Hacker News new | past | comments | ask | show | jobs | submit login
What is Chia cryptocurrency and why is it bad news for hard drives? (tomsguide.com)
161 points by fortran77 8 months ago | hide | past | favorite | 254 comments

I just... do not understand how smart, even _brilliant_ people, the sort of people who know enough to make something like this, can look at Bitcoin... the waste, the way it has utterly failed to achieve what it set out to do[1], everything, and thinks, I know, I want to do that, but instead of blowing up the market for GPUs it'll be SSDs. Why? To get rich? It even seems doubtful to me that the people who developed Chia are the ones who'll make the money off it, if there is any to make.

[1] Who uses Bitcoin for peer-to-peer payments? And the cryptocurrency ecosystem as it exists right now, far from allowing people to transact without third parties, is driving a cambrian explosion in third parties, because most people are incapable of transacting Bitcoin themselves.

> It even seems doubtful to me that the people who developed Chia are the ones who'll make the money off it, if there is any to make.

Oh I wouldn't be so sure about that. From their FAQ:

" Why are we pre-farming?

Chia has a novel business model to both, lower volatility of the coin and increase adoption. By loaning Chia and managing the interest rates of those loans as well as other tools like buying our stock with chia coins, we hope to lower the quarter to quarter volatility of the coin. To drive adoption, we intend to loan Chia to Global 5000 companies who will use it to pay their international vendors quickly, less expensively, and more securely. We also intend to use the strategic reserve to aid development and adoption. We plan to do things like invest in promising startups in the Chia ecosystem, potentially increase farming rewards during limited periods of time to spur additional farming, and fund corporations paying 105% of the value of their international payables in Chia instead of fiat currency.

Is Chia doing an ICO?

No. Chia is not planning an ICO. Instead, our goal is to take the company’s equity public on an American stock exchange. This way, shareholders can share risk and return with management with transparency and disclosure and we can use well understood corporate controls to make binding statements about how Chia Network intends to use the Strategic Reserve. The Chia digital money is meant to be a useful payment instrument and not an investment opportunity. Chia intends to complete a fully compliant SEC registered equity IPO and will come to market as market-timing is amenable after the launch of mainnet. "

The people who developed Chia hold a massive pre-mined collection, and plan to go public (as in stock, not ICO) later, so definitely seems like they could profit.

That then makes this a very transparent pump-and-dump case - going public would mean the SEC would be right on them.

They've been developing new technology and algorithms for many years to make chia possible, hardly a pump and dump.

> Who uses Bitcoin for peer-to-peer payments? And the cryptocurrency ecosystem as it exists right now, far from allowing people to transact without third parties

Bitcoin may not be what it set out to be, but how is that a bad thing? The market decided whatever it should be.

What bitcoin did is inspire creation of many, genuinely interesting currencies, some even learning from bitcoin's failings: Stellar.org that aims to form a new kind of backbone for instant worldwide asset transfers; Monero that enables anonymous payments; and MakerDAO that makes it possible to transact across borders with a stable digital-currency, DAI.

Banks (like eco, onjuno, dharma) and other financial instruments on the Ethereum network (like curve, aave, compound) are a reality today. The space is far more than just currencies. Besides, networks like Cosmos, Polkadot are going to enable an even richer crossover among various currency networks that today is exclusive to Ethereum altcoins via uniswap, mdex et al.

I agree that mining seems like a lot of waste but Delegated Proof of Stake and other advances aim to address those issues.

Is there an accessible introduction, a YouTube video perhaps, to crypto instruments beyond Bitcoin? I have a basic understanding of how bitcoin works from SecurityNow and would like to learn just a bit more to test the waters.

> Is there an accessible introduction, a YouTube video perhaps, to crypto instruments beyond Bitcoin?

I take you're specifically interested in what's known as "DeFi" (decentralised finance)?

Multiple things going on on that front, but you could start here: https://archive.is/bJQt4 (coincentral); this video's a pretty good overview on Ethereum's DeFi ecosystem https://www.youtube-nocookie.com/embed/k9HYC0EJU6E and this one on what's happened in just 2020 https://www.youtube-nocookie.com/embed/qFBYB4W2tqU

DeFi, today, is poised to grow beyond Ethereum. Check out cross-chain networks like Polkadot (https://archive.is/GZVlM) and Cosmos (https://archive.is/1pWHF). matic.network (now Polygon) is another interesting side-chain network that seeks to take Ethereum tokens cross-chain.

If you're looking to build a DeFi product on Ethereum, docs at 0x.org, aave.com are good places to start (since they integrate out of the box with decentralised exchanges like Uniswap, Balancer etc). Circle.com, Sendwyre.com, Simplex.com, and Nowpayments.io are some of the leading providers for fiat <-> crypto-currency transfer. I am sure Binance has some offering in this space too.

Great! Thanks for the links!

The real meat of any the cryptocurrencies is in their consensus algorithm, which can be difficult to understand. Chia has these resources listed:

> We have academic papers and presentations that give detail about our new consensus algorithm and blockchain software. In 2019 we revealed our Green Paper outlining the construction of Proofs of Space and Time and illustrating many of the design choices of Chia.


> We have updated our consensus algorithm and you can review and commment on our working document.


> There is also a 2019 talk from Mariano Sorgente at MIT on how to achieve Nakamoto consensus with Proof of Space and VDFs.


> Bram Cohen presented at Stanford on February 2018 on Proofs of Space. Our advisors, Dan Boneh, Benedikt Bünz, and Ben Fisch published a survey of VDFs which are the underlying technology of Proof of Time. Lipa Long published an explanation of class groups that our Proofs of Time is based on. Bram presented Beyond Hellman’s Time-Memory Trade-Offs with Applications to Proofs of Space at BPASE ‘18 in January 2018 based on the academic paper and these slides by Hamza Abusalah. Ben Fisch gave a talk at BPASE 2018 in January 2018 on Verifiable Delay Functions. Bram spoke at Blockchain at Berkeley (which starts about 20:00) in March 2018 with slides. Bram gave a talk at BPASE 2017 in January 2017 on removing waste with Proofs of Space and Time (slides).


> Bram gave a talk at SF Bitcoin Devs Seminar about data structures for scaling Bitcoin with slides and Merkle Set code. Bram gave a talk at SF Bitcoin Devs Seminar about removing waste from cryptocurrencies.

Thank you.

These people likely believe they are getting in on a trend that's comparable to the internet/web pre 2000. They're getting in on the ground floor of something that has the potential to drastically change a lot of things. The number of opportunities someone gets to do that in their lifetime is very low.

The main thing about money, Bud, is that it makes you do things you don't want to do.

That means you actually wanted to do those things, even if you hadn't thought of them before. Money is inanimate, it can't force you to do things you don't want.

It also lets you avoid doing the things you don't want to do.

> The main thing about money, Bud, is that it makes you do things you don't want to do.

I'm not your Bud, Pal.

Every time there are talks about cryptocurrencies it seems like HN is stuck pretending that only Bitcoin exists and tech hasn’t moved since Bitcoin.

Most cryptocurrencies are clones of Bitcoin with the same problems. Ethereum is more interesting, though it still has some of Bitcoin's problems.

That's not true at all. Most cryptos aren't clones of BTC at all - they have wildly different consensus systems, interaction models, goals and capabilities. They all have their own issues and capacities, but saying they are all clones of BTC is like saying all existing programming languages are clones of FORTRAN and have the same problems as FORTRAN does. There's a huge variety, and we're still in the very beginning of it, nobody knows what emerges out of it and what would be successful and what would fail. It's probably decades in the future, maybe more.

Ethereum is also part of the shitcoins given the ridiculous amount of premined, but their marketing is considerably better.

why do you care about premining?

Most crypto are running on Ethereum at this point. A lots of useless Bitcoin clone indeed.

But a lot is happening still. Heck, I think some of those chains even serve a actual purpose, without the computational issues. Crazy right ?

As for Bitcoin, I hope the money in it trickle down to actual useful things. But, trickling down. Well, we know how it goes .

Can you list the problems here? For each problem I'll give you an alternative of Bitcoin that aims to solve it. I'm not joking.

That you for proving their point.

I run a Bitcoin lightning node on a small Raspberry Pi, and I use Bitcoin for p2p payments. So from my perspective, Bitcoin is both a great store of value and a good currency.

But since this thread is about an altcoin: Almost no other altcoin enables me to run my own fully-validating node on cheap hardware. And that is 100% a requirement I have since the whole point of crypto to reduce trust to a minimum. I don't know about the details of Chia, but as far as I know most coins with "smart contracts" built into the blockchain require expensive full nodes, so I personally do not see much appeal.

Check gridcoin.us , can run on most hardware even raspberry pi [1]

[1] https://www.reddit.com/r/gridcoin/comments/7mdnud/get_into_g...

