Microsoft is shutting down its Azure Blockchain Service • Hacker News Board

Re: Microsoft is shutting down its Azure Blockchain Service

#161

Post by lordlimecat » Thu, May 13, 2021, 4:13 PM UTC

speeder wrote: post #45

koheripbal wrote: post #34

Earlier quoted context omitted.

I don't see why these would need a blockchain rather than just a centralized database. I'm very skeptical of these use cases.

The purpose of blockchain is to solve what happens when neither side of a transaction trust the other. For example, suppose US and Iran agreed to pay each other some amount as proof of their sactions deal... US doesn't trust Iran or its central bank. Iran doesn't trust SWIFT, since it can be easily manipulated by its enemies (as it has been, before). US doesn't trust the alternative to SWIFT, since it is run by Russi…

>The purpose of blockchain is to solve what happens when neither side of a transaction trust the other.

Normally the courts solve this problem, so this is only an issue if one or both parties are operating outside of the law.

Which, for normal financial enterprises trying to do business, creates regulatory problems: Chase can't exactly transact with the PLA without getting bulldozed by regulators, and you can't exactly hide the transaction on blockchain (since its immutable).

In the scenario you describe, there are ways of cheating even in blockchain (51% attack), and there are ways of solving the issue outside of blockchain (military force, embargos).

Re: Microsoft is shutting down its Azure Blockchain Service

#162

Post by nonameiguess » Thu, May 13, 2021, 4:14 PM UTC

speeder wrote: post #45

koheripbal wrote: post #34

Earlier quoted context omitted.

I don't see why these would need a blockchain rather than just a centralized database. I'm very skeptical of these use cases.

The purpose of blockchain is to solve what happens when neither side of a transaction trust the other. For example, suppose US and Iran agreed to pay each other some amount as proof of their sactions deal... US doesn't trust Iran or its central bank. Iran doesn't trust SWIFT, since it can be easily manipulated by its enemies (as it has been, before). US doesn't trust the alternative to SWIFT, since it is run by Russi…

We actually already do this. See this job description: https://acpol2.army.mil/fasclass/search_fs/search_fs_output....

This would generally be responsible for paying parties that actually trust you, using Treasury checks and possibly small amounts of cash, but there is an equivalent position within the uniformed Finance Corps where you pay untrusted parties who also don't trust you. For that purpose, you deliver extremely large briefcases full of cash. Basically, the same idea as a drug deal where the buyer and seller don't trust each other. Everything is transacted in cash, and both parties are carrying a lot of guns. This is generally how we make payments to local warlords in theaters of combat where they are not by any means an internationally recognized legitimate authority you can make any kind of a treaty with, but you need to buy their cooperation anyway.

As for trusting they'll actually deliver what you're buying, again, that's what the guns are for.

Heck, as it stands right now just for normal consumers, buyers and sellers don't need to trust each other. They just need a legal authority who can prosecute each side and they need to trust that legal authority. Barring that, they need an extra legal authority they trust with guns.

As it stands, the obvious limitation of a blockchain is it only helps if your transaction is solely the exchange of data stored on the blockchain itself. Otherwise, you're back to the same issue. The fact that a transaction is recorded on a ledger doesn't mean either party actually received the goods they expected, and it doesn't mean they have any means of remediation, legal or otherwise, if they don't. The fact that a ledger says you own something doesn't mean you're in possession of it, and it doesn't mean the party that is in possession of it has to give it to you.

Re: Microsoft is shutting down its Azure Blockchain Service

#163

Post by rodiger » Thu, May 13, 2021, 4:15 PM UTC

whateveracct wrote: post #154

rodiger wrote: post #83

Earlier quoted context omitted.

There's a weird insistence (especially here on HN) that if blockchain isn't doing something that can't be done in another way then it is useless. In reality just doing normal, established processes in a distributed way often is the value-add. See decentralized finance[0]. All these tools and primitives exist in traditional finance, but when it's decentralized all market participants can get a cut of the rewards inste…

dfinity is pretty centralized. they cut corners and basically hand-pick their mining data centers. they only qualify as "decentralized" because they work with multiple unrelated data centers. i am pretty sure multiple hard problems that they need to solve to reach true decentralization have been punted because they hand pick their miners. also, more than decentralization, they mostly sell their dev-friendly programmi…

They don't hand-pick miners at the end of the day- it's an on-chain vote (by token-holders and I'm sure they own a large sum so point taken).

Maybe not the best example- I'm not too familiar with the structure and they definitely over-hype everything about the project. Still the parallel is there of making a centralized thing more accessible / open to more market participants.

Re: Microsoft is shutting down its Azure Blockchain Service

#164

Post by dragonwriter » Thu, May 13, 2021, 4:15 PM UTC

pornel wrote: post #160

pragmar wrote: post #142

Earlier quoted context omitted.

Here are a couple. Identity. Once you log into a website a MetaMask address, a lot of things begin to come into focus. Create as many as you want, move contents around, etc. It's a lot more empowered than google/fb oauth feudalism. Physical and intellectual property registries. Could be far more efficient, accurate, and resilient to bad actors.

Property registries are one of those stupid ideas for blockchain. If someone was going to steal your property, they're not going to care what some blockchain says about it. If you only need an informational registry for well-behaved cooperating parties, you don't need it to be completely trustless. There are many other ways to decentralize storage and ensure consensus when you don't require mexican-standoff level of…

>Property registries are one of those stupid ideas for blockchain.

