Hacker News Reader

Smart contracts can’t even do anything interesting without a trusted source of data input. ...for a certain definition of "doing anything interesting", just as well as "trusted source"*.

As the locus of economy shifts more and more away from physical widgets and into information, the value of information goes up in relation to physical widgets. This makes Blockchain a practical, if not all-encompassing, tool for verification.

Compare a plain old car - the software onboard is a small fraction of its value - to any half-decent workstation, where the value of software can easily exceed the hardware's value several-fold.

Proving I had certain information at certain time point - e.g., a license for the software, or CAD design or other valuable file I've created - can easily provide a lot of value. And in this case "trusted source of information" is the mere fact of existence of a cryptographically secured license file naming me as the licensee at certain time point. Easily securable with the Blockchain.

> Proving I had certain information at certain time point

The blockchain solves the problem of proving ownership of a transferable asset. If you just want to prove that you have certain information at a certain time point there are much simpler solutions: Publicly release the hash of a message (let's say: "I created this using the licence key XXX-XXX-XXX" + signature from your private key) on Twitter/Facebook/Github/Archive.org etc.

Or even better: Sign a contract and send it to your law firm. Once someone breached a licence/contract you need to involve lawyers/regulators anyway.

an interesting way of describing a crypto blockchain like bitcoin or ethereum is that it's a way of incentivizing the storage and preservation of information. In particular, the information of a transaction from one address to another is attached with it a fee for storing it. Your "much simpler solutions" require a trusted third party that has no financial incentive to keep your information around, and depending on the lifetime of your needs may not persist long enough. 10 years ago your solution may have involved myspace or further back geocities or tripod, which are defunct / have refreshed their databases. Spending under 10 cents on a miner fee to store your hash is also a much better solution and cheaper solution than the lawyer escrow. Lawyer escrows and safes are vulnerable to local information loss that bitcoin/ethereum protects against through federated datastores (lots of important papers stored in the world trade center were destroyed on 9/11).

For ~20$ you can publish a hash in a classified ad. That's highly likely to outlast any and possibly every block chain you care to name. Further, because the way hash functions work you can use a single hash to timestamp unlimited amounts of information or individual documents.

This has been true for the last 30+ years yet almost never happens because it's not actually that useful.

The "best" use of blockchain type technology I have seen is the NIST randomness beacon, though I can't think of many applications other than a lottery or where you need to formally prove you used random numbers to pick something.

[0]: https://www.nist.gov/programs-projects/nist-randomness-beaco... [1]: https://beacon.nist.gov/home

> why would any real business want to use a public blockchain?

I think it is only important for a business when public trust is important. For example, a news agency could publish its articles in a public blockchain so the original article and any updates to that article would forever be auditable from that blockchain.

Why? News sites don’t even care about maintaining an edit history. They just make minor fixes and leave editors notes when they fix things. News on the block chain is not something anyone wants.

I do. Censorship resistance and removing the ability to rewrite history would be good. even if current media practice is different doesn’t mean it can’t change in the future.

Forbes just recently partnered with Civil, a blockchain journalism project from Consensys Studios, which is the commercial powerhouse of Ethereum’s cofounder.

I've worked in publishing, this initiative sounds like many other "save journalism with technology" initiatives that I've seen go nowhere, like the Coral Project.

But back to Civil. Why would I care that Forbes is going to publish SOME of their articles on this blockchain? Their site uses HTTPS, I'm always sure that the content I'm looking at actually came from Forbes. I never need to question that the person in their byline is who they say they are, and I'm not sure how a blockchain would help if I didn't: it seems like their big plan is to just put badges on articles?

This sounds like a grasp at relevance via blockchain hype.

I'm willing to be convinced that I'm wrong, but I feel pretty safe predicting that this will go nowhere. It doesn't solve a problem that Forbes actually has.

Precisely, and we already have tools at our disposal for that which also work very well in the news agency example (albeit internally): Wiki and Git. Blockchain adds nothing, and has disadvantages of wasting resources.