> I just... do not understand how smart, even _brilliant_ people, the sort of people who know enough to make something like this, (...)

You can say the same about people in adtech.

Can you? It seems the adtech makes a lot of money, but also gives businesses, especially small businesses, far greater reach than they could achieve 20-30 years ago. Seems like an apples to oranges comparison IMHO.

If these people actually cared then I'm sure the same effect could be achieved without the shitshow that is advertising today. Also, ads likely cause overconsumption and thus impact the climate. And the big companies (some of them in adtech) gobble up all the wealth so small businesses are not necessarily in a better position.

My small business can advertise to people, much more than I would have been able to in the 90s. I don’t even begin to grok your point.

You can't prove that you wouldn't be able to advertise as much without adtech in its current form because 1. since the 90s the internet happened and 2. nobody ever did research into an approach which would serve society better. Which tells me that these smart people are after the money, and not in pursuit of a service that actually helps society.

For example: a simple switch on my browser to turn on/off all ads would be much more consumer friendly, and would still allow consumers to find products if/when they wish to find them. And this is just a simple example that just popped into my head.

Many people keep suggesting Chia will cause the next storage shortage crisis the same way current crypto-currencies are with GPUs. This is understandable given that the article describes this is already starting to happen. However, everything right now is all due to _speculation_. When there is liquidity in the market and people actually know the value of the coin, the economics might look very different.

Why? Because by the very nature of storage, it is infinitely more profitable to farm only unused extra storage than it is to buy new storage specifically for farming. I'll follow with a more detailed explanation below.

Since the marginal cost of using extra space on a hard drive you already bought is near zero, this should theoretically drive the farming rewards for Chia so low that it's not economical to buy a hard drive specifically to farm Chia.

In contrast, consider compute. If I try to use extra cpu power on an average computer or server to mine bitcoin, compared to the high end GPUs and ASICs this extra amount of compute I have within my average computer is so small it would barely generate any rewards. Given that using more CPU also causes more power consumption, it also costs more.

So putting this together, due to the existence of specialized/more-performant hardware and the fact that using more CPU/GPU incurs a cost, most of the people buying high end GPUs (or definitely ASICs), are buying it primarily for mining. In contrast, with space, since it's always near free to use storage you already bought, and everyone is on a near-level playing field, the market is only profitable for those with no sunk costs. At least, that's just a theory.

Just a thought, but who knows. We'll see in time.

Saying that Bitcoin blew the market for GPUs reveals that your knowledge is years out of date. Bitcoin is predominately mined on ASICs.

Viagra was originally developed as chest pain medication. Did it utterly fail, simply because a new use-case was unveiled?

Which begs the question, what is the new use case for Bitcoin? Some say digital gold, but only time will tell.

Out of all my bitcoin friends, no one I know owns an ASIC

Everyone mines on GPUs because they are cheaper and casual dual purpose (play games and mine afk)

Aren't ASIC normally limited to a single algo as well?

Your friends are most likely mining ethereum, not bitcoin.

> Which begs the question

Heh. I think you just unintentionally used this correctly by assuming there even is a new use case.

It was intentional. I am thinking probabilistically here, and the jury is still out on whether “digital gold” is the use case.

"Begging the question"[0] is a logical fallacy where you provide an answer to a question that depends on first assuming the answer is correct. i.e. a circular argument.

That was intentional?

[0] https://en.m.wikipedia.org/wiki/Begging_the_question

> It even seems doubtful to me that the people who developed Chia are the ones who'll make the money off it, if there is any to make.

Premined, just like Ethereum and all the other shitcoins.

I think many of us underestimate the power of FOMO and the desire to get rich quick.

The latter isn’t necessarily greed, lots of people just don’t have the good fortune to make a comfortable amount of money (as those of us in tech do) so are more susceptible to bubbles.

> lots of people just don’t have the good fortune to make a comfortable amount of money

The people who can imagine and implement a novel new type of cryptocurrency certainly do

Some people look at the lives of poor people vs the lives of ultra-rich people, and decide to get rich. Crypto is a decent way to get rich right now.

My understanding is that the way they intend to operate it, the coin will start out rather highly valued, won't have too much capacity to grow in value, and the number of coins that can be "farmed" is already known, and will reduce in the future. At the rate it's growing, it'll probably be at the point where it isn't worth it to buy new SSDs for this sometime this week or next. And it will likely rapidly get to a point where it isn't worth it to buy HDD soon thereafter.

I don't understand how a forum called Hacker News is so dismissive of an invention that should be right up a hacker's alley. Innovative, technical, cryptographic and decentralized. Cryptocurrencies feel like something straight out of a cyberpunk novel.

Because it doesn't work.

I wish it worked! It does feel like something out of a cyberpunk novel. I loved "Crypronomicon." There's nothing punk about Bitcoin anymore, though. IBM, the least punk organization in the history of mankind, wants to sell you a blockchain. The promise of Bitcoin is the idea that individuals, not large financial concerns, have the power. Coinbase just IPO'd for an eye watering amount of money on the premise that people who want to buy and sell Bitcoin are probably going to have to do it through them. What's less punk than an IPO?

> The promise of Bitcoin is the idea that individuals, not large financial concerns, have the power.

Bitcoin is not the only cryptocurrency out there.

Besides, the fact that coinbase just had an IPO doesn't mean they control bitcoin, I don't need them or anyone else to move around cryptocurrencies however I want.

That’s a weird argument, because the big guys are trying to get into cryptocurrencies doesn’t make it less punk.

it literally does mean that. Punk is a rebellion against the system, which is made out of the big guys.

Maybe the rebellion is happening and Bitcoin&friends are slowly winning

Oh - just like punk did?

>Because it doesn't work.

Bitcoin works well in plenty of ways. Bitcoin just doesn't work very well as a currency, but that ship has sailed and it's ok.

> that ship has sailed and it's ok.

I've seen this comic, and the dog is melting in the next panel.

That's the weirdest take on why blockchain doesn't work I've seen so far. The entire point of these projects is that people do whatever they want with them. Monero exists if you want a purist approach and so does ergo.

> Cryptocurrencies feel like something straight out of a cyberpunk novel.

And that is exactly why I dislike it. At least among ones that I encountered cyberpunk is strongly dystopian.

> I don't understand how a forum called Hacker News is so dismissive of an invention that should be right up a hacker's alley

I hope that Chia dies and falls on its face before SSD and HDD will repeat story of GPU.

I like to use this resources for actually useful things, not gambling.

> decentralized

So far BTC-like stuff is centralized, just in a novel toxic way.

> At least among ones that I encountered cyberpunk is strongly dystopian.

Back in the day when we were sitting in dark basements with big Unix beards drinking Jolt cola, we all dreamed of living in a crypto-anarchy, becoming cyborgs with chip implants and fully submerge in the Metaverse. I guess times have changed.

As someone who started reading cyberpunk as a young teenager this is definitely the future I imagined, just with different names.

Cryptocurrency, the internet, mobile devices, huge powerful corporations, Anonymous, black hat hackers, hi tech protests like DDoS, electric vehicles, shady IT security companies, Moxie Marlinspike, billionaires going into space...

Cyberpunk is corporate oligarchy, and we are getting there.

Glasses is as far as I am happy to go with cyborgization. Maybe implant in case of total hearing/eyesight loss.

I know how buggy software is, I would not feel safe with it directly coupled to myself.

> living in a crypto-anarchy

I know how well communism went, I hope that next radical society reengineering will happen sufficiently far away to not impact me.

Wasting massive amounts of mostly carbon-producing electricity as the planet dies to do absolutely nothing productive is a pretty cyberpunk evil megacorp thing to do, agreed.

> a cambrian explosion in third parties

Which is all that is needed to create a fair bargaining position for the consumers.

Theres nothing wrong with exchanges on top of bitcoin, in fact, that will be the solution for the tx costs and energy problems. But HN is too stubborn to get it, remember, if you do the opposite of what HN youd become rich :-)

Bitcoin did solve the problem it alleged to be solving: money which is not controlled by state, and a system which us pretty hard to seize.

It does allow you to have an "account" outside any jurisdiction, and allows you, given good opsec, to transfer huge amounts in transactions not controlled by any state. Like, millions of USD.

Bitcoin is obviously useless to pay for daily goods, like a coffee. But gold would be also impractical for such a case.

Crazy speculators pumping the value of bitcoin may look unreasonable, but again, for those who have millions (and maybe even billions) in bitcoin, they are weirdly... useful.

Read the Bitcoin paper:


It does not, at any point in there, say that the benefit is to avoid state control. It talks about how the cost of requiring a third party in transactions discourages microtransactions, for instance (there are other touted benefits as well). But it's not an anarchist manifesto. And all of the benefits of Bitcoin listed in the original paper require it to be a useful medium of exchange. The Bitcoin paper posits that it should be useful for transactions _much smaller_ than a cup of coffee. It isn't. It so obviously isn't that even Bitcoin advocates can admit that.

