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

261–270 of 724 posts

Re: AWS and Blockchain

#261

Post by fareesh »

Renevith wrote: post #150
fareesh wrote: post #127

From an Indian pov my take is the boundary markers use-case sounds pretty good. I don't expect an American flying around from EWR to Seattle to ever understand this intuitively without being exposed to the realities in this part of the world, but having thought about it for all of 20 seconds it sounds like a very good idea. Yes - it is a solution to the problem of power and corruption. In India there is corruption at…

Can you help me understand how this would help? The way I see it: if the government were on board with recording property boundaries in a public ledger, it could do it without a blockchain. And if it's not on board, then I don't see how a random third-party ledger would help, since the government wouldn't defer to it.

It would have to be a solution that involves the government

Re: AWS and Blockchain

#262

Post by fareesh »

alexjurkiewicz wrote: post #135
fareesh wrote: post #127

From an Indian pov my take is the boundary markers use-case sounds pretty good. I don't expect an American flying around from EWR to Seattle to ever understand this intuitively without being exposed to the realities in this part of the world, but having thought about it for all of 20 seconds it sounds like a very good idea. Yes - it is a solution to the problem of power and corruption. In India there is corruption at…

But how will poor people own enough computation to ensure rich people can't mount a 51% attack? If the database relies on trusted nodes run by eg, the government, why can't the government use MySQL and periodically publish a dump?

It would have to be a mix of government and non government nodes for it to work.

Re: AWS and Blockchain

#263

Post by nathias »

paulgb wrote: post #253
nathias wrote: post #244

Earlier quoted context omitted.

the original purpose of an invention does not limit the scope of its application

No, but ignoring the design criteria that an invention solves for is a recipe for using the wrong tool for the job.

'right tools for the right job' seems like just re-stating the original point in different terms, how right a tool is depends on the actual context a big part of that context is culture, preferences, politics etc. not just engineering concerns

Re: AWS and Blockchain

#264

Post by unionpivo »

inkyoto wrote: post #249
jillesvangurp wrote: post #239

Earlier quoted context omitted.

> Git is absolutely the wrong tool for the job as hashes in git are used as identifiers to identify a specific change, but not to cyrptographically checksum the document/the change to the data. Git commits can also merged, rewritten, squashed or deleted – something is absolutely unacceptable in the compliance space. Read up on git internals, it's a very elegant and simple design and you seem to assume a few things ab…

> Read up on git internals, it's a very elegant and simple design and you seem to assume a few things about it that are flat out wrong. I am well acquainted with git internals and object types, which include blobs, trees, commits, and tags. But I do acknowledge that I was wrong about git not hashing the object's contents – it does. > A git repository is in fact a Merkle tree. Every commit is chained to the previous c…

You can do the same things with ledger databases, it's just no so convenient, as just typing --force

If you have permission to replace database with with your new one (or spin up new instance of DB, adn redirect app to use that, depending on how you are set up), it's just not as convenient (by design).

I used centera WORM disks at my previous jobs, for added guarantees. (not by choice, stay away of you have a choice)

And once the GDPR came out, and we had to remove some content, that's what we did, made a copy, with bad data removed, nuked centera clean, (created new partition, and disabled old one) and restored now clean data on. As far as our applications were concerned nothing happened.

Re: AWS and Blockchain

#265

Post by jacknews »

dale_glass wrote: post #211
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…

Trust is expensive, lack of trust is even more so. Say, how would you build a skyscraper without trust? Are you going to set up an on-premises laboratory to test the quality of the concrete and steel? Are you going to personally verify the entire building's plans? Check whether the geological study was accurate? Follow every worker around to make sure they don't cut corners anywhere? The less you can trust that the m…

I think what you mean by trust here is just a thin layer over a vast system of enforced checks and balances.

Materials standards, building codes, fair contract law, etc, etc, etc, and the courts and police to back it all up.

Re: AWS and Blockchain

#266

Post by hiq »

BatteryMountain wrote: post #201
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…

What you are referring to is a distributed ledger, built on top of a blockchain data structure. The word Blockchain is so overloaded, people forget you can build a "blockchain" in 50 lines of code if not less. "Mining" new blocks is added on top of that, it's not a hard requirement - you can just create new blocks and link them up. You can choose to "mine" on CPU (slow) or on the GPU, if you want "mining" at all. You…

https://www.schneier.com/essays/archives/2019/02/theres_no_g... explains why talking about "blockchain" for anything but a public blockchain (including consensus etc.) is uninteresting.

Re: AWS and Blockchain

#267

Post by Nursie »

paulgb wrote: post #250
netfortius wrote: post #242

Earlier quoted context omitted.