The only thing the Blockchain has proven to be useful for is cons in the form of speculation, akin to Ponzi scheme. I've seen it argued that it allows people to move away cheaply from bad currencies. Companies like PayPal and Stripe allow for the same already, and they use real currencies.

So what if a story or bit of information has to be retracted by law?

I second my sibling, newspapers have no interest in transparency.

> So what if a story or bit of information has to be retracted by law?

Then the news agency issues a retraction. It's already the case that the Internet will not forget, whether via the Internet Archive or http://www.newsdiffs.org/ .

How can data be retracted from a blockchain?

Both those sites are also subject to laws and can easily retract information if necessary.

> How can data be retracted from a blockchain?

In much the same way data can be retracted from printed copies of newspapers that already have widely circulated copies in both personal and institutional/library hands: by publishing a subsequent retraction later in the chain, like any other change.

The change and what was previously there is left for anyone to see in the blockchain.

And printed copies are not?

Sure. But in the days of it systems and digitalization less and less is being printed. It’s also quite difficult to find and access these printed copies someone might have stored somewhere.

> newspapers have no interest in transparency

I disagree, I mean... the whole point of journalism is to get to the truth -- in other words, to make an event transparent.

Why a blockchain?

ipfs [0] already provides persistent immutable links.

Creating a torrent with a feed for updates [1] would also provide distributed storage and hashing of the content with the ability to publish updates.

Or they could create a git repo and encourage others to mirror the git repo, such that they couldn't rebase away history without those mirrors noticing.

Or, and here's a radical idea, they could physically print and deliver papers to people, which means any updates would have to be included as noted amendments in future delivered papers.

However, why would any business want to do this? News sites these days are built on complex dynamic content and ads, which is antithetical to distributing the content in any such ways. If you can download just the text of the article, they're losing ad views. If you download the ads too, it now changes each time a new person purchases an ad on their ad network.

This is blockchain trying to dream up a problem for publishers which those publishers don't think they have. If they did have that problem, a git repo would be the most mature and simple way to implement it, not "blockchain" which includes nothing as mature as git under its umbrella.

[0]: https://ipfs.io/#uses

[1]: http://www.bittorrent.org/beps/bep_0039.html

> This is blockchain trying to dream up a problem for publishers which those publishers don't think they have.

Fair enough, it was just an example... maybe a bad example of when public trust is needed for something.

But, that level of trust and audibility would be nice in this era of conspiracies and fake news. I remember watching CNN on 9/11 and they reported that Flight 93 was shot down over Pensylvania. Maybe that was bad reporting, maybe it was accurate. But, now I can't find that footage anywhere.

I would love to find that footage now. Find out exactly what they said. Find out who their source was. Find out if it was bad reporting with bad sources, or find out if it was something more. Without that footage, it's just hearse and I'm sure we'll never know the truth about it now.

If only there was a group of people dedicated to archiving content... I dunno, maybe archive.org (aka "The Internet Archive")?

They even have a CNN section including a full recording of their broadcast on the morning of 9/11 [0]. I'm not going to waste the time watching this to see if they miss-reported some detail or not because I've got better things to do.

I don't see how blockchain helps here still; the problem with storing video is that copyright is complicated and there's a lot of it which costs a lot to store, not that we don't have the means to store and audit it.

> that level of trust and audibility would be nice in this era of conspiracies and fake news

Being able to see changes to articles doesn't actually help with those problems. Better journalistic practices help. People thinking critically helps. Removing the absurd click-bait headlines we've ended up on might help. Making news not be for-profit might help (related to the former). Changing the format of the fake news doesn't make it any less fake.

It seems like blockchain is being used as this idealistic thing which surely, if implemented, would not suffer from as many or more flaws than the status quo, when in reality no blockchain based thing has reached mainstream usability or mainstream users yet (arguably excepting bitcoin).

[0]: https://archive.org/details/CNN_20010911_130000_CNN_Live_Thi...

It's exhausting arguing with the same bogus "use cases" every other week but here goes in short:

- In your scenario you're not decentralized, you have a clear authority: the news agency

- The news agency can broadcast a signed merkle tree of the articles it publishes alongside the normal newspaper/news program etc... For instance through their website.