Okay, so maybe you think that it's okay that Bitcoin doesn't meet its original goals given that it meets these other goals. But the goals you state, the lack of state control and the difficulty of seizure, are... like, Bitcoin does not put you the person outside of the state's law enforcement mechanisms, including the taxation people. Ironically, if Bitcoin was actually useful for small microtransactions, like the paper says it should be, that might actually be useful for avoiding state control. It's not! And nothing about Bitcoins makes them hard to seize if you are within the physical jurisdiction of the state.

The original point of Bitcoin was most certainly to avoid state control. Read some of satoshi's forum posts or this: https://en.bitcoin.it/wiki/Genesis_block

On Bitcoin you can do micro transactions using the lightning network. On Ethereum you have a plethora of layer 2 scaling projects that are gaining traction as we speak. Other networks like Avalanche achieve high throughput and low finality out of the box. It seems to me that there are plenty of things to be excited about in the space but you have to keep an open mind instead of being a stickler about the verbiage used in the original Bitcoin white paper.

Random fun fact: the EU Investment Bank just did a 100 mln bond issuance on Ethereum.

The financial rails of the world will be moving to crypto asset networks, slowly at first, and there is nothing that can stop this shift.

"But documentation says it isn't for that..."


Hi I work in the space email me if you want to understand my motivations

I'm not asking for a pitch deck or a prospectus. I'm asking for a little bit of humility and learning from history. Like, credit where credit is due, the original Bitcoin paper posed a real problem (real in the sense that solving it could actually be net beneficial for society as a whole) and proposed a novel, clever solution to it. That's great! And from the perspective of when the paper was written, it was hard to see the ways in which Bitcoin would fail to actually solve the proposed problem. But now, not only can we see how Bitcoin failed to deliver on its initial promise, we can see how someone who is clever enough to come up with Bitcoin could fail to grasp the implications of it.

Like, from where I sit, Chia looks like it's trying to solve the unintended consequences of Bitcoin, which is a tacit admission from the Chia people that Bitcoin has those problems. It also looks like Chia has an easily foreseeable set of unintended consequences as well: doing to SSDs what Bitcoin has done to GPUs. It doesn't seem to me like the Chia people are worried about that, and I don't know why.

> what Bitcoin has done to GPUs

I understand your argument, but Bitcoin hasn't been mined on GPUs since 2011/2012. Broaden your argument from Bitcoin to "proof of work schemes" more generally, and from GPUs to "profligate (mis)use of resources", and it will land more effectively.

If you are against cryptocurrencies more broadly, inclusive of proof of stake and other more recent approaches that aim to address the resource consumption issue, then that position didn't come across clearly.

I don’t think that is a valid defense.

Bitcoin started the whole crypto craze, and etherereum wouldn’t exist without bitcoin.

And the eth (and who knows how many other coin) crowd certainly does still mine on gpus.

> And the eth (and who knows how many other coin) crowd certainly does still mine on gpus.

And is actively on the path to proof of stake.

> proof of stake

Is this past the point of basic research?

Cold fusion has been on a path for 50 years.

It went live December 1st, 2020. There are over $10 billion currently at stake. You can track the chain here https://beaconcha.in. Please don't spread misinformation. https://cointelegraph.com/news/eth-2-0-confirmed-for-dec-1-l...

The level of ignorance and yet strength of belief here is astounding

It's all right. For every unjustified cryptocurrency sceptic such as myself, there are 10000 cryptocurrency zealots pumping and dumping every day. Hodl, right?

I am not against proof of stake, exactly. But proof of stake has been around for a long time, and is a very small part of the crypto market right now. I don't really understand why, but until proof of stake is able to get adoption, I don't think the existence of proof of stake is a very good defense against the accusation that cryptocurrencies are wasting resources.

Proof of stake is even more of a Ponzi, with the rich in a given cryptocurrency getting even richer with time.

At least for miners in a PoW system, there’s a pressure for some liquidity by miners selling to cover the mining costs.

Recent related threads:

Thoughts on a “HDD shortage” from an industry insider - https://news.ycombinator.com/item?id=27008295 - May 2021 (106 comments)

SSD makers start warning that mining products like ChiaCoin will void warranty - https://news.ycombinator.com/item?id=27001848 - May 2021 (381 comments)

Chia surpasses 1 exabyte of hard drive space - https://news.ycombinator.com/item?id=26975734 - April 2021 (151 comments)

New Chia coin crypto with Proof of Space may create shortage of storage device - https://news.ycombinator.com/item?id=26879278 - April 2021 (19 comments)

Chia Blockchain mainnet successfully launched today at 14:00 UTC - https://news.ycombinator.com/item?id=26515025 - March 2021 (14 comments)

I am... Suspicious of all the chia articles popping up on HN. I have not seen the same kind of buzz happening in other crypto spaces about it. And I am also suspicious about the claims that chia is the one that is such a threat to data storage supply, considering there are other crypto currencies with a similar approach.

Overall, I am inclined to say that this is a submarine and HN is falling for it.

Regardless, more to the point of the article, the argument and moral panic rings hollow. Chia just incentives other people to store data for people who want to store data. The place, overhead and mechanism of data retrieval is different, but, still, someone wants to store the data there and they pay for it. So how then is this more of a threat than cloud providers for example?

Later Edit: I mistook Chia as having a similar approach as Filecoin, but, on closer inspection, it seems that they do not store data people want they just store data to prove it's stored and use that to distribute mining rewards. So I was wrong at this end. Still, I maintain my opinion this is a submarine.

Your comment broke the site guidelines, which ask you not to post insinuations of astroturfing, shilling, etc., without evidence. Multiple articles showing up about a topic isn't evidence of abuse; it's perfectly normal. Any slightly unusual angle like the hard-drive thing is more than enough to explain a small flurry of articles.

"Please don't post insinuations about astroturfing, shilling, brigading, foreign agents and the like. It degrades discussion and is usually mistaken. If you're worried about abuse, email hn@ycombinator.com and we'll look at the data."



We have that rule because it's extremely easy to fall into imagining this sort of thing based purely on a few data points one dislikes. In the absence of evidence, that's likely what's happening. Moreover, the submission histories of most of the accounts that posted those threads is evidence against what you said, and that information is public and trivial to check.

Apologies dang, you are right. Won't do this again. I really appreciate the work you're doing here.

> Chia just incentives other people to store data for people who want to store data.

This is untrue, Chia requires wasting storage space to store pointless garbage.

I learned about Chia last week by googling "hard drive shortage" to figure out why suddenly hard drives prices are spiking where I live.

Definitely not a submarine.

Hard to say, really. It could be organic, but then, it is in the interest of Chia to spread the stories about "hard drive shortage" - like all crypto schemes, their best user base are people who will see an article about cryptocurrency causing a social-wide issue, and read it as an opportunity. The people who see a disaster and their first thought is about how to profit off it.

I'm a bit worried it's a self fulfilling prophecy:

"People are buying hard drives to farm despite price increases, it must be profitable, maybe I should buy one too..."

Could you or the parent commenter define "submarine" in this context? I searched and didn't find anything relevant.

It's a reference to a pg essay: http://www.paulgraham.com/submarine.html

The part introducing the term is at the beginning:

"Why do the media keep running stories saying suits are back? Because PR firms tell them to. One of the most surprising things I discovered during my brief business career was the existence of the PR industry, lurking like a huge, quiet submarine beneath the news. Of the stories you read in traditional media that aren't about politics, crimes, or disasters, more than half probably come from PR firms."

It means an advertisement disguised as a news article.

> I have not seen the same kind of buzz happening in other crypto spaces about it

If all the usual crypto charlatans aren't "to the moon"-ing it, that is a strong indicator of value for me.

The reason it is getting play on HN and other tech focused outlets is because it is an interesting new project from the designer of BitTorrent, who clearly knows a thing about designing protocols.

> Chia just incentives other people to store data for people who want to store data.

I think you are confusing Chia with Sia

You are correct and my later edit was added to address this.

Not sure what is the proper etiquette if I realize my initial statement was false is some way, add a new paragraph addressing the issue or just edit the original post to reflect updated stance?

New paragraph could add confusion

Editing the initial post might make it seem like I'm trying to look smarter than I actually am.

So in other words, there's a campaign (centralized or not) to pump Chia on HN? :(

It's not that many articles, and they're not all favorable. Please don't break the site guidelines like this.


> they're not all favorable

That doesn't matter. It still generates buzz. People will pour resources into a crypto-currency even when it's established up front as a joke.

> don't break the site guidelines like this.