Specifically, the thing that is essential for most property registries is recognition and enforcement by a meat-space government. So, while a digital ledger is a good idea compared to, well, lots of existing property registries, decentralization/trustlessness isn’t important, adoption by a trustworthy and capable enforcer is.

Re: Microsoft is shutting down its Azure Blockchain Service

#165

Post by rodiger » Thu, May 13, 2021, 4:16 PM UTC

Spooky23 wrote: post #157

rodiger wrote: post #102

Earlier quoted context omitted.

I can be a market maker for an exchange with just $100... not sure how that isn't anything new. I can get a flash loan for $10,000,000 with zero credit to profit off of arbitrage opportunities. Yes, this is all focused on money but nothing to do with speculation or ponzi schemes. DeFi is sort of billed as an "internet of money," so it makes sense that it's focused on financial systems. Unfortunately many do just dump…

The pump and dump hype around cryptocurrency turns a lot of people off.

Yeah it does. It's a shame and hopefully the risks are made clearer over time.

Re: Microsoft is shutting down its Azure Blockchain Service

#166

Post by SubiculumCode » Thu, May 13, 2021, 4:16 PM UTC

jandrese wrote: post #132

matthewaveryusa wrote: post #70

Earlier quoted context omitted.

>contributing compute power is economically incentivized I think this is the aspect in blockchain that has really been stretched. Most problems are within closed networks and solved with digital signatures -- is the act of signing a record proof of work? If you squint hard enough and razzle-dazzle it enough I could see how you could call it that -- or simply call it a database with a signature column :)

I'm always saddened that the energy that goes into those blockchain signatures is basically wasted. Why couldn't bitcoin work by having the miners fold proteins or something? Proof of work comes from advancing the state of the art just a hair each time. Imagine if all of the computation power being used to prop up bitcoin were also doing something productive?

Proof of Stake is taking over, in Ethereal, Polygon, and other alt coins.

Re: Microsoft is shutting down its Azure Blockchain Service

#167

Post by SubiculumCode » Thu, May 13, 2021, 4:20 PM UTC

google234123 wrote: post #137

peterkelly wrote: post #14

> Microsoft's documentation suggests users start migrating to an alternative now So, a database?

But can a database handle more than 5 transactions per second?

If you think modern block chain is still stuck at 5 tps, then you have not been paying attention.

Re: Microsoft is shutting down its Azure Blockchain Service

#168

Post by pragmar » Thu, May 13, 2021, 4:20 PM UTC

iudqnolq wrote: post #155

pragmar wrote: post #142

Earlier quoted context omitted.

Here are a couple. Identity. Once you log into a website a MetaMask address, a lot of things begin to come into focus. Create as many as you want, move contents around, etc. It's a lot more empowered than google/fb oauth feudalism. Physical and intellectual property registries. Could be far more efficient, accurate, and resilient to bad actors.

So a bad actor gets a fraudulent invention registered. In the current system some people think really hard and then consider deleting it. This has many flaws, but the flaws are in the human bits. Where they write down their decisions is an implementation derail. How will blockchain make it better? Unless you've figured out how to have a computer figure out what inventions should qualify for protection and what infrin…

>>> How will blockchain make it better?

You're focusing on patents. Countries operate copyright offices--clerks, submission fees, red tape, human error. With blockchain, I can produce and store an md5 of a audio file, pdf, or video for next to nothing. You now have cryptographic signed proof of generation at a particular time, and a system that stores the data perpetually. Yes, you still need courts and judges to adjudicate conflicts, I never said it was a 100% solution.

Re: Microsoft is shutting down its Azure Blockchain Service

#169

Post by SubiculumCode » Thu, May 13, 2021, 4:21 PM UTC

alkonaut wrote: post #129

_pdp_ wrote: post #3

What are the practical uses of private blockchains anyway?

Every company on the planet seems to be asking that. It’s like “AI”, one of those technologies that you can’t risk everyone else using while you don’t. So for some reason having a solution and look for the problem is suddenly acceptable. Few seem to find the problem though. If you read through typical descriptions of how blockchains can be applied in industry, you’ll notice that they are extremely vague. That should…

Ummm Vet? Theta? Real worlds solutions to supply chain and video streaming, respectively.

Re: Microsoft is shutting down its Azure Blockchain Service

#170

Post by mumblemumble » Thu, May 13, 2021, 4:22 PM UTC

dale_glass wrote: post #159

throwawaygh wrote: post #141

Earlier quoted context omitted.

Git is used almost exclusively in a centralized fashion. The use patterns really aren't all that different from SVN, even if having a local copy is kinda nice. But, like you said, having a local copy isn't really the same thing as decentralized. "master" is still a thing. More importantly, the next big shift in version control technology is/will be back toward centralization . You already see this with eg Google Docs…

Git is very much used in a distributed fashion. Eg, on github. Right now, I'm working on one of the forks of a dead commercial project. Our fork in turn has 61 forks right now, some of which may diverge further and have multiple people working on them, which at some point may or not contribute things upstream.

But not in the blockchain sense. All those forks aren't independently contributing to a single source of truth in a peer-to-peer manner.

In practice, those forks generally serve one of two purposes. Either they're for working separately on changes that you intend to submit upstream to the agreed-upon central repository, or you're intending to legitimately fork the project and create your own new central repository that's relatively independent of the original.

Despite from the fact that, thanks to our industry's love for overloading technical jargon, we happen to use the word "distributed" to describe both use cases, they're really quite different in practice.