- Anybody who cares about it could maintain a copy of this signed tree locally.

- If the news agency censors or edits an article anybody with a copy of the original tree can prove it by showing that their tree has a valid signature from the news agency and yet the hashes don't match.

There, problem solved without having to worry about proof of work, proof of stake, transaction costs, 50% attacks of what have you.

Or, instead of paying to implement, operate and maintain that system forever, and also hoping that someone independent will store copies forever (because without that independent third party, the news agency's whole system is irrelevant)... maybe they could permanently store a hash of published revisions on a public blockchain for $0.05 each.

It's just another option that might turn out to be useful for some.

Although, I feel like there are other complications in the news agency use case. For instance, many potential changes to an article are irrelevant to meaning, so they don't imply bad intentions, despite having a totally different hash. If the only copy of an article is on the newspaper's site, then a simple format update a year later could break all the hashes. [edit] I mean so it would become impossible to check whether the article had been changed. A typical hash function doesn't give any sense of edit distance.

GP post doesn't seem to consider the apparent need to also keep unchanged copies of the original articles. Those would be much too expensive to store on today's typical blockchains.

I'm sure those problems have been discussed before, maybe solved, but I wonder if these issues make this news agency case less suitable for a simple Blockchain implementation?

If I understand what you're proposing you want to store the entire articles themselves in the blockchain. That indeed offers many benefits but it's also prohibitively expensive. In practice people will only store small hashes in the tree which can be then used to authenticate offline data. That means that you need somebody to archive that offline data to really make sense of the signed data so you have exactly the issues that you're describing if nobody cares to do it.

The fact that storing data on the blockchain is expensive is also not an optimization problem that can be solved easily with incremental optimizations, it stems from the fact that the data is transferred and mirrored forever on thousands of nodes. If storing data on the blockchain was cheap you'd quickly run into scalability problem. People who run Blockchain nodes don't want to mirror Wikipedia, the output of all the world's newspaper and child porn archives on their servers.

No, of course I'm not proposing anything like that. On the contrary, I was saying that it seemed impossible. Can't you see we actually agree about that!

I pointed out other possible problems with the original suggestion that news agencies should use blockchains. But you seem to completely misunderstand those words as being an attempt to support that original suggestion. Why?

I'm amazed that your reply to me repeats almost exactly what I had just said: "[blockchain could] store a hash of published revisions ... [but] copies of the original articles would be much too expensive to store on today's typical blockchains."

I hope this won't sound rude, but I've got to suggest that you read what people are actually saying, instead of assuming you know what they're going to say.

No wonder you have so many "exhausting arguments".... you're not listening to anyone but yourself.

Just because I feel like arguing a bit, here goes:

>- In your scenario you're not decentralized, you have a clear authority: the news agency Only due to current technical limitations. Shift to Blockchain, get distributed news agencies, with reporters[1] residing all over the world. Think of the time and costs saved by not having to send reporter out of a central office

>- The news agency can broadcast a signed merkle tree of the articles it publishes alongside the normal newspaper/news program etc... For instance through their website. Doesn't help a bit, given how they have total control over the content of the website, they could just as easily swap out the merkele root and claim it is the original one. Blockchain denies them this sleight of hand.

>- Anybody who cares about it could maintain a copy of this signed tree locally. ...and there's a well known tool for distributing that in a verified & timestamped format - Blockchain.

>- If the news agency censors or edits an article anybody with a copy of the original tree can prove it by showing that their tree has a valid signature from the news agency and yet the hashes don't match. Introduces a 'he-said, she-said' problem. Which the Blockchain neatly avoids.

[1] possibly with a lot more "citizen journalists" who do the reporting as a side activity, rather than mostly plain old full-time journalists.

>Shift to Blockchain, get distributed news agencies, with reporters

You'll have to break that one down for me. Can the blockchain also cure male pattern baldness? Asking for a friend.

>and there's a well known tool for distributing that in a verified & timestamped format - Blockchain.