I take it the rule was triggered by the word "campaign":

>> So in other words, there's a campaign (centralized or not) to pump Chia on HN? :(

I pretty clearly indicated that I don't think it's necessarily a coordinated thing. It's more of a phenomenon than a "brigade" or "astroturfing", but what should we call it? What should I say? How should I feel about a disturbing trend that you've pointed out?

Furthermore, you've already shared your data, so there's no reason for me to email you about it per the guidelines. I appreciate the vigilance and hard work, but I think we can agree you're mistaken this time.

I've also seen it in various "technology" subreddits.

I'm suspicious that HDD and SSDs manufactures and/or merchants have been hyping it after getting jealous with how much money GPUs have been raking in.

this incredibly viral "chia will cause a shortage of hard drives" meme (true or not) is a huge gift wrapped win for the coin's promoters.

every impression of this meme that results in someone's decision to avoid chia (which, had they never heard of the coin in the first place, would have been the outcome anyway) will be offset by a half dozen others who will spot a quick 10x return opportunity for themselves, even if they recognize that the long term result is a net negative for everyone else.

it's Streisand Effect rocket fuel.

Bingo. The mods are asleep at the wheel on this one.

It took me a while to actually understand how proof-of-work mining works because seemingly every article one the subject gives too abstract of an explanation. Turns out it's not that complex if you understand what a cryptographic hash is. Just take a block (a list of recent transactions) and take a SHA-256 hash if it over and over with slight variations until you get a hash that starts with enough consecutive zeroes.

I'm having the same issue with understanding proof-of-space. Could someone give a technical explanation for those of us that understand basic cryptography (but aren't professional cryptographists)?

Proof-of-space is essentially stored Proof-of-work. You are pre-filling your drive with many cryptographic hash's. When a challenge comes in the software will do a quick check to see if any of your plots could possibly have a proof (a hash with sufficient zeros for that challenge). If any of your plots pass this filter then the plot is looked up on the HDD to check if it does indeed contain a proof. The difficulty, the number of zeros needed for a hash to be considered a proof, adjusts as the network grows to ensure a steady supply of coins. With proof-of-work you are printing lottery tickets every time a challenge comes in, then throwing them away and starting again, with Proof-of-space you are printing your tickets once and then storing them for years. There is extra layers, Chia is actually proof-of-space-and-time, so that you can't win by faking proof-of-space with proof-of-work.

For someone uninitiated: what if you had high computing power? Instead of scanning your HDD, can’t you generate these hashes on the spot?

Yes, you can. That's called a grinding attack. There is mitigation in the proof-of-space construction for exactly that. That is also the reason that they will have to move to more expensive proofs as time goes on, and hardware improves.

> with Proof-of-space you are printing your tickets once and then storing them for years.

So then where's the issue with SSD wear, and wouldn't it be more interesting for people to have spinning drives (for the better ratio of GB / $)?

> There is extra layers, Chia is actually proof-of-space-and-time, so that you can't win by faking proof-of-space with proof-of-work.

Is it because of this? So you'd have to print your lottery tickets often enough that a faster drive may help you?

Yes you absolutely want to be farming with spinning drives for the better price to capacity ratio as you mention! (Hopefully in a few years this will change and SSDs will overtake spinning disks in this metric)

The most efficient way to start your farm is to build your plot with a temporary drive (an SSD) and then move it to a destination drive (Spinning HDD) where the actual farming takes place.

The building of the plot is what does heaps of writes and reads and puts the wear on an SSD.

You absolutely can plot directly to Spinning Disk however it is slow and you will really be trashing the heads if you try and plot multiples to the same disk.

The solution is an enterprise SSD with proper endurance. If a consumer SSD doesn't state it's TBW then you can be pretty sure its not very much.

> wouldn't it be more interesting for people to have spinning drives (for the better ratio of GB / $)?

You do trade off for a slower read rate. If the rate of requests coming in is fast enough, your spinning drive/s may not be fast enough to go through the indexes for each one.

Spinning disks are more than fast enough for farming. Many use network connected spinning disks. Tape drive is not fast enough however.

I'm probably missing something here. CPU/GPU cycles are irreversible, you can't get them back after you have used them. Storage space is reversible, you can get it back at any time by deleting what has been stored. I'm sure they have a way of detecting and preventing this, but wouldn't that imply that constant network connectivity is required to check that the storage is still allocated?

Think of proof of work, where you create loads of hashes for each challenge that comes in, once people came along with ASICs they could create so many more hashes in the same amount of time! The chances of finding a proof with an ASIC is directly correlated with how much more hashes you can create than a CPU in a given time. With proof-of-space you are storing the hashes on disk. The more hashes you have stored the more chance you have of finding a proof. You can delete those hashes at any time if you would like (to use the space for some other purpose) and the only effect is that you will just be lowering your chances of finding a proof for any future challenges. You don't have to be connected to the network all the time, however you will only have a chance of winning Chia while you are connected to the network. The positives of this approach: 1) Once you have plotted you drive and the plots are farming you use very little electricity (many people farm with Rasberry Pi's) 2) You can utilize unused storage on drives you already own and then as you need that space for some other purpose you can delete the plots. 3) Your chances of wining are directly correlated with your share of the network space ie. someone who plots twice as much has you has on average twice the chance of wining a block, no more, no less. 4) Building a special ASIC for proof of work serves no useful benefit beyond what it was intended to do. Whereas if someone was to build a cheaper more efficient form of storage, well that would be a benefit to everyone!

One way is using zero knowledge computational integrity proofs. They allow me to prove to you that I know of pre-image of some hash without revealing what it is.

Say the network has seen 1TB of data before with hash H. It can generate a random number as a challenge, and request a miner to prove they have access to the 1TB worth of pre-image at the same time it has access to the random challenge. The only way for a miner to compute such a proof is to have the 1TB of data stored somewhere.

What you describe is an instance of the Hashcash Proof-of-Work. There are many other types of PoW, which don't amount to just computing a hash function [1].

[1] https://cryptorials.io/beyond-hashcash-proof-work-theres-min...

There are multiple flavors of proof of space and they seem to be based on quite complex math to the extent that a simple explanation may not be possible.

There always is a simple explanation:

Resulting from (1) removing irrelevant details (details crucial for an implementation, but irrelevant for an initial mental model) and (2) making use of analogies built around already known concepts.

"Complex maths" is never an excuse to not find that explanation.

Disclaimer: I am a pure mathematician.

I don’t think I have ever heard of proof of space, is it the same as proof of stake?

Completely different things.

In Proof of Stake, the network (roughly!) allocates rewards for validating transactions based on the amount of coins you have, i.e., your stake in the network.

In Proof of Space, the network allocates rewards based on you proving that you are really storing the data you were supposed to store. If you fail to produce a proof for some data item the network expected you to have stored, you lose the reward.

More info:



> In Proof of Space, the network allocates rewards based on you proving that you are really storing the data you were supposed to store.

Sort of? But you're usually not really storing data for proof of space. That's a different kind of system.

Yeah, that is called proof of storage in Filecoin:


Another important wrinkle is that Chia’s proof-of-space needs you to prove you’ve stored the right cryptographic noise - rather than, for example, any useful data.

Another proof of waste shitcoin. Just what the world needed.

They are different. Chia isn't Proof of Stake because how many XCH has no impact on your odds of winning more.



Sorry, dropped a couple words: "how many XCH you have has no impact"

No - it’s a bit more like preprinting big ranges of lottery tickets, that require giant look-up tables moreso than raw computation.

So it requires the dedication of some hard/rivalrous real resource, storage space, other than either strictly energy/CPU (PoW) or the protocol’s own tokens (PoStake).

See the Chia ‘Green Paper’ for more details.

If someone could create a cryptocurrency using proof that require people to pick up garbage, the world would be garbage-free in a month.

People would litter just to create more garbage to “mine.”

We had (have) this here in Belgium with a very real case. Separate collection of plastics and the creation of plastics collection and recycling (organizations)companies were introduced for environmental improvement.

Unfortunately, these have turned into the biggest plastic packaging lobby and have caused an explosion in the use of plastic packaging in retail. More plastic use is unfortunately a net benefit to those whose economic/financial incentives depend on it.

I would have thought that Belgium had learned from its past [1] to think carefully about providing incentives with potentially (horribly) harmful side effects.

[1] https://en.wikipedia.org/wiki/Atrocities_in_the_Congo_Free_S... -- "... The collection of hands became an end in itself. Force Publique soldiers brought them to the stations in place of rubber; they even went out to harvest them instead of rubber ... They became a sort of currency ..."

There are many examples of perverse incentives.

As for the reference you provide, note that the atrocities were at the time when private companies using private militias ruled the region that was the property of a private holding of Leopold of Saxe-Coburg-Saalfeld who at the time also was the King of Belgium.

It was the Belgian state learning of these atrocities that prompted Belgian state to annex the region and bring these to a halt.

