Hacker News Reader

"...the understanding of blockchain on reddit is non-existent."

The understanding of a lot of things are almost non-existent on reddit, or rather the people who know what they are talking about are buried by an algorithm that vastly prefers populist comments typed as quickly after OP as humanly (or robotically) possible. Of course this makes reading the article (or other posts) and having your comment visible mutually exclusive.

I didn't realize how little correct information was available on reddit until I became very specialized in a subject, went to that subject's sub, and saw that practically nobody there even knew the basics--yet they happily chat about the subject anyway, pretending they were experts. The rare exceptions were nearly invisible, lost in in-jokes and references.

Be-very-fucking-ware when asking advice about something you are unsure about on reddit, they will channel so much smoke up your ass you'll burp clouds. And once one person makes a confident declaration which is totally false or misunderstood--pretending they're some pro or expert--almost immediately every other guy in the thread is copy-pasting his response to every other instance of that question as quickly as they can to get points or whatever. It's silly.

For the record I try not to be so delusional as to believe I'm the only one who knows anything and everyone else is stupid--I'm totally average, lots of people are brighter than me. This is a systematic thing brought on accidentally by their forum design.

I don’t disagree. Do you have any examples of where Reddit got blockchain all wrong? While I have said many critical things that get me downvoted to oblivion on Reddit, I wouldn’t say everything I read on a sub where I understand the subject matter well is wrong.

I gave up reading comments there quite some time ago. As I said, reddit doesn't collectively get things (like blockchain) wrong--individuals get things right or wrong, and reddit most often promotes the wrong answers because wrong answers are more spectacular, more exciting, and quicker to type than the truth.

Reddit's chronological weighting is a very big factor in this. If your comment comes a few minutes after a post, such as the time it takes to read the content of the post--or if it takes time to thoughtfully type out a response--it is scored vastly lower by the black box and is less likely to be seen and voted on.

There are certainly exceptions, but unless you already know a lot about the subject at hand it's very hard to pick them out among the deluge of jokes, smark, and people rapidly parroting the misconceptions of others.

I was foolish enough to speak up on a networking related subreddit. Silly me, 20 years of experience, I should have known better than to share.

I got pummeled by a handful of small time networking guys who think their one rack data center is hot shit and there is just no other way to do things than to buy Cisco and parrot what their sales guy says. I kid you not, one of them seriously was calling his Cisco sales guy and asking about some Cisco news, got some spin from his sales guy and, and then posting what his sales guy told him.... (dude even said he was doing that) I was done.

Reddit is great for art, fashion and humanities related stuff.

Technical, it's great to locate yourself on the map and get a compass - then you have to leverage outside sources to get to the destination.

That's an eloquent way of putting it. Great point.

> Most techies moved on

But techies are not necessarily where the money is.

A large number of financial execs have spotted the hype, and see it as their "contribution" to their bank. It gives them the chance to do something different, maybe run their own team, and make a name for themselves.

From the stories I've heard they are not all clueless about the real potential either; They just see that this is likely to eat up even more investment, and they want to be on the gravy train.

I've heard that several of the consultancies are into this area as well, though they really have no clue what they are doing. Some of them have zero experience but they can still sell projects on the strength of their brand.

I agree. There are a few interesting areas, such as how a blockchain scheme might be useful to the Tor network, but it is all hype. I really laugh at those who think blockchain tech can/will replace property registries, real estate agents, and lawyers. I guess these were the same people who thought SecondLife would replace teleconferencing.

In the end you don't need decentralisation for verifying integrity - there is and was a large body of research for verifying integrity of such records like timestamping services.

Hint: blockchain is a timestamping service. What is new compared to the body of research you speak of is that it doesn't need trusted third party.

why not? for sure it's awesome to have tamper proof asset registries on the blockchain that are compatible worldwide. it will revolutionize everything really.

> for sure it's awesome to have tamper proof asset registries on the blockchain that are compatible worldwide

For one thing, the blockchain doesn't provide tamper proof asset registries because the blockchain can be forked (see the DAO fiasco for proof).

Second "temper proof asset registeries" aren't actually a problem that really needs to be solved. The hard part of any registry is making sure the data that goes in is valid.... and that is a meatspace warm-fuzzy-human problem, not a technical problem.

there is no contentious forking if it's true decentralization. But this is more of a thing if governments would participate and run nodes along with anyone that would like to attest too.

I don't believe in the saying that you need a problem to be solved. For sure the country I live in now cannot lookup easily what other real estate properties I own in every part of the world. All governments would need to select a distributed database and encode everything on it. Well this is now called Ethereum or insert your favorite platform here and they will develop on it because the technology makes it easy.

The "DAO fiasco" and Ethereum Classic is, in fact, proof that you are wrong about the lack of tamper-proofness.

oh, you mean a dictionary object containing a hash of the properties of the previous dictionary object?

Sooo innovative...

show me some dictionary objects that are replicated in computers in all parts of the world to keep them tamper proof and I will give you one satoshi

It's awesome until your private key gets stolen and suddenly a hacker in Russia owns your house.

how do property title considerations currently work?

what keeps a Russian from stealing your house today?

I'm working on a pretty cool project. It's called Regen Network. We're a distributed verification network for ecological data with a strong bias towards encouraging more regenerative agriculture. And we're using a blockchain as part of our verification model.