I am sorry, but I think I am missing something here: isn't the "double spend problem" actually needing to be resolved by the choice of including one specific transaction (and no others), from a set of conflicting ones, truly relevant (vs. irrelevant)?

Yes, a specific transaction needs to be chosen as the consensus transaction. But if there are multiple to choose from (e.g. in the case of an attempted double spend), the miner can choose either one. The blockchain doesn’t guarantee, for example, that the transaction which appeared earlier will always be chosen (in practice, it’s likely the one with the highest transaction fee)

IIRC there were some hacks on bitcoin ATMs using this sort of knowledge.

The attacker would set up a withdrawal of (IIRC) CAD on the ATM, then transfer the bitcoin. The ATM would see the transaction, and dispense the cash. However the attacker would immediately, before the next block was generated, send the same BTC to another address which they controlled, with a higher fee. The miner would discard the first one and roll the update into a new block.

I may have one or two of the details wrong (is there an explicit way in the bitcoin protocol to mark a transaction as superseding a previous one?), but anyway, like many things, it worked until they got caught.

Re: AWS and Blockchain

#268

Post by TimJRobinson »

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…

I used to think this, but working with DeFi on Ethereum for a while I've realized the killer feature is actually permissionless composability. Which is why enterprise block chains make little sense.

Having one neutral platform, controlled by no one, with standardized API's and immutable open programs that anyone can permissionlessly build on - is amazing.

We've never had this before, and it's incredible how fast the DeFi space is moving because of it. I work for a platform (Balancer) that has had over 20 different companies (Aave, Element, CopperLaunch, Gyro, Aura, Hidden Hand) build financial applications on top of our tech stack in the last 2 years.

Then there are different wallets (Metamask, Rainbow, Ledger), aggregators (1inch, Matcha, 0x), portfolio management tools (Zapper, Zerion, DeFi Saver).

All of these work with each other mostly out of the box and without any formal partnerships between anyone. This is a radical new way of building finance and it's the fastest paced industry I've ever been a part of.

It's similar to AWS where it got exponentially more valuable as each new service was added - but the entire world can contribute to it. Like AWS people didn't understand the value initially, but over time they realized how big of a deal it was as more and more components were added.

Re: AWS and Blockchain

#269

Post by Terretta »

imglorp wrote: post #35

Tim didn't mention a contemporary sibling tech at AWS which should not be thrown out with the bathwater: private (centralized), immutable, signed ledgers. Their version is called QLDB and does have concrete applications where you need an provable audit trail of a stream of transactions with search and index on top. I was on a team at a fintech closely engaged with AWS to prototype a money movement system using that a…

Read tbray's article looking for QLDB. There were keywords along the way suggesting the possibility, but it didn't get there.

As a NYC financial services CTO, a few years back I spoke to Jassy personally, asking him to please not make 'traditional' blockchain nonsense, but to please productize and release a distributed ledger primitive. A few weeks later, AWS got back to us and said they were willing to work on such a thing, and a few groups (perhaps such as yours) jumped on as well. In our case, we were looking to use it to help manage multiparty contracts where we as a bank were trusted by the multiple parties who didn't need to trust each other.

When AWS did eventually* announce AWS Blockchain at re:Invent, they announced QLDB alongside. One was what everyone was begging for, one was useful. I'll leave which is which as an exercise for the reader. :-)

TL;DR:

I encourage anyone serious about solving financial services problems where blockchain comes up to look at QLDB instead, but also anyone looking at security, compliance attestation, or other multi-party trust or regtech related problems.

* Note: "eventually" as in, quite late to the blockchain hype party, and dragged kicking and screaming by the sales org, as tbray describes, so that at least they'd stay in the room with "big IT" signing up for IBMs hyperledger and other failed experiments.

Re: AWS and Blockchain

#270

Post by hiq »

davide_v wrote: post #243

In Italy a famous news website is using blockchain since 2020 to, I quote literally: "help readers check source of news", "strengthen bonds of trust between its organization and its readers and customers", "trace the history and source of each news item." News (EN): https://www.ansa.it/english/news/science_tecnology/2020/04/0... Example of news in the blockchain: https://blockchain.check.ansa.it/landing/b150aff9f028f…

What does that bring on top of TLS?

If you want to check whether an article ever appeared (even if it's no longer there), then something like https://en.wikipedia.org/wiki/Certificate_Transparency would be enough.

No that's not a blockchain, at least not for a useful definition of blockchain (see https://www.schneier.com/essays/archives/2019/02/theres_no_g...).

Post reply on HN