It could be argued that indeed the public sector should have learned not to trust the murky amalgam of old and new capital

I live in Belgium and I never heard of that. Do you have any source? I’m interested.

Rats of Hanoi - explains the exact thing you are saying :) https://www.atlasobscura.com/articles/hanoi-rat-massacre-190...

yep, we would then have to deal with "flash trash" appearing all over the place.

Carbon coin. Proof of removing carbon from atmosphere.

Proof of effective taxes paid as per your tax bracket

Interesting. There are a bunch of variants, but it seems like there are three main ways of handling crypto.

Bitcoin is proof of work, which of course requires a lot of electricity.

Some coins, like Neo/CutCoin are proof of stake, which basically let you stake some money, which I guess you can't spend and use that to handle transactions, so instead of running mining rigs, you just stake the money and earn a return. Ethereum is planning to move to proof of stake. Kinda like how your FDIC insured bank can loan your deposit out.

Chiacoin, is proof of space and time, IE proof of assigned storage through the use of stored data, plus a verified delay function (slow to calculate, fast to verify hashing basically). The latter part needs only one processor, so whoever calculates fastest is accepted, so you aren't running a bunch of miners racing to benefit.

I'm not sure that I see the benefit of proof of space and time except in the case of things like filecoin, where you are basically paid for storing stuff on your hard drive without changing it. Chia coin is, if I understand things correctly, just sticking random numbers on a SDD to "farm" them. I mean the VDF stuff is cool to be sure, but it seems like in the case of Chia it's just wasting memory space with random numbers rather then incentivizing storage of existing data.

Is there a better system than proof of stake, in terms of just not being absurdly wasteful in terms of energy or storage capacity?

For the purpose of responsible disclosure, I do have less than $20 bucks of ethereum atm, and that's the only amount of any coin mentioned above. I have no investments in the companies behind any of the above coins.

I am fascinated by the POS coins you chose to list. Why list neo and cutcoin over Cardano? I've never heard of either!

Honestly? Because I'm ignorant.

I just quickly checked DDS for a list of actual proof of stake coins, since I knew about proof of stake only as something that was happening eventually for ethereum, picked a random result on the first search results page, and then picked two choices randomly from the listed coins, using a rng.

I figured if I picked two ignorant options people would probably say that, but I'm comfortable being ignorant. Now I've heard of Cardano.

And Cardano's Proof of Stake is groundbreaking. I don't know why the commenter chose to list those coins.

why is it groundbreaking? how does it work vs ETH try at PoW?

All of these proof of X are just mechanism to find the next “leader” who will be in charge of proposing the next page of the ledger. Proof of work has proven to be wasteful, I personally don’t really see a great usecase in proof of space, proof of stake is basically the way to go. That being said, you have leaderless consensus protocols (avalanche) and even consensusless protocols (AT2).

> This “farming” process differs from mining in the sense that you don’t need specialised equipment to participate; anything from a laptop to a smartphone can be used to farm Chia. The more free space you have, however, the more likely you’ll be rewarded.

This seems like an ad for Chia disguised as a hit piece. It doesn't help that the author's only other article is trying to pump Dogecoin.

Field Notes, Terran Expedition 6, Chief Archivist Blaxarll:

"Terran life took a sudden dramatic shift after the fall of Hominid dominance. The turning point appears to have occured when the sacrificial demands of Blockchain worship spiraled out of control after fuel-based offerings were supplanted by info storage based offerings. This, in and of itself, was not terribly harmful, but the model was used by the Cult of Aquaricoin. The so-called Water Miners enshrined a marketplace for non-fungible indulgences depicting popular figures. This was backed by a Proof of Potable Water algorithmic catechism, and swiftly resulted in total extinction for the species.

It is this researcher's opinion that we are fortunate the Terran Hominids had only just discovered space travel, for they very well could have extinguished the resources of our own homeworlds for such religious decadence."

this is one of the best things i have ever read thank you

To be honest I kind of wish someone would build a proof-of-bandwidth crypto. Miners would overtake data centers with cheap transit, driving the build out of more long-haul network. Some would then move to different neighborhoods just to get FTTH access, enabling more incentive for last-mile build out.

Orchid is a decentralized crypto market for bandwidth:


You can run a node and sell bandwidth to people who want to proxy traffic through your node

Curious about this idea. What happens if illegal stuff happens? Assuming bandwidth = internet.

Side thought is renting out your mobile device for peripheral testing. That's something I was looking for recently video/audio specifically.

It looks like you can't become a "miner" (sell bandwidth) without becoming a partner. All of which are existing VPN companies - who probably have their abuse handling well ironed out at this point.

(I am the lead developer of Orchid.) Can you tell me where you read that bit about "without becoming a partner"? It is sort of correct--like, I can see how that argument could be made--but not really true in any fundamental sense... it certainly isn't true enough that I would have personally said it to someone, as if you want to provide bandwidth you absolutely can, it is just highly unlikely anyone will buy it, as they would have to explicitly leave the only-loosely-fenced-off garden we provide by default (but I also don't work on the current UI or the documentation, so for all I know this is extremely hard to know). Really I am just curious what specifically you saw, as you seem to have come to that conclusion pretty quickly for something I only consider half true ;P.

Definitely, though: you join a market like this because you know what you are doing and what you are getting yourself into; I personally could not in good faith recommend that an end user sell access to their home Internet connection.

The only "documentation" I was able to find researching how to mine it was a Reddit thread suggesting that you have to stake a large amount of OTX to in turn be able to sell bandwidth and in turn make OTX. Previously in my hunt for documentation I stumbled upon the partners page with 4 VPN providers featured at the top of the page and the whole thing sort of clicked. I did find a Github mono-repo for what looked like the end user clients (at least from what I was able to find in the docs directory).

If my statement was incorrect (which by your half-acknowledgement I don't believe it was), then I am sorry for misrepresenting your project. A healthy dive into howtos for miners, and unwalling your garden, would clear that up.

But I do quite like the closing jab.

(I didn't read this comment before as I simply forgot about this thread at the time; I imagine you won't see this reply, now, though ;P. And, sadly, I can almost guarantee I won't see a reply to this message if you did leave one, for reasons even deeper and more compounded.)

FWIW, I feel like there is nuance that is being lost in your sentence about whether your statement was somehow either incorrect or correct. My half-acknowledgement of the premise would to me indicate that neither absolute on such a statement could be "correct"?

The server--which is in that exact same monorepo you found, it being a monorepo and all ;P--can be easily compiled (or merely downloaded from the GitHub releases) and it doesn't take too much messing around with it to run it... as a developer (not an end user) You certainly don't need to be a "partner".

The client, also, will happily connect to you... if the user "opts our" of only using a list of people that happens to right now only include "partners". This is where I appreciate that the software--right now--has not a "walled" garden but a lightly-fenced garden: the difference is important, as anyone who wants to can go play outside the walls. This isn't like Apple or Facebook: the term "walled garden" simply doesn't apply.

The "half-acknowledgement" is, of course, that if not many people are playing outside the garden then you are likely going to be lonely. But as there is no wall, the project absolutely let's you sell bandwidth there... there just aren't many buyers. You see how that's different, right?

In some sense, being allowed to do something isn't a guarantee that it is a good idea. Imagine a world where anyone can make a restaurant, but almost all customers search for food on Yelp... is it true that only Yelp partners can sell food? No: anyone can sell food! But I appreciate--in that it isn't black and white--that it isn't exactly true that anyone can succeed at selling food.

And this is the rub, right? It is like the difference between a right to sell something and a guarantee of customers, which are related in a certain sense but aren't even slightly the same :(.

This is then why I wanted to ask after where you came to the quick idea that, with Orchid, one """can't become a "miner" (sell bandwidth) without becoming a partner""", as that statement isn't correct (but I appreciate neither is its opposite!!) in a way that makes me sad, as there are a lot of projects for which it is correct: where there are walls of authentication and permission or softwares that must be licensed... and I am very proud that Orchid isn't that?

That said, I feel like--with your emphasis on documentation (not in the passing way I mentioned it, which would include high-level discussion, but some expectation of "here is how to do it")--you are somewhat "moving the goalpost": instead of analyzing whether the software does or allows something, you are asking whether the project encourages something... and that is just very very different to me.

The reality is that--partner or not--you will probably be unhappy selling bandwidth as there isn't much traffic being purchased on the system right now, and I would rather not guide people down paths that will make them unhappy... particularly when I know that, if it were actually profitable, there would be a large cottage industry of third-party guides telling you how to do it ;P.

Why would anyone sell their bandwidth then?

Potentially for more favorable routing? There are some other potential niche things like maybe load testing, or doing load testing from weird IPs. It's not entirely clear to me either without having some kind of storage (in which case it would make sense as a CDN).