I have to agree that for the most part we haven't seen any real products come out of the crypto space. And there's a lot of chaff. It actually blows me away sometimes how many icos there are

You wouldn't download a car

Possibly the most succinct refutation of the reasonableness of IP rights ever devised. Rather than proving that copyrights should be respected, it forces people to realize that when computers allow you to completely encompass all human effort into the design or creation of a work, and remove entirely all human effort from the duplication or distribution, dollar value as a measure of scarcity or rarity becomes meaningless.

Also, dammit, we're getting close: https://cults3d.com/en/3d-model/various/mkultra-3d-printable...

I think it’s a reference to the famous anti-torrenting campaign that equates real world theft with torrenting.

https://m.youtube.com/watch?v=HmZm8vNHBSU

Yes, and we're saying that that campaign has inadvertently made one the strongest extant demonstrations that torrenting is actively morally good.

And that's amusingly ironic.

"You wouldn't download a car" is a reference to an IT Crowd skit which parodies the anti torrenting campaign you linked. IT Crowd is a wonderful British sitcom, I recommend any techy would enjoy it.

Or this is one of those historic moments where human civilization enters a new phase of international value creation and value transfer with monetary policy forever decoupled from government. That is the crypto narrative. ICOs and security tokens used for international crowdfunding are part of that bottom up borderless future.

>You wouldnt use blockchain for social media

Not saying you're wrong, but why not?

Social media content is not a scarce, fungible resource, it doesn't require global consensus and doesn't require an arms race of useless hashing to prevent double-spends (since there's nothing to spend).

If you want to prevent others posting as you, just sign posts/actions with a private key. If you want transferrable ownership of short usernames, that can be done by having the previous owner sign a timestamped file stating the (public key of) the new owner, maybe with some hashcash thrown in to prevent a double-transfer. Since there's no need for global consensus, you can limit spam using whitelists (i.e. ignore almost everybody); it can be reduced further by adding hashcash to posts/actions.

(Yes, hashcash didn't work for email; that's because email must interoperate with existing non-hashcash senders, and hence can suffer a downgrade attack; plus email allows clients to disconnect after sending, so the servers have no way to request them to include more hashcash, which drives down the price they're willing to accept to 0).

I'm very curious how you plan to explain this to non technical folk in a way that they'll actually care enough to slough through any of that. Blech.

I don't understand; messages should be tailored to their audience, and Hacker News commenters are technical folk. I wouldn't give that explanation to non-technical folk.

My non-tech answer would be that blockchains prevent some specific types of fraud in virtual currencies (e.g. trying to create new money out of thin air). Social media doesn't involve virtual money, e.g. it's fine to create new posts, friendships, etc. out of thin air. Hence they don't need a blockchain. The problems that social networks do have (e.g. posing as someone else, sending spam) can be dealt with in other, less expensive ways.

Blockchain (Bitcoin Cash) is great for social media. Checkout https://memo.cash. It's like an uncensorable twitter.

Just a few seconds ago I was wondering if anyone of importance (technical importance) did in fact talk positevely about blockchain(s). If anyone is or knows someone like that, I'm interested.

There are many. It's reasonable to assume that there will be use cases outside of the cryptocurrency space. Just because there's a ridiculous amount of hype doesn't mean there's nothing there.

> You wouldnt use blockchain for social media

You might if you wanted it to be censorship resistant and cryptographically verifiable: https://steemit.com

Does a censorship resistant social network need all the "overhead" of a blockchain type solution? (PoW with difficultly level, for one)

Although I could see PoW as a possible antispam measure, and it's been done before (prior art Hashcash)

The fact that you need to store all previous blocks (and content) to fully validate later blocks is an advantage in this case.

I was super into using hashcash for antiabuse in social stuff pre-bitcoin, but the difference in hashing rate in-browser versus in C even then rendered it somewhat ineffective at anti spam as an abuser could hash 10-100x faster, and the delays on the UX side for legitimate users made it a nonstarter. With GPUs and hashing asics now that gap is even worse.

(FWIW these days the Steem blockchain that holds the steemit content is delegated proof of stake, for block production.)

How does Steem store user generated content like images and videos? Are they in the blockchain itself? If so, how do they solve the problem of people downloading the blockchain being liable for illegal content added by other users?

Not sure what they do, but it's entirely possible to build a solution where all image/video content is stored on-blockchain as well.

>If so, how do they solve the problem of people downloading the blockchain being liable for illegal content added by other users?

Divide each file into N shards, make sure every miner has no more than N/2 shards of any file at any point in time. You only 'get' a complete file if you specifically ask for it.

I'm not a lawyer, but I believe it is unlikely this solution will hold up in court.

The Steem blockchain only holds unicode text: words, html, json. Images and video are embeds or img tags, which point to URLs on ipfs or elsewhere. That’s a separate challenge.

It seems like any on-chain user generated content is likely to be a problem; arbitrary binary data can be encoded as text. I suspect this problem already exits on the bitcoin chain though, so it's not like it's unique to Steem.

I never understood how Steem purports to be "censorship resistant". steemit.com is just a centralised, privately operated website hosted on AWS, so it's not any more censorship resistant than anything else hosted on AWS.

Every single post, comment, and vote is in the distributed Steem blockchain, copied in seconds to every full node, only a small fraction of which are run by Steemit—same as any other blockchain/p2p system.