News agencies typically already have a way to distribute information since that's literally their job. Even if they don't they could just tweet it or whatever. That's not really a problem.

>Introduces a 'he-said, she-said' problem. Which the Blockchain neatly avoids.

No it doesn't, if you can provide a valid signature from the "News Org" public key that doesn't match the article's checksum then it proves that "News Org" at some point signed conflicting data, proving that something has been tempered with. It doesn't matter if the person pointing that out is some anonymous user on 8chan, as long as the signature is valid it can only mean that the news organization did something shady or that their key got compromised.

That's the whole principle behind public key cryptography. I signed this comment before posting it, which means that if tomorrow I delete this message and you kept a copy you could show that I in fact authored it (or at least endorsed it somehow): https://pastebin.com/XKnQSFew

> Shift to Blockchain, get distributed news agencies, with reporters[1] residing all over the world

The associated press didn't seem to need a blockchain to have reporters scattered across the world. They simply recruited reporters that lived across the world.

What you're talking about sounds more like twitter or facebook where you have "reporters" (aka users) write poorly written updates from around the world.

> Doesn't help a bit, given how they have total control over the content of the website, they could just as easily swap out the merkele root and claim it is the original one

If you've ever force pushed to a git repo, you know why this is hard. Regardless, blockchain doesn't solve this, per the next point.

> Introduces a 'he-said, she-said' problem. Which the Blockchain neatly avoids.

The blockchain avoids it by insisting whoever spent the most energy is the source of truth.

Let's say an attacker wishes to insist the correct hash is Y, not X, and thus paint the news agency as a liar for having a story up with hash X.

All that person has to do is use more energy to publish a longer blockchain that includes Y... and the news agency doesn't really have an incentive to spend a lot of energy protecting the blockchain probably, so it won't be hard.

The incentive of mining makes sense for money. It doesn't make much sense for this case.

> and there's a well known tool for distributing that in a verified & timestamped format - Blockchain.

I'd go with "git", "rsync", "ftp", or "bittorrent" as well known tools for distributing some files, whether they're timestamped or not.

If I personally published a signed merkle tree... I think I'm too much of a nobody for people to download it to keep me honest you know. Unless I pay them to do it of course.

So then the problem is, I need to go to some entity that is famous enough. And I need my tree to connect to theirs. And all this on the cheap. I'm not a aware of any such thing in existence.

Another nice thing about the blockchain is that the timestamps of the signatures are provable.

And also nice is that there are no record keepers to whom which pressure can be applied.

> If I personally published a signed merkle tree... I think I'm too much of a nobody for people to download it to keep me honest you know. Unless I pay them to do it of course.

In the original scenario I assumed that there were third party people interested in detecting or even preventing modification of articles after they're published. If nobody cares about it then the whole thing is moot to begin with.

If your scenario is that you'd like to be able to timestamp data that has value to you but not others (say, something patent-related) then it's a different scenario and then yes, the blockchain might make sense, although you could also use many other trusted, albeit centralized, third parties to achieve the same thing. Making a public post of Facebook or some other big website for instance. The incentives would have to be very high for these companies to modify post dates for a scam. Even more so if you post it to several websites. I think there are also services who will add your signature to a merkle tree and then publish it in a newspaper for instance.

But I agree that in this situation publishing the hash on a blockchain is actually a good solution. It's simple and easy to program for. But keep in mind that this only works as long as the currency is valuable enough to keep miners honest and rewarded for their work (otherwise people will be able to attack the blockchain and rewrite history). So if a cryptocurrency fails and its value collapses then suddenly your proofs become worthless as a side effect.

> Smart contracts can’t even do anything interesting without a trusted source of data input

At least read the writings of Szabo before you decide what smart contracts are and aren't good for.

There are many several interesting applications here and now around trustless atomic swaps. Not necessarily between assets but cryptographic keys in general (for example DRM systems).

> It all seems like a marketing scheme to sell cyber beanie babies

Well, if it has a marketing budget then it isn't really trustless as there is some central authority to market it. Be careful out there. Rapidly increasing valuations attract scammers like sugar to flies.