"We believe that by harnessing the power of blockchain, we can offer users a better VPN experience"

So, is deleting articles the new deleting vowels? facepalm

Assuming Orchid even harnesses blockchains at all--I built it and notably don't have any clue what the marketing people meant when they wrote that sentence (I hate that website)--I agree that "blockchain" should be plural.

Except what happened to GPUs would happen to bandwidth. We'd all be fucked.

Exactly. The proof of bandwidth would probably require you to waste that bandwidth for mining.

> To be honest I kind of wish someone would build a proof-of-bandwidth crypto.

https://pkt.cash/ exists "built to incentivize the growth of infrastructure," and "PKT Network is designed to decentralize internet access around the world by enabling anyone to become an ISP." [0]

A key component of the PKT Network is cjdns (https://github.com/cjdelisle/cjdns).

[0]: https://pkt.cash/PKT_Network_v1.0_2021.02.01.pdf

If I were to own a transit for the purpose of profitting from proof-of-bandwidth, ignoring the regulation and law for the communication infrastructure, would I?

A) Let everyone have their fair share of bandwidth

B) with DPI, choke other miners so I'm relatively better in the whole miners.

Also, bandwidth will be wasted just for the sake of proof-of-work instead of traditional work that have real values on its own.

Folks here might be interested in Chia's smart contract language: https://chialisp.com/

Basically the designers saw how easily many Ethereum/Solidity contracts were hacked and saw that stateful side effects were the main culprit so they made their language have no side effects! Pretty cool.

I've been wondering if non-turning complete languages might be the way to go for many smart contracts, which often resemble vending machines more than what general purpose languages are used for. Those should be captured by pushdown automaton or even a finate state machine and the corresponding languages (but except from regex, I'm not really aware of any..).

My CS knowledge got really rusty, but I think it should be possible to put tight bounds on the computational cost of these, as well as statically guarantee that certain states cannot be reached, both of which would be very valuable in this case.

> they made their language have no side effects

Yeah, I doubt that

It's not that they "made" their language have no side-effects, it's more that they made a choice to use a lisp/functional design for the language which gives the advantages of that family of languages languages... advantages like limited side-effects and formal approaches for proving "correctness" of programs. It's a really clever design choice IMO for a smart-contracts language.

But what you really want for smart contract is a language with integrated formal verification. The ability to formally prove some properties of your smart contract is much valuable than functional vs imperative paradigm.

Functional programming might makes it easier to detect flaws in your programs, but when there is billions at stakes and no recourse in case of a bug, you really want your program to be flawless.

Tezos has a similar pitch (having a better language that is less bug prone) and is a ghost chain now.

I asked this recently but the topic died before I got many answers. Basically, I’m not a cryptocurrency denier. I just don’t understand it. Do people want cryptocurrency because they want to spend it on stuff? Or do they want cryptocurrency because they want to sell it for other currencies? In other words, at what point will a cryptocurrency stand on its own as something that’s worth a pair of shoes? Instead of being something that people invest in to eventually sell for dollars with which they they buy shoes? Again- I know I just don’t understand. My original question:

1 point by theshadowknows 4 days ago | parent [–] | on: The Last Days of Satoshi: What Happened When Bitco...

Bitcoin and other crypto/alt currencies are so interesting to me. Partly because I do not understand them: If I mow your lawn and you pay me in the locally accepted fiat then I can use that fiat to buy hotdogs and fuel for my lawn mower because the gas station itself will accept that fiat. The gas station will accept that fiat because its vendors, creditors, and employees also accept it. And the chain goes on. If however I mow your lawn and you pay me in bitcoin I'll need to find a gas station that accepts it. Many likely exist. But, how many of their vendors, creditors, and employees also accept it? Some, no doubt, but I'd wager not all would. So, even though these currencies no doubt represent massive potential they're still somewhat limited in terms of adoption. Obviously this is becoming less true over time which I think is a good thing - competition isn't something fiat currencies tend to have to deal with, but they sure could use it. My real misunderstanding is 'why' people accept these currencies in the first place. In other words - people accept a United States dollar because it is worth one United States dollar. However, it seems that most people accept bitcoin (or bitcoin fragments) not because they are worth some percentage of bitcoin but exactly because they are worth some amount of United States dollars. Is that true? Is the primary reason that bitcoin has become more popular that an investor can spend 10,000 fiat on it - wait a year - and then extract 15,000 fiat from it?

> Do people want cryptocurrency because they want to spend it on stuff?

Some of us did. Last month marks 10 years since I was introduced to the community and registered on the bitcoin-otc. We bought and sold all sorts of random crap (food items, laptops, gadgets, one regular even paid people bitcoin to have them buy him lunch. A brief time existed where some of the more idealists expected bitcoin to replace Paypal. That obviously didn't happen.

Starting with Satoshidice gambling the network performance started to crap up and fees began to climb. Eventually it became worthless for any sort of fast/cheap transacting and everything centralized onto a few exchanges. Every asshole in the world has proposed ways to improve this issue, but with all the money sunk into bitcoin and a lot of arguments we aren't in a better situation today.

Now every naive person from 20-some year olds to 70-year olds are hitting me up and asking what I think of NFT's and the shitcoin of the week because they saw some heads on Bloomberg gushing over it and I'm just tired of it all.

What about BCH? It seems like a reasonable solution still in the running

BCH solved one of BTC's fatal flaws (transaction rate) but all the other fatal flaws like volatility, irrevocable transactions, being your own bank, etc. are still there.

Hm, but if we solved for those issues, we would probably just have USD. I thought the appeal of a fixed supply, non-double spend asset was to be able to keep up with previously exclusive asset inflation .

That depends whether you want Austrian digital cash or Keynesian digital cash.

Lol, it’s the first time we’ve had an Austrian money, and the market seems to like it.

The last 3 things you listed are benefits, not flaws. At least volatility is positively skewed.

I love positive volatility. Don't you want to make money?

No such thing as free guaranteed money, all of this is just a fancy way to gamble.

A lot of it, certainly not all of it. If something better exists, we’ll probably move over to it, and the transition period could be turbulent.

It's more like the promise that you could spend $10,000, wait five years, then retire. Bitcoin's historical >10,000x returns have triggered massive amounts of envy, greed, and FOMO.

Disclaimer: I'm a Bitcoiner, which seems to be an unpopular stance here. I'll do my best to explain some of my motivation and answer to your questions.

Q1: Why would people accept these currencies in the first place? I want to hold Bitcoin because of its monetary policy. It has a fixed, capped supply of 21 million BTC, whereas fiat can be printed at will. This property is what fuels the current hype, and the fact that central banks around the world are inflating money supply as we speak certainly helps the narrative. So for bitcoiners, money saved in fiat is likely to lose buying power over time, but in Bitcoin we see, on average, a steady increase. Since I see Bitcoin as a superior form of money, I would certainly prefer getting paid in Bitcoin than in fiat.

Q2: Why do people that accept Bitcoin still denominate prices in USD? Bitcoin is far from being an established currency, and it is still very volatile (though far less than a few years ago). Most of the population are not bitcoiners. The US are denominated in USD. So, until Bitcoin is both more stable and more popular, it is the most convenient to do what we're used to and denominate things in USD.

Q3: Is speculation the primary reason for Bitcoin's rise in popularity? We're in the middle of a bull run right now, so probably? But if you zoom out and look at the price evolution over 5 years it looks different, and there is a steady growth.

Q4: Do people just want to increase their USD? ("spend 10,000 [...] and then extract 15,000 fiat") Some do for sure, but I don't see much appeal in exchanging hard money into money with arbitrary monetary policies. Bitcoin tripling in value does not change the initial motivation I had to buy it, so why sell? Of course, if I want to buy something with btc then that amounts to selling some. But my motivation to be in Bitcoin is to protect my time, the most limited resource I have in my life, from inflation, and that does not change with price.

The way I understand bitcoin is the following:

1. Main advantage of it is that it can't be stopped.

2. Since it can't be stopped - it allows to evade any regulation from the government or the bank. If it replaces banking system - you can move around the globe with your capital and spend it the way you like. This is not the case at the moment

3. Bitcoin could become a currency if it was fast enough and fees were low enough. Which was quickly discovered to be not true since the complexity increased, transactions slowed down and no one agreed to upgrade block size since this is additional costs to mining.

4. Since bitcoin now is slow and expensive industry found another way to use it - store long term value without risk of government to inflate currency and steal your money. point#2 is still valid. Recent surge in prices is caused by this reasoning in my mind.

5. Since everyone earned enormous amount of money in past months people will keep hodling expecting the growth to continue, but it will slow down. Once it slows down - people will start selling to get their profits and there will be huge correction.

6. In the very long run bitcoin can go through multiple cycles of such growth and decline and either will become more predictable (community will come up with some rules on how bitcoin fees and speeds are regulated etc.) or will stay this a bit more dangerous but a but more free way to transfer value geographically and also over time.

