Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

AWS and Blockchain

tbray.org

Post reply on HN
Print view View original

191–200 of 724 posts

Re: AWS and Blockchain

#191

Post by everfree »

autotune wrote: post #152
marcell wrote: post #132

Blockchain skeptics are all defending the tech of the current financial system, and implicitly arguing it can’t be improved. They are wrong. The tech underlying the current system is really bad. For example: (1) How can I perform an atomic transaction across two stock exchanges? (2) How can I take stock held on one exchange, and use it as collateral on an unaffiliated lending platform? (3) How can I launch a new fina…

The underlying system is mostly written in COBOL and desperately needs a replacement before all the COBOL engineers die out with many large bags of money going to their descendants. Blockchain, no matter how good it might be or might have been, is clearly not that system due to its now tarnished reputation at this point in time.

> Blockchain, no matter how good it might be or might have been, is clearly not that system due to its now tarnished reputation at this point in time.

I never understand how people can consider the reputation of glorified timestamped linked lists to be "tarnished". This isn't a company or even a particular piece of software, it's a class of data structures. You're trying to tell me that a particular way of structuring data has a reputation that can be tarnished?

Re: AWS and Blockchain

#192

Post by EFreethought »

fabianfabian wrote: post #188
EFreethought wrote: post #178

Earlier quoted context omitted.

Is there a double spend problem in fiat currencies? Honestly, I do not think there is. So what is the point of cryptocurrency?

yes, the entire reason a bank is needed in all online transactions is because nobody trusts YOU to not double spend, but they do trust the bank to make sure they don't let you spend your money twice.

Then the banks have solved the double-spending problem, so in effect there isn't one.

I was asking the question because if double-spending has been solved, then why do we need crypto/blockchain?

Re: AWS and Blockchain

#193

Post by everfree »

Nursie wrote: post #128
subradios wrote: post #123

Blockchains are an incredible technology. It is obvious that their first attempted uses are contractual and monetary, because blockchain guarantees can make things better fastest in those arenas. The unfortunate part is it doesn't seem like anyone has really figured out the how yet. I am reminded of the Web, Amazon was the first company to really "get" e-commerce and made all the money - there will likely be a simila…

We're already 14 years in. Where is the evidence it can make anything better? Why hasn't the amazon thing happened in the last ten years of blockchain?

To be fair, Amazon didn't hit their stride as a retailer until the 2000s, roughly 20 years after the internet's "big bang" date.

And while we're 14 years into distributed ledger technology, we're only 7 years into turing-complete distributed ledger technology, which I would argue is the actual innovation.

Re: AWS and Blockchain

#194

Post by Terr_ »

runeks wrote: post #167

Blockchain was invented to solve one particular problem: distributed consensus on a sequence of transactions, where the choice of which transaction to include from a set of conflicting transactions is irrelevant. The latter property here is key to understanding where blockchain is useful. It was created to solve the "double spend problem", ie. two transitions that spend the same coin but send it to different recipien…

> Only if your problem exhibits the above property (and it's a distributed system) does using a blockchain make sense.

I think we also need to specify that it's also an extremely distinct and rare type of "distributed": When you must support unrestricted and unlimited creation of new participant nodes.

That's the fundamental requirement that drives an entire cascade of other blockchain design choices, such as proof-of-work to prevent a takeover by sockpuppet-nodes, and mining to incentivize regular consistent participation by a presumed-benign majority. Without that requirement and workarounds, we're left with a kind of regular distributed database, pre-hype technology that already featured hashes and linked-lists.

This requirement ("anybody can become part of the network anonymously") also makes for a great litmus-test for whether you need blockchain.

For example, in national elections we probably don't want/need literally any number of computers anywhere in the world to jump in and deciding what happens based on their CPU power. It would be dramatically faster/easier/safer to simply define a fixed list of nodes, where each state runs a couple plus the federal government, and then rely on the fact that evildoers would have to compromise too many different systems and agencies.

Re: AWS and Blockchain

#195

Post by v-erne »

smt88 wrote: post #24

I had a real experience that perfectly reflects the research presented here (safer ledgers useful, blockchains are not). I accidentally got wrapped up in a project to automate some HR functions, and the product manager demanded that it must be blockchain because blockchains are the future. It turns out that append-only databases are well-suited for HR records, and (especially when dealing with things like background…

>>It turns out that append-only databases are well-suited for HR records

It may seem so, bu then You find out what are Your governments limit's for storing workers data, and Your company lawyers forces You to make it possible to delete everything that past that limit to limit legal risk and be gdpr compliant. And at this moment You find out that world is not constant and far from Your ideal model of spherical cow.

Re: AWS and Blockchain

#196

Post by Vespasian »

jiggawatts wrote: post #118
amluto wrote: post #67

Earlier quoted context omitted.

