IPFS or other file hosting coins and protocols do to stop that problem. And as hard as it is to swallow, there are plenty of other ways to obscure the uploading of illegal data across all sorts of centralized services - but we don't just get rid of the internet.
Where are the applications?
Unmoderated discussion doesn't scale. The noise floor rises faster than the population of the service until it collapses under the weight of the trolls.
So maybe you decided to make public filter lists, but then the people making the filters end up being the defacto operators of the platform and you're back to square 1.
I don't think we would be back to square 1. We would still have the possibility to create new clients and build whatever applications we want without being under the mercy of Big Tech holding the data hostage. Sure, maybe one such app talking to the open platform would become dominant and the de facto operator but at least it wouldn't have the power to restrict, limit or shut down other apps.
How does this improve on a self hosted blog with RSS feeds and associated aggregators?
"Everything will be magically good and people will live in love in peace once there's a decentralized open database with bo gatekeepers" is just as valid a statement if you "don't say anything about implementation".
Now, can you stop being obnoxious?
My point is: You keep saying things without offering anything concrete. When people ask you concrete questions you hide behind "I never said aything about implementation"
> I (me, personally) want to
I also want many things.
> I also want many things.
And how would you feel if people started arguing saying you don't actually want those things? That's what you are doing.
Its quite obvious that user services would form their own communities around their moderation styles, and its not like all data needs to be posted on chain anyways.
We don't need blockchain for decentralization. That's what we're dismissing. Blockchain is a set of cancer cells that have metastasized into the existing zeitgeist surrounding decentralization efforts. It offers nothing, and just poisons our existing work.
Decentralization doesn't have inherent value, and has a whole host of problems that cryptopeddlers don't want to either acknowledge or pretend to solve.
I think that when building the internet and the web, decentralization was assumed to have a lot of value, do you disagree with that?
It wasn't inherent.
The value was "in the case of global nuclear war it would be nice to have independent centers that could still continue to operate in the aftermath". It was a solution to a very specific (though thankfully theoretical) problem.
"web3" offers no such problem, but loves to scream "but decentralization!".