It probably could be more dangerous to world governments if point #3 realized

If more countries outlaw exchanging bitcoin with fiat commercially (for example because of its detrimental effect on the environment), bitcoin's value will decrease.

> Is the primary reason that bitcoin has become more popular that an investor can spend 10,000 fiat on it - wait a year - and then extract 15,000 fiat from it?

It seems to be true.

Here's an alternative mental model... Let's talk about Filecoin (related to Chia) for file storage. Assume for the moment that all 2 billion tokens are in circulation. At any given time, there is a market of people looking to store files and people willing to store those files for a fee. You're transacting (roughly) in Gb-years at today's level of technology -- i.e. willingness to store 1Gb of data for 1 year. Regular market forces apply. You could also pre-purchase credits for later use... either to get higher priority for your data when you want to use your tokens, or as an investment in said ability. And the prices you'll pay will automatically flex based on improvements in technology since the global supply of tokens is fixed.

You could then reasonably think of Filecoin as a futures contract on generic file storage capacity. Ethereum is a futures contract on computational horsepower. Bitcoin is something else(?) - perhaps a futures contract on a general long-term abstract form of value storage. In short: Some of these tokens act as futures contracts for decentralized compute capabilities. The most-common current use case happens to be heavily related to finance (DeFi); applications for censorless content networks are also in development; and the applications are really quite general purpose.

Anything today that requires a ledger is a reasonable problem to tackle too, where eliminating centralization could be beneficial. E.g. ENS instead of ICANN for domain names, tokenized stock certificates instead of relying on opaque organizations owned by major banks (e.g. the DTC and problems around naked shorting).

So I'm trying to check out FileCoin as I like the idea it tries to solve but: the website has some nifty visuals but after reading a few pages, I have no idea:

1- how to really get started and participate if I don't have a dedicated server (I mean I would expect something easy to onboard, but I don't want to setup a whole dedicated machine for that, despite the $100 rebate on Digital Ocean's bill).

2- How are the coins allocated and what the heck you can do with them?

3- what if I'm just a user who wants to store some files? No idea how to do that, through who I should go, what tools I should use. How do I pay for the storage to start with?

4- no idea how this economic model works: spend on hardware to become a miner? but I don't see how you get back this heavy investment. I understand miners get paid, but see 3).

5- who is this for exactly? It doesn't seem to target individual users, small businesses or even large companies. Is it just an experimental project, is anyone actually really using it and building stable 'stuff' on it?

6- how does this compare to the other storage solutions in terms of availability, andwidth and cost?

7- What's the adoption so far? Is there any storage available in my region?

I mean, it's a great idea but even as a 'tech guy', I'm lost as to how to use that infrastructure.

I'm sure that I could invest a few hours into it to find out[1], but should I really have to put that much effort into something that should be clear in a couple of paragraphs without the confusion of big word marketing[2] (I mean that page has all the buzzwords but doesn't explain much by the time you reach the bottom it).

I think the time spent on that nifty animated intro[3] of the front page could have been put to better use. I see no onboarding effort although the docs look good.

Maybe this is too much of a rant and I'm being unfair, I'm sure I must have missed some important things, but I don't see how this can get wide adoption if the basics are not clear and the onboarding is not better.

Or maybe it's just me; I still don't know what it can do for me. Maybe I'm not the target?

  [1] https://docs.filecoin.io/about-filecoin/what-is-filecoin/#for-storage-providers

  [2] https://filecoin.io/store/

  [3] https://filecoin.io/

Filecoin is meant for stable long-term storage; thus, mining the unused space on your personal harddirve is not the intended use case.

Today, there is a lack of Dropbox-like systems to make the decentralized system broadly useful. It's a shortcoming that people are working on solving. It's akin to being given a big ole harddrive: Until you have a filesystem & file browser, it would only be relevant to a small population -- but that utility would explode once tooling is complete.

Some resources to address you question on utilization:



Also: spending less than an hour trying to grok something new, and then ripping into it seems pretty disconnected. Thinking back to my learning curve on emacs, linux, IRC, BBSs, early www, etc. Hell, I remember spending days(!) wire wrapping that first computer and flashing the ROMs. It took substantial effort to be involved in the "obscure hobby", but look where computing and the web are now! Don't mistake ease of use today for potential impact tomorrow.

These velocity of money models can't account for even a small fraction of cryptocurrency valuations. For example, if your model says Filecoin is worth $10 but it's currently trading for $160... clearly the buyers are not pre-purchasing storage.

https://file.app will give you a reasonable approximation of the current value of Filecoin in Gb-years. From what I can tell, it still compares very favorably to various AWS offerings even at today's price. You could make a reasoned case that recent price run-up was more related to discovery than fundamentals... But regardless, my stated supposition assumed we reached steady state on token issuance (i.e. many years into the future).

New paradigms are likely to exhibit a lot of volatility during their onset, but timing matters less than giant rising tides. Time will tell.

Outside the options you mentioned, people buy crypto because the tech is cool, and the idea of a trustless decentralized system apeals to them.

The end goal isn't really clear. Perhaps the main thesis is like VC investing. Cool stuff now might become useful and valuable later.

I bought some address space (like for a web server) in ETH the other day. I think the Ethereum contracts system is quite nice and logical to put on the block chain.

Changing stock trading protocols to go on a block chain with full public transparency would be a useful application. No need for PoW, you can just have the Fed mint new blocks out. This would prevent a lot of the shenanigans and gatekeeping that occur in capital markets. It would greatly improve price discovery.

On a less directly applicable level, I feel that the only realistic strategy for governments in the long term will be to become more competitive with the crypto-ecosystem (currencies, contracts, encryption, etc.). The hope being that by allowing individuals to freely choose their own legal and financial rule-sets the systems that will remain relevant will become better. Having more freedom of choice for which government you'd like to be a part of is ultimately a good thing. Whether that will actually shake out is anyone's guess, but I believe something along those lines is the fundamental hope of this area of technology.

If only someone could invent a cryptocurrency that incentivises something good instead of something harmful.

Gridcoin is designed for this. It is based on scientific research and instead of sha256 hashes of random data, coins are earned by proof of your computer doing scientific research.

From their homepage:

> Examples include protein structure prediction (Rosetta@home), mapping the Milky Way galaxy (Milkyway@home), and tackling problems in public health and clean energy (World Community Grid).

How do you prevent malicious research projects in a decentralized system?

On-chain voting is used to decide what projects are whitelisted.

What would you consider malicious?

A project that provides credits without doing valuable research.

Proof of Social Good. Increase your neighbor's credit score and you will earn a coin.

It will be gamed.

There's FoldingCoin:


Things that would make sense:

- SETI@homeCoin, ala https://en.wikipedia.org/wiki/SETI@home

- ArchiveCoin, ala https://wiki.archiveteam.org/index.php/ArchiveTeam_Warrior

These would actually be useful.

If I fold for free, i get nothing.

If I fold trough one of your examples I can get money.

Where is that money coming from?

Blockchain seems like an artificial extra step to pay folders in this case.

The problem is that it is really hard to link non digital things to the digital world.

oh great, another item that will be impossible to buy soon

thanks, miners

the sooner someone introduces the law to ban all proof of work crypto, the better - or at least banning companies from taking or using PoW crypto would be a good start

The India model is right -- give people 6 months to sell off their assets and make it illegal.

You are right, hard drives should only be available to mega international corporations like Google and Facebook to make money off of.

Can anyone clarify for me why/how this incentivizes SSDs over HDDs? Seems if it's really just about having "proof of space", whatever HDD is cheapest would make sense? How much data gets read back how fast during validation?

It's the initial generation of the 'plots' that make up the proof of space that's disc IO limited. So faster SSDs for that and then moved onto HDDs for long-term. According to Chia, it uses 332 GB to generate a 101.3 GB proof of space:


So just need a computer with 350GBs of ram?

That works of course, but RAM also costs per gigabyte an order of magnitude or so more than SSDs.

Doesn't wear out though. Question is will they require more ram in the future once 350GBs becomes cheap?

First we had proof of work, now we have proof of space. Both are rather wasteful.

The challenge appears to be to create a proof of something beneficial for society, like "proof of clean energy generated", "proof of not flying", "proof of vegan lifestyle" or "proof of CO2 sequestered".

There already exists a cryptocurrency that is both environmentally friendly, fast and fee-less - Nano [0]. It uses something called directed acyclic graph (DAG) and while it does require proof of work it’s only for sending transactions to prevent spam, not for mining. It was created a few years ago but only recently it starts to get more traction. I used it a couple of times and it worked great.

[0] https://nano.org/

Proof of X unfortunately has to be capable of being verifiable by a computer without human intervention.