An NFT representing ownership of farmland is, frankly, absurd. If I own farmland and lose my private key (or screw up a smart contract and lock it away forever, etc), should that land never be farmed again? If a river moves and boundaries must change but the NFT isn’t aware of this, what happens? If someone with guns takes my land, who cares about my NFT? Maybe technology can help create a system where transactions r…

Not to mention that the only thing that actually matters is what's in the government's land title ledger. They're the ones with the guns that enforce (or don't enforce) ownership. These crypto crazies absolutely crack me up. Can you imagine some dirt poor farmer in the arse end of nowhere confronting the local constabulary that were bribed by their rich neighbour? "You see here in my wallet how my Landz Titles NFT cl…

> Some people are so naive and clueless that it's a parody of entitlement

It's nice to think that one can stand his/her ground and not have to rely on others especially in societies were rugged individualism is a concept.

Reality disagrees with that and has proven throughout history that organized and coordinated groups of people are much more powerful than the sum of their parts.

It's the same with prepers (the crazy political kind), who think that in an apocalyptical world of madness and warlords their 1 family home will not just be sieged down immediately with them either bending the knee or getting killed.

Re: AWS and Blockchain

#197

Post by delaaxe »

bredren wrote: post #185
Vespasian wrote: post #179

Earlier quoted context omitted.

As the article's author says at some point society is always built on trust in another entity. Crypto fully shifts this from the database operator to the software developer and is willing to accept significant trade offs to do so. Most of the unique complexity and challenges in blockchain technology originates from this.

The author also says all of the implementations he witnessed were reliant on databases and did not require the unique complexity and challenges of the blockchain.

Whether or not the underlying storage mechanism uses a database is not the point.

The point is who can edit that database and under which rules (imagine that database holds a lot of your money and you live in an oppressive regime).

Re: AWS and Blockchain

#198

Post by autotune »

everfree wrote: post #191
autotune wrote: post #152

Earlier quoted context omitted.

The underlying system is mostly written in COBOL and desperately needs a replacement before all the COBOL engineers die out with many large bags of money going to their descendants. Blockchain, no matter how good it might be or might have been, is clearly not that system due to its now tarnished reputation at this point in time.

> Blockchain, no matter how good it might be or might have been, is clearly not that system due to its now tarnished reputation at this point in time. I never understand how people can consider the reputation of glorified timestamped linked lists to be "tarnished". This isn't a company or even a particular piece of software, it's a class of data structures. You're trying to tell me that a particular way of structurin…

Someone comes out with a new database called "BigBucksDB" that uses a previously unknown technology called "blockchain." Suddenly every scammer and grifter decides to pump up this new "BigBucksDB" and emphasizes how safe and secure it is, thanks to blockchain, because all transactions are public and the ledger or whatever can't be modified in any way except to append new data. Then the companies using "BigBucksDB" decide to transfer all of their customers funds into their own accounts at a real bank and declare bankruptcy. What was the point of implementing "blockchain" in the first place? What use case can it possibly have for future services that would be worth taking a risk on it to implement when its primary use case for crypto transactions just fell apart? I know practically nothing about blockchain except how it was touted in the media as something that was supposed to make it difficult to modify crypto transactions and commit fraud. That fraud has still managed to be committed. What was even the point?

Re: AWS and Blockchain

#199

Post by AlotOfReading »

everfree wrote: post #181
AlotOfReading wrote: post #156

Earlier quoted context omitted.

Healthcare records frequently need to be modified (e.g. a test was done incorrectly and needs to be voided). Sure, you can stick another entry in there saying "modify this other entry", but then any operations involve canonicalizing the state by scanning the entire subsequent chain. All the caching/indexing optimizations I can think of to make this not a complete nightmare also break all the data privacy and RTBF pro…

> then any operations involve canonicalizing the state by scanning the entire subsequent chain Sounds exactly like what blockchain node software does when you first start it up and it syncs the chain?

The scales are completely different orders of magnitude. Crypto chains are miniscule, a few hundred GB max and even that's problematic for a lot of devices. A single CT scan is something like 20+ GB. There are ~220,000 of those in the US every single day and probably billions or trillions of other little things being logged. You can store offchain, but then you're a. Centralizing things again and b. Probably violating patient data privacy.

Re: AWS and Blockchain

#200

Post by asah »

jacknews wrote: post #141

I think the cost of trust is underestimated. All those shiny skyscrapers that the banks build? It's for trust, along with a good deal of banking licensing requirements and regulation. The example with the farmers' fields is quite valid. We forget in our relatively well-governed western nations that some governments really can't be trusted to simply keep records straight and not 'lose' vital documents, in cases where…

Yes, and #include cost of military, courts and police ("guys with guns")
Post reply on HN