Secretive Bitcoin Startup 21 Reveals Record Funds, Hints at Mass Consumer Play

http://blogs.wsj.com/digits/2015/03/10/secretive-bitcoin-startup-21-reveals-record-funds-hints-at-mass-consumer-play/

by yuvadam • 11 years ago

130 95 11 years ago

blogs.wsj.com

softdev12• 11 years ago

This is interesting because it's the first time I've heard about 21 and they have raised $116 million pre-launch. Given that companies who've raised an order of magnitude less are all over the news, it's surprising they have been so stealth.

I'm also still not clear what 21 does. According to the article they are building out the blockchain into the internet of things (IoT). That sounds really vague. Plus, the article mentions a contractual twist - like smart lawyer contracts. These two things don't seem to obviously go together.

zmanian• 11 years ago

It sounds like some variant of the hardware Bitcoin bank idea that has been floating around the valley.

The rough sketch of the idea is

-Hardware device runs a full security bitcoin node in your house -Provides a multisig for your mobile wallet. -Automatically moves savings into cold storage -Provides a basic environment for implementing crypto banking features

alvarosm2• 11 years ago

yeah, another bitcoin pipe dream... what's needed is probably something like integrating a wallet+bitreserve+exchange+payment processor in a single business. And then forking/enhancing btc itself. Some of the top btc companies have got ridiculous investment rounds that could be used towards that goal, easily.

diminoten• 11 years ago

The blockchain ends up being useful as a public secure ledger, so perhaps it's going to be used in that way somehow.

Otherwise, I don't know of many companies who've actually worked into the space of the less well-known features of bitcoins that can be used via the embedded pseudo-language.

giant_squid• 11 years ago

Their jobs page provides some hints as to what they're up to: https://21.co/#jobs

kordless2• 11 years ago

They do if they allow you to do seamless software installs and payment chains. More than obvious, if you've been thinking about it.

sajid• 11 years ago

The vision is to use bitcoin to enable automated micropayments between devices. One example that Marc Andreessen has talked about is sharing your network connection with neighbours:

http://www.wired.com/2015/03/opinion-bitcoin-may-gets-us-rea...

21 aims to provide the tools and build the full stack infrastructure to make that happen.

phodo• 11 years ago

"These two things don't seem to obviously go together." That actually makes a lot of sense: if you consider some of the security challenges associated with IoT, e.g. an IoT door lock, then a smart contract where you have m of n signatures needed to unlock it (think of them like a modified escrow with policy side, too), then you can see the play here. The transaction to open the door gets signed only if that policy is met at each of the m signatures. You might have a lock device signature, a home signature, and a home owner signature. There are lots of microtransactions in IoT (good and potentially bad). Smart contracts can mitigate / isolate risk and ultimately reduce attack surface vulnerabilities. Block chain is a necessary part of this to work. For more info check out "bitcoin 2.0".

comrade1• 11 years ago

Are there ways to reduce the size of the block chain? Or will we get to the point where only large service providers can hold the block chain?

I ask because I was maintaining my own wallet for awhile but stopped because the block chain database was getting unwieldy - it would take up to a week to resync if I ever turned my wallet software off for a significant amount of time.

You know though... I actually don't care. Bitcoin was an interesting curiosity and useful in some situations (transferring money to people in rural areas of 3rd world countries, for example), but it's slow (up to 15 minutes for the transaction to be approved?), has high fees in its transactions and even higher to convert from/to cash, it's unwieldy...

To use it from consumer devices would require someone else managing the wallet, and at that point I'll instead just use cash or a credit card.

wmf• 11 years ago

SPV or ultimate blockchain compression can reduce storage requirements significantly.

alvarosm2• 11 years ago

Blockchain pruning would drastically reduce the size and it could be implemented relatively easily but it's not on the standard client(or any other afaik). You only need to keep unspent transactions (after you're confident a block reorganisation wouldn't go as deep as the block where you're pruning old tx).

dollaaron2• 11 years ago

I believe there are proposals to allow pruning the blockchain, but I'm not 100% sure on the details.

To guarantee a transaction goes through you have to wait for it to be confirmed, but most transactions really aren't that important/high risk, so beyond checking that the address does actually have the money, it's instantaneous. Many payment processors (I know Bitpay has this option) actually take the 0 confirmation transaction risk, allowing customers to pay instantly.

Transaction fees are not high, with a typical fee of .0001 bitcoin (3 cents), which is small compared to credit card fees. While work needs to be done converting to/from cash, actually using bitcoin (that you already have) to buy things is a better experience than using credit cards. You simply grab your phone, scan the QR code, and hit send. This is much better than entering your credit card for the thousandth time on another website, or being redirected to paypal, entering your password, and choosing your preferred payment option for the thousandth time (stop asking me to use Paypal Credit!).

Using an Android wallet such as Mycelium, or a desktop wallet like Electrum doesn't require you to download the entire blockchain, and although that comes at the cost of some trust, I personally don't find that problematic.

Basically, bitcoin has some problems, but I think some of your criticisms are no longer applicable to most everyday usage of it.

CPLX5• 11 years ago

actually using bitcoin (that you already have) to buy things is a better experience than using credit cards

It's just not. It's not easier than swiping a magnetic strip and signing something, or typing in sixteen digits (or using autofill or the site's previously saved CC info) and pressing buy.

It's not good that I can't demand a return or have recourse in the event of fraud. And it's not a benefit that with bitcoin I have to spend the money first (to buy bitcoin) and wonder if it will hold its value, versus just buying stuff when I feel like it on credit and covering the transaction within 30 days at no interest cost.

I know there are a lot of people who really want your above quoted statement to be true. But, for now and the forseeable future, it isn't.

padelt• 11 years ago

There is the concept of pruning in the original paper[1] which basically removes all fully-spent transaction trails. That does not yield a reasonable upper bound on the size though[2]. Beyond pruning, it seems you need to trust another network component (external wallet, server).

[1] https://en.bitcoin.it/wiki/Scalability#Storage [2] http://bitcoin.stackexchange.com/questions/4650/upper-limit-...

richcollins• 11 years ago

There is an iOS wallet that connects directly to the network that downloads the necessary info in seconds:

https://itunes.apple.com/us/app/breadwallet-bitcoin-wallet/i...

EC1• 11 years ago

Wouldn't this be really power draining? Having to sync and download constantly.

deweller• 11 years ago

This uses Simplfied Payment Verification (SPV).

SPV helps with some scalability issues, but doesn't solve them all.

From http://bitcoin.stackexchange.com/questions/4649/what-is-an-s...:

SPV: A Bitcoin implementation that does not verify everything, but instead relies on either connecting to a trusted node, or puts its faith in high difficulty as a proxy for proof of validity. BitCoinJ is an implementation of this mode.

chatmasta• 11 years ago

This is an important question since the focus of the article's speculation is on the Internet of things. Obviously embedded devices are not going to be lugging around gigabytes of blockchain any time soon. Even with pruning you're not going to get the size down to a realistic number for devices with little to no hard drive capacity.

I imagine in any IoT scenario, a fleet of sensors talks to a trusted central hub. If Bitcoin is involved then the hub could run the wallet and the sensors could delegate transactions to it.

puranjay• 11 years ago

I'm in India and I can't even buy it easily. Paypal hates it, as does India's Reserve Bank.

Too much effort for too little reward. Right now, the only thing I'd actually want to buy with BTC is illegal stuff off the darknet. There is virtually nothing else I need that bad that normal cash can't do.

666_howitzer• 11 years ago

I'm from Bangalore, I have used localbitcoins quite a few times actually, all the trades went smoothly. The only issue I have with LBTC is you have pay a small premium compared to other exchanges for added convenience.

https://localbitcoins.com/

jafaku4• 11 years ago

Paypal hates it, as does India's Reserve Bank.

I can't believe this doesn't immediately make you think that maybe, just maybe there is something good about Bitcoin. Are you 0K with the central bank stealing wealth from you and your family through the debasing of your currency? Are you 0K with working your ass off so that bankers and other powerful people can live a life of wealth and pleasure? How can you take advise from such people?

"Bitcoin is bad, don't use it! It's a drug currency... very shady... also, terrorism!"

That is basically what every central bank said to the people in their country.

anilgulecha• 11 years ago

unocoin seems to make it quite easy to buy /sell in India.

mrb2• 11 years ago

Yes there is a solution. The reason it took you days to resync was because you ran the reference implementation, Bitcoin Core, which is meant for "full nodes" running 24/7. Simple users like you should instead run "thin client" or SPV wallets (Simple Payment Verification) which do not need to store the whole block chain, allowing your wallet to resync pretty much instantly. For a list of