Just bought a new Synology NAS this week and started putting together disks and they're scarce and the prices are through the roof.

Couldn't find what I was planning on buying. What I found was through the roof. I bit the bullet and bought 4 new ones. Not enough to fully populate the NAS, but at least it'll give me some space for now to do what I need.

We've seen scarcity and what it does for prices after the flooding in.. 2010?(ish).

I just did some NAS upgrades a month ago and feel as if I dodged a bullet.

I'm curious what you ended up paying. I bought drives on sale, and got nine 16TB external drives to shuck (one is a spare) for ~$2250 USD taxed and shipped.

I will warn you though that burn-in testing is going to take a while. I did a very exhaustive test[1]... and two drives at a time over USB3 it took about twelve calendar days to test them all before I put them into service.

-- ----- ----- ----- -----

[1] Write out block pattern, read back. 2 hour break. Head thrash random writes/reads for 2 hours. 2 hour break. Head thrash random writes/reads for 2 hours.

$399.99 for 14TB seagate expansion ones. Still waiting for them to arrive. Everything else cheaper I found was out of stock.

NAS has 6 bays, I only bought 4 to get a RAID 6 setup and I'll expand when I find better prices.

If anyone has links, I'll take them!

Did you look at large external USB hard drives' prices?

I did.

I basically ended up getting external seagates that I’m planning to schuck after doing a ‘burn-in’ test. Still high for what they were.

Is Chia's real play investing in HD companies?

I can see Eth existing in 10 years time. I can't see Chia existing in 10 years time.

Crypto verse in general feels like a 50/50 bet...but the storage based stuff feels like significantly worse odds

why u think that? curious

Why would Chia be the threat here? Filecoin(IPFS) is already a threat, it already has a capacity of 4.81 EiB of data, 4810 PetiBytes! (Is that how you spell it? In contrast to Petabytes)

And to store 1 GiB of data per year it only costs the user $0.00082 USD.

That shows harddrives are already being gobbled up by miners at an incredible rate. Has this made a noticeable rise in storage prices? Anyone know of any data that can show that?

Supposedly Filecoin is storing 'real' data, so it's not using any more storage than AWS or any cloud provider would. Is that not the case? At this point I wouldn't be surprised if the the number of 'real' data stored was only a fraction of the total network size. Chia is just storing 100% useless data outside of its use by Chia itself.

i.e Filecoin's and similar coins should have value directly tied to their use to store data while Chia is just another crypto that has no 'intrinsic' value outside its use as a currency and computation platform.

>Filecoin(IPFS) is already a threat, it already has a capacity of 4.81 EiB of data, 4810 PetiBytes! (Is that how you spell it? In contrast to Petabytes)

4.81EiB is ~5.54EB. Filecoin has been around for ~81.5 months. That's an average growth rate of only ~68PB/month, but we can of course assume that the current growth rate is higher. For example, last year there was a period where it grew 250PB in a month.

Chiacoin has supposedly added ~200PB to its mainnet in the past 24 hours. That growth rate is also still climbing.

Seagate has been quoted as having an current HDD manufacturing capacity of ~350EB per year. The general assumption is that between Seagate, Western Digital and Toshiba there's likely a total manufacturing capacity of close to 1000EB a year.

Of that 1000EB, a large percentage goes to OEMs and SIs, another large percentage goes to direct customers (Alibaba, Amazon, Apple, Facebook, Google, Microsoft, etc.), and the remainder goes into the sales channel (retail, VARs, etc.). OEMs, SIs and direct customers have contracted supply and pricing, whereas the sales channels do not. Channel is likely 30-40% of current production volume. So that's 300-400EB a year of capacity across all manufacturers going into the sales channel where end users can buy from.

-- ----- ----- -----

At 0.2EB/day of growth, that is 18-24% of all channel volume being consumed just by new Chia farm plots. At that rate, that'll cause channel supply issues very, very soon for certain. Once channel supply becomes exhausted, prices will rise since the demand is there.

The issue is... what ends up happening to the sustained growth rate? If it were to reach something like 1.0EB/day in growth, that is basically all of the normal channel capacity being spoken for just by Chia farm plots. That's an actual disaster for end users.

Manufacturers can't respond to those demand spikes instantly. They can respond some, for certain, but if some component hits maximum manufacturing capacity, you need to expand or build new factories (whether the manufacturers, or contract manufacturers within the supply chain for certain components). Building new factories is not quick. Not just construction, but ensuring that the demand will be sustained so you don't end up with these huge capital expenditure factories going idle when demand wanes.

> And to store 1 GiB of data per year it only costs the user $0.00082 USD.

How can filecoin possibly be that cheap?

Providers like B2 and wasabi are 6-7 cents. Sia spent a while in the single digits is currently around 10. This is less than a tenth of a cent?

I think Filecoin miners are currently being heavily subsidized by mining rewards so they don't need to make a profit on storing data. At those prices, I wonder if storing data is more trouble than it's worth.

What is this space used for? Generally curious

Do they have a benchmark for how many tps that Chia can handle ?

Bram Cohen has been quoted as estimating ~20 TPS at Layer 1.

Great, now nobody's going to have CI so we can get 3TPS on BTC, nobody's going to have GPUs so we can get 12TPS on ETH and now nobody's going to have SSDs so we can have 20TPS on Chia. The future is now.

I don't blame the people buying up the GPUs to mine these cryptocurrencies but the vendors who allow them to do. It is absolutely criminal that they don't filter out these miners when they sell these products. They can easily make it possibly to only allow maybe 1 or 2 of these computers are shipped to a certain address or only one graphics card per a credit card. Whatever way they choose. Even putting a captcha to knock out the robo buyers would probably be enough.

They have an obligation to video game makers, hospitals, and other businesses to sell them so that people that need them for other purposes can purchase them at reasonable prices. It's also on the govt to make sure that this happens.

Just my 2 cents.

Is there a blockchain for webhosting?

Is the disk space used for anything useful in parallel to Proof of Space?

What would have been smarter would be to make the space allocated to Chia also serve as a decentralized, worldwide storage facility (e.g. compete with S3/Glacier). You'd get better utilization of the resource, and have an extra revenue stream for the system.

What does this do that Filecoin can't?

What stops me from running multiple nodes pointing at the same stored data? Or would that not be beneficial

The disks store cryptographic lottery tickets - your chance of having the right one increases the more space you have. Two nodes pointing at the same batch wouldn’t.

Only one "wins" the reward.

When do we stop this madness.

Does this inherently support Filecoin like use cases?

Can anyone explain the algorithm a bit?

Here's the whitepaper: https://eprint.iacr.org/2017/893.pdf

Unfortunately, it's a lot harder to follow than Satoshi's at least for me.

I hold no Chia or crypto at all, however, from what I am seeing around me, Chia is going to be really, really big.

We can always hope that trading Chia is made illegal, along with proof of work algorithm based currencies in general.

Governments so-far failing to regulate existential threats to themselves is honestly odd.

Chia is not Proof of Work, FWIW. Their technical and business whitepapers are helpful resources for learning more about what they call "Proof of Space and Time".

All of them are Proof of Waste.

Is Proof of Stake proof of waste? It has some of its own unique challenges, but I think it seems better in that it doesn't require committing lots of physical resources.

You're right, proof of stake isn't proof of waste. Proof of work and proof of space are, though.

I guess that depends whether an exabyte is a lot.

Talking about Proof of Stake, not Proof of Space.

>Governments so-far failing to regulate existential threats to themselves is honestly odd.

Is it the job of the government to hang onto power at all costs?

Officially, no, but it is an emergent property of their other responsibilities.

For instance, a government cannot maintain law and order without power, and a society without law and order is the definition of a failed state ([edit] well, more specifically one in which the government has lost its monopoly on violence). It's not an unreasonable interpretation to therefore say any non-failed state must retain power.

Bank bailouts in 2008 justified? Save the rich, screw the poor and the middle class. At least the society is peaceful.

Mmm not to me no, I believe in a capitalist system that lets losers lose. However the argument was ordinary folks would pay.

It is, sort off, the inherent goal of a (nation-)state. At least, that is a common and rather accurate historical and geopolitical model.

maybe it's not the job, but it's the nature of governments

> Governments so-far failing to regulate existential threats to themselves is honestly odd.

Factions within governments all around the world are using it to their own ends.

We can hope, but, if you're going to need hard drives in the foreseeable future, you might want to go buy them now.

From people hoping to mine and then immediately sell it, or from people who are planning on buying and holding?

I don't know what their plans are. Just from the amount of buzz I'm hearing around me. Reminds me of the early days of Ethereum.


Just the amount of buzz and interest I'm hearing. One data point (which is the only one I can share without revealing personal info on this public site) is that my local hardware store sold out of 8TB+ HDD drives and, according to the employee there, that hasn't happened in many years.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact