most electric cars are teslas so like a statistical analysis would be mostly correct.
propaganda against electric cars is generally just propaganda against cars in general, for example:
if full self driving becomes a thing you can expect traffic to become MUCH worse as like one minivan for a parent with 3 kids is replaced by four self driving cars that drive each kid around and one for the parent. This could lead to traffic jams at every subburb
I’m just saying the original post is conflating criticisms against a particular implementation of blockchain with all blockchain.
“The wright brothers invented a useless machine they called the airplane. It only held one person and could fly for only a limited duration. It was also extremely dangerous.”
Well that’s very different than anything happening in the thread so yes its very silly.
The old newspapers are full of critiques of the ford model T and conflating them with all cars, most of those criticisms apply only to early cars. I think that’s what you’re getting at here.
But realistically in context the criticisms make sense. I would go as far as to say its more a problem that enthusiasts havent renamed bitcoin as a “cryptoasset” since its not useful as currency and apply “cryptocurrency” only to proof of stake designs.
Or as an analogy, it would be like calling a motorcycle a type of bicycle, which is more or less true, but its so goddamn different in use they had the sense to rename it so normal people would be able to tell them apart.
I think its mostly for most people “crypto” = “cryptocurrency” = “bitcoin” even though there’s some categorical umbrellas there and I don’t think there’s a way to fix it without redefining cryptocurrency to only include currencies and specifically exclude bitcoin and similar coins. Blockchain has a similar problem because its used as a synonym for crypto.
Sure. Sure. Insert your own example of applying the sins of a particular implementation of a technology to a hierarchical supergroup. That was an example I came up with in a few minutes on the shitter, not a deep philosophical argument.
_ is bad, because (specific implementation) has these flaws!
That’s the issue being discussed here. It’s misleading without precision.
Immutable so long as no one party or group owns more than half of the coins on a given blockchain… then the ledger is whatever they say it is and it propagates down because they can manufacture their own “consensus”.
It’s not 51% of the coins, it’s 51% of the computing power on the network. Both of which are virtually impossible in the case of Bitcoin, though not entirely impossible. I just wouldn’t consider a 51% attack even remotely a threat to the network compared to something like government crackdown
You are making my point. Blockchain is not crypto. Blockchain can be useful in private, internal use cases (like a transaction ledger for bank branches) where trust is largely implicit.
Distributed / immutable databases are not solely a feature of blockchain either.
It’s a very interesting thing in a vacuum. Basically any application of it so far (with the possible exception of the original one, if it weren’t just a speculation investment machine at the moment) runs into the problem where it has to interact with reality at some point. And most of the problems Blockchains solve are already solved by a variety of other systems, for less time/currency/hardware investment.
Because it’s an immutable ledger, not just a database. It maintains a history of every previous transaction/entry. Blockchains are used by banks and in the supply chain because it makes backtracing and identifying discrepancies trivial. For things like cryptocurrency, blockchains allow “don’t trust, verify” but for something where you already have trust, they allow “trust but verify”
Cryptographically immutable append only ledgers (aka merkel trees) have existed since at least 1979. A blockchain is different because it has distributed consensus. If your consensus algorithm is trust, then it’s not a fucking blockchain.
A blockchain is nothing more than a data structure. It’s essentially a linked list using the hash of the previous block. Distributed consensus is something blockchains are useful for, but it doesn’t define it
This is a good comment that makes all good points. But I just wanna say let’s stop saying “blockchain” singular and with no preceding article like we’re tech CEOs and it’s some immutable god. They’re blockchains, plural, like any other data structure there can be more than one and there are. eg The blockchain of ethereum is distinct from the blockchain for bitcoin but they are both blockchains.
Valid point! But then how do you refer to the data structure/architecture/model concept? Sometimes we want a concise term (like bittorrent or ActivityPub) for the abstraction
It’s a novel data structure, we can refer to it like we do other data structures: Linked lists, hash tables, primitives. The branded implementation of these things is what we typically make singular: Bitcoin, ethereum, monero (bittorrent, activitypub…)
Bittorrent implements a torrent swarm, activitypub implements a federated social network.
The only alternative to proof of work is proof of stake. And if the world ever ran on proof of stake crypto, it would make today’s wealth inequality look like a Marxist paradise.
What matters is whether or not there’s any significant profit - the reward being bigger than the cost. PoW minimizes this with difficulty adjustment flushing most profits down the toilet. PoS must always have a return competitive with other investments.
Profits should be low, or else it promotes inequality.
You get one asset with worse interest that can be spent (like dollars), and one asset with better interest that can’t be spent (like bonds). The poor hold the former, the rich hold mostly the latter.
If there’s a low barrier to entry then you’re just reinventing POW because someone can run as many nodes as possible with minimal stake in order to maximize returns. Can you give an example of one successful “fixed rewards” Crypto coin?
You are exactly the type of person I’m talking about :\
Crypto is not the only use case for blockchain.
Blockchain can be useful in inherently trusted, closed ecosystems.
Depending on the use case, proof doesn’t even need to be anything more than a valid certificate (not work, not stake)
Consider a bank that develops its own blockchain ledger for internal use only, replacing their branch ledgers (which require daily reconciliation and propagation).
An immutable, distributed ledger has plenty of valid, valuable use cases without looking like crypto.
“Internal use only” blockchain is an oxymoron. If all contributors are trusted entities, then what does it matter if the data is stored in a blockchain vs any other data structure? If anything the amount of extra work to maintain and modify the blockchain in the case of errors just makes it unnecessary.
If that’s the best example of its many “valid, valuable use cases”, then it’s still a pass from me, dawg.
Pretty obvious that its use in the first place was some FOMO executives trying to get in on “blockchain” technology, just like they’re doing with AI and LLMs now. Funny how BTC and ETH both plateaued a year ago, right around the time AI became the new thing.
I’m not going to bother reading the rest, I already wasted too much time arguing with a true believer. GL with the crypto that you say you don’t have.
Lol you’re a judgemental prick. I have zero crypto because that shit is absolutely moronic.
It’s like you talked to someone who supports nuclear power and responded “good luck with the nuclear bombs you definitely don’t have, true believer”
Sounds like you just want to hate on shit you don’t understand. Go on with your bad self though 👍
EDIT: LMAO from the article you linked they even point out that the tech itself isn’t the problem, but the willingness of businesses to invest in the improvements (which is, like, an incredibly common problem in business that does not in any way make the tech bad)
Its plan to digitize global supply chains hasn’t received sufficient cooperation and support to remain viable.
That is the downside of corporate blockchain projects. They need everyone to share a mindset and commit to a long-term plan. Unfortunately, businesses face ever-changing conditions and financial problems. Few can warrant the cost of buying into such systems under the current market conditions.
You probably preach about how nuclear power is terrible, how if it were so great there would be more of it, and people sticking to coal is proof that coal power is better.
But as long as you can have a trusted central authority, you don’t need blockchain without needing to trust all the clients. The central authority can enforce authentication and authorization for what each client can do, combine with proper logging it is already sufficient. Blockchain is only needed if there’s absolutely no central authority (which is not the case for any traditional business as by nature they already have a hierarchy where the top would act as the central authority, therefore any business that implements blockchain internally is just BS-ing).
Then why hasn’t a better blockchain based currency gained any popularity? If they don’t have critical mass then your distinction is meaningless. It turns out there is just zero real world need for an untrusted distributed ledger. Databases and governments solve the problem much better.
Questioning the technical virtues of an alternative product based on lack of critical mass adoption is pretty funny, when you consider we’re on the fediverse. I know that doesn’t defray your argument, but just an amusing observation.
I see why you might draw the comparison, but I actually don’t think the comparison is valid at all. Forums/communities can still be useful and fun with only a few people. Discord is also massively popular with a small community model, for a more successful example to compare with the fediverse. However a currency that nobody uses or accepts is entirely useless until mass adoption happens. That’s why they typically get mandated by force by governments.
Etherium and virtually the whole rest of the crypto scene that is “not bitcoin” has pretty soundly rejected the wasteful Bitcoin design. There was even a fork of Bitcoin that would have used the much more efficient proof-of-stake, but since that would be bad for everyone with a proof-of-work “mining” rig it didn’t take over.
It turns out there is just zero real world need for an untrusted distributed ledger
An “untrusted distributed ledger” is literally the backbone of modern software development. While you could plausibly split hairs and assert that git requires “trust”, I don’t think you’d wind up in a spot that both supports your assertion and a cognizable difference for anyone but mathematicians and security nerds. (And even if you did, the exact same sort of non-scam usages of blockchains are ones that operate like git, with the ledger used for something else.)
People bring up crypto because it is the only use of blockchain that isn’t worse than already established methods. And crypto is only “better” because it’s unregulated and allowed a bunch of scams to be pulled.
If [other applications of the blockchain, which has now existed for an extremely long time] don’t have critical mass then your distinction is meaningless.
I’m not convinced there’s any internal use for blockchain. Internal implies under a specific umbrella, some overarching organisation, who can then be the central trusted server that makes blockchain unnecesary.
That said, non-public but open uses, such as tracking dealings between companies in markets with little trust and no single governments (the shipping example in your referenced comment) is indeed the thin slither of a plausible use-case.
Another limitation is that blockchain loses its benefits if anyone tries to design over the complexity of using it directly (using a ui that under the hood uses blockchain is no different to using a ui that talks to a central database, you’re trusting the central ui provider, you need to (at least be able to) build your own interface to realise the benefits of blockchain.
That means blockchain basically will never benefit individuals, it can’t. Sure, you could have multiple compatible uis shared around, but that’s no different security-wise to multiple central banks with an interoperable transfer system.
The only place blockchain has real benefits is when multiple large corporations/governments are interacting and don’t trust eachother/anyone.
See the link in my other replies for some examples of internal uses that still benefit from immutable, distributed ledgers.
Large organizations still have loss and risk from individual bad actors. Operating a central authority that validates every single transaction in a ledger, and validates ledger history and consistency, can be prohibitively complicated. A well designed blockchain implementation can resolve most of these issues.
A great example is a pharma/healthcare company that wants to manage medicine batch and expiration tracking, as well as distribution/patient assignment. With a traditional infrastructure every participant needs to phone home to a central authority. In a blockchain setup, peers can report ledger events one hop up and propagate it through the chain.
That’s a very simple example but I hope it gets my point across
Cryptocurrency development makes a whole bunch of arbitrary value-guided decisions during creation, all of these decisions have tradeoffs such that nobody has figured out a way to feature them all at the same time, or would they want to.
For example, bitcoin is fully auditable. Anyone with a copy of the bitcoin blockchain can review every single transaction in bitcoin’s history, and trace the flow of every last satoshi from it’s mining to today. This is because the developers of bitcoin place a high value on verifiable auditability and security. Conversely monero was developed for the purpose of being a completely untraceable, unauditable currency that still has a knowable supply. And ethereum was created in a manner that intentionally supported scripting, so that it could be used as a platform for novel applications and contracts. None of these primary features could be ported to either of the other two without breaking them completely, because of the deep programmatic implications of the requirements.
It’s not really a question of better or worse, but of use case. The fact of the matter is that the reason these three examples are the leading currencies for their use case is literally because nobody has yet been able to do a better job. And for bitcoin at least, at this point it’s security rests just as much in it’s wide adoption and interest as it’s design intent, so it’s unlikely that anyone ever will.
sigh
Once again:
Blockchain is not synonymous with cryptomining
Blockchain does not require proof of work
Cryptocurrency and NFT grifting does not devalue blockchain as an immutable distributed ledger
I swear to god people just copy paste whatever makes them feel good without any effort at understanding
True… But Satoshi did invent Bitcoin, which is proof of work, and is everything in OP
The “technically true” nature of it reads like propaganda against electric cars as a whole, does it not? I’d argue that applies here too.
“Popularized” the electric car, maybe?
There were electric cars since there were cars
Popularized it is. I was careful to not put invent, but morning brain not work good.
Fair enough, I forgot that OP said “blockchain,” instead of “bitcoin.”
most electric cars are teslas so like a statistical analysis would be mostly correct.
propaganda against electric cars is generally just propaganda against cars in general, for example:
if full self driving becomes a thing you can expect traffic to become MUCH worse as like one minivan for a parent with 3 kids is replaced by four self driving cars that drive each kid around and one for the parent. This could lead to traffic jams at every subburb
Doesn’t have to be cars.
I’m just saying the original post is conflating criticisms against a particular implementation of blockchain with all blockchain.
That’d be a silly sentence, would it not?
Well that’s very different than anything happening in the thread so yes its very silly.
The old newspapers are full of critiques of the ford model T and conflating them with all cars, most of those criticisms apply only to early cars. I think that’s what you’re getting at here.
But realistically in context the criticisms make sense. I would go as far as to say its more a problem that enthusiasts havent renamed bitcoin as a “cryptoasset” since its not useful as currency and apply “cryptocurrency” only to proof of stake designs.
Or as an analogy, it would be like calling a motorcycle a type of bicycle, which is more or less true, but its so goddamn different in use they had the sense to rename it so normal people would be able to tell them apart.
What are the differences that you see?
I think its mostly for most people “crypto” = “cryptocurrency” = “bitcoin” even though there’s some categorical umbrellas there and I don’t think there’s a way to fix it without redefining cryptocurrency to only include currencies and specifically exclude bitcoin and similar coins. Blockchain has a similar problem because its used as a synonym for crypto.
Your example is not even close to technically true.
Sure. Sure. Insert your own example of applying the sins of a particular implementation of a technology to a hierarchical supergroup. That was an example I came up with in a few minutes on the shitter, not a deep philosophical argument.
_ is bad, because (specific implementation) has these flaws!
That’s the issue being discussed here. It’s misleading without precision.
Except it says blockchain in the toot?Immutable so long as no one party or group owns more than half of the coins on a given blockchain… then the ledger is whatever they say it is and it propagates down because they can manufacture their own “consensus”.
https://www.investopedia.com/terms/1/51-attack.asp
and most use cases around things like “smart contracts” end up still requiring a trusted third party at some point
https://pluralistic.net/2022/01/30/the-inevitability-of-trusted-third-parties/
It’s not 51% of the coins, it’s 51% of the computing power on the network. Both of which are virtually impossible in the case of Bitcoin, though not entirely impossible. I just wouldn’t consider a 51% attack even remotely a threat to the network compared to something like government crackdown
You are making my point. Blockchain is not crypto. Blockchain can be useful in private, internal use cases (like a transaction ledger for bank branches) where trust is largely implicit.
If you have trust, why do you need a blockchain?
Distributed / immutable databases are not solely a feature of blockchain either.
It’s a very interesting thing in a vacuum. Basically any application of it so far (with the possible exception of the original one, if it weren’t just a speculation investment machine at the moment) runs into the problem where it has to interact with reality at some point. And most of the problems Blockchains solve are already solved by a variety of other systems, for less time/currency/hardware investment.
Because it’s an immutable ledger, not just a database. It maintains a history of every previous transaction/entry. Blockchains are used by banks and in the supply chain because it makes backtracing and identifying discrepancies trivial. For things like cryptocurrency, blockchains allow “don’t trust, verify” but for something where you already have trust, they allow “trust but verify”
Cryptographically immutable append only ledgers (aka merkel trees) have existed since at least 1979. A blockchain is different because it has distributed consensus. If your consensus algorithm is trust, then it’s not a fucking blockchain.
A blockchain is nothing more than a data structure. It’s essentially a linked list using the hash of the previous block. Distributed consensus is something blockchains are useful for, but it doesn’t define it
No, that’s a Merkel tree. Been around since 1979.
Nah, you can pretty easily define a Blockchain as a special case of a Merkel tree with a distributed consensus method.
Here, I provided a bunch of examples months ago of other ways blockchain is being used: https://lemmy.world/post/36683795/19677963
Why do you think LLMs are so popular?
This is a good comment that makes all good points. But I just wanna say let’s stop saying “blockchain” singular and with no preceding article like we’re tech CEOs and it’s some immutable god. They’re blockchains, plural, like any other data structure there can be more than one and there are. eg The blockchain of ethereum is distinct from the blockchain for bitcoin but they are both blockchains.
Valid point! But then how do you refer to the data structure/architecture/model concept? Sometimes we want a concise term (like bittorrent or ActivityPub) for the abstraction
It’s a novel data structure, we can refer to it like we do other data structures: Linked lists, hash tables, primitives. The branded implementation of these things is what we typically make singular: Bitcoin, ethereum, monero (bittorrent, activitypub…)
Bittorrent implements a torrent swarm, activitypub implements a federated social network.
The only alternative to proof of work is proof of stake. And if the world ever ran on proof of stake crypto, it would make today’s wealth inequality look like a Marxist paradise.
Why? If the rewards are fixed and independent of the amount staked then there is no issue.
What matters is whether or not there’s any significant profit - the reward being bigger than the cost. PoW minimizes this with difficulty adjustment flushing most profits down the toilet. PoS must always have a return competitive with other investments.
Profits should be low, or else it promotes inequality.
No. It needs to offer a reward enough to pay the minimum number of validators needed to establish a secure network.
2x the number of stakers means 1/2 the return for each validator, but the network doesn’t care about that.
With too low a return they just invest elsewhere.
You get one asset with worse interest that can be spent (like dollars), and one asset with better interest that can’t be spent (like bonds). The poor hold the former, the rich hold mostly the latter.
Low returns and people leaving is not a problem.
There will still be enough validators to secure the network. The remaining will get a higher return.
What’s the minimum you mentioned earlier?
Minimum Viable Validator Count Ethereum requires at least 2,000-4,000 validators to maintain strong network security. Currently, there are over 900,000 validators actively securing Ethereum, making attacks prohibitively expensive.
If there’s a low barrier to entry then you’re just reinventing POW because someone can run as many nodes as possible with minimal stake in order to maximize returns. Can you give an example of one successful “fixed rewards” Crypto coin?
Ethereum.
The overall network issuance rate is fixed and currently roughly 4.5% per year. If more people stake then less is received per person.
Minimal stake is is $75k per validator.
This stake is slashed if you are dishonest.
You are exactly the type of person I’m talking about :\
Crypto is not the only use case for blockchain.
Blockchain can be useful in inherently trusted, closed ecosystems.
Depending on the use case, proof doesn’t even need to be anything more than a valid certificate (not work, not stake)
Consider a bank that develops its own blockchain ledger for internal use only, replacing their branch ledgers (which require daily reconciliation and propagation).
An immutable, distributed ledger has plenty of valid, valuable use cases without looking like crypto.
“Internal use only” blockchain is an oxymoron. If all contributors are trusted entities, then what does it matter if the data is stored in a blockchain vs any other data structure? If anything the amount of extra work to maintain and modify the blockchain in the case of errors just makes it unnecessary.
If that’s the best example of its many “valid, valuable use cases”, then it’s still a pass from me, dawg.
I have a large number of examples in this comment from four months ago. Please read them: https://lemmy.world/post/36683795/19677963
Literally the first example you gave was shut down a year later: https://www.reddit.com/r/CryptoReality/comments/1mpeh9z/when_crypto_bros_are_asked_for_a_blockchain_use/
Pretty obvious that its use in the first place was some FOMO executives trying to get in on “blockchain” technology, just like they’re doing with AI and LLMs now. Funny how BTC and ETH both plateaued a year ago, right around the time AI became the new thing.
I’m not going to bother reading the rest, I already wasted too much time arguing with a true believer. GL with the crypto that you say you don’t have.
Lol you’re a judgemental prick. I have zero crypto because that shit is absolutely moronic.
It’s like you talked to someone who supports nuclear power and responded “good luck with the nuclear bombs you definitely don’t have, true believer”
Sounds like you just want to hate on shit you don’t understand. Go on with your bad self though 👍
EDIT: LMAO from the article you linked they even point out that the tech itself isn’t the problem, but the willingness of businesses to invest in the improvements (which is, like, an incredibly common problem in business that does not in any way make the tech bad)
You probably preach about how nuclear power is terrible, how if it were so great there would be more of it, and people sticking to coal is proof that coal power is better.
Big brain you’ve got there 😂🫡
This is not true, even inside a bank. Employees commit fraud. Branches don’t trust each other.
Once built, maintaining is easy.
Er. The whole point is that blockchains are immutable.
But as long as you can have a trusted central authority, you don’t need blockchain without needing to trust all the clients. The central authority can enforce authentication and authorization for what each client can do, combine with proper logging it is already sufficient. Blockchain is only needed if there’s absolutely no central authority (which is not the case for any traditional business as by nature they already have a hierarchy where the top would act as the central authority, therefore any business that implements blockchain internally is just BS-ing).
How Walmart Canada Uses Blockchain to Solve Supply-Chain Challenges
Then why hasn’t a better blockchain based currency gained any popularity? If they don’t have critical mass then your distinction is meaningless. It turns out there is just zero real world need for an untrusted distributed ledger. Databases and governments solve the problem much better.
Questioning the technical virtues of an alternative product based on lack of critical mass adoption is pretty funny, when you consider we’re on the fediverse. I know that doesn’t defray your argument, but just an amusing observation.
I see why you might draw the comparison, but I actually don’t think the comparison is valid at all. Forums/communities can still be useful and fun with only a few people. Discord is also massively popular with a small community model, for a more successful example to compare with the fediverse. However a currency that nobody uses or accepts is entirely useless until mass adoption happens. That’s why they typically get mandated by force by governments.
Or if you need it for some other particular application, like drugs on the dark web.
Great points!
Two points:
https://www.forbes.com/digital-assets/categories/proof-of-stake-pos/
Etherium and virtually the whole rest of the crypto scene that is “not bitcoin” has pretty soundly rejected the wasteful Bitcoin design. There was even a fork of Bitcoin that would have used the much more efficient proof-of-stake, but since that would be bad for everyone with a proof-of-work “mining” rig it didn’t take over.
https://git-scm.com/
An “untrusted distributed ledger” is literally the backbone of modern software development. While you could plausibly split hairs and assert that git requires “trust”, I don’t think you’d wind up in a spot that both supports your assertion and a cognizable difference for anyone but mathematicians and security nerds. (And even if you did, the exact same sort of non-scam usages of blockchains are ones that operate like git, with the ledger used for something else.)
Blockchain is not synonymous with crypto. Why are you bringing up crypto specifically? Crypto is garbage. But Blockchain is not crypto
People bring up crypto because it is the only use of blockchain that isn’t worse than already established methods. And crypto is only “better” because it’s unregulated and allowed a bunch of scams to be pulled.
Critical mass is not required for internal systems. Not all implementations of blockchain are intended for public use.
I’m really tired of this. Blockchain. Is not. Crypto.
Here’s the research I did for everyone four months ago: https://lemmy.world/post/36683795/19677963
I’m not convinced there’s any internal use for blockchain. Internal implies under a specific umbrella, some overarching organisation, who can then be the central trusted server that makes blockchain unnecesary.
That said, non-public but open uses, such as tracking dealings between companies in markets with little trust and no single governments (the shipping example in your referenced comment) is indeed the thin slither of a plausible use-case.
Another limitation is that blockchain loses its benefits if anyone tries to design over the complexity of using it directly (using a ui that under the hood uses blockchain is no different to using a ui that talks to a central database, you’re trusting the central ui provider, you need to (at least be able to) build your own interface to realise the benefits of blockchain.
That means blockchain basically will never benefit individuals, it can’t. Sure, you could have multiple compatible uis shared around, but that’s no different security-wise to multiple central banks with an interoperable transfer system.
The only place blockchain has real benefits is when multiple large corporations/governments are interacting and don’t trust eachother/anyone.
See the link in my other replies for some examples of internal uses that still benefit from immutable, distributed ledgers.
Large organizations still have loss and risk from individual bad actors. Operating a central authority that validates every single transaction in a ledger, and validates ledger history and consistency, can be prohibitively complicated. A well designed blockchain implementation can resolve most of these issues.
A great example is a pharma/healthcare company that wants to manage medicine batch and expiration tracking, as well as distribution/patient assignment. With a traditional infrastructure every participant needs to phone home to a central authority. In a blockchain setup, peers can report ledger events one hop up and propagate it through the chain.
That’s a very simple example but I hope it gets my point across
Identifying anomalous behavior from bad actors is already a solved problem with databases and governing bodies.
Nobody said it wasn’t? But different models have different benefits and drawbacks?
Cryptocurrency development makes a whole bunch of arbitrary value-guided decisions during creation, all of these decisions have tradeoffs such that nobody has figured out a way to feature them all at the same time, or would they want to.
For example, bitcoin is fully auditable. Anyone with a copy of the bitcoin blockchain can review every single transaction in bitcoin’s history, and trace the flow of every last satoshi from it’s mining to today. This is because the developers of bitcoin place a high value on verifiable auditability and security. Conversely monero was developed for the purpose of being a completely untraceable, unauditable currency that still has a knowable supply. And ethereum was created in a manner that intentionally supported scripting, so that it could be used as a platform for novel applications and contracts. None of these primary features could be ported to either of the other two without breaking them completely, because of the deep programmatic implications of the requirements.
It’s not really a question of better or worse, but of use case. The fact of the matter is that the reason these three examples are the leading currencies for their use case is literally because nobody has yet been able to do a better job. And for bitcoin at least, at this point it’s security rests just as much in it’s wide adoption and interest as it’s design intent, so it’s unlikely that anyone ever will.
Shut up shut up shut up shut up shut up
Don’t ruin a good thing we’ve got going on here
First my buy order at 100K CAD just missed executing before the bottom bounce, now this?
What can I say? I woke up this morning and chose violence 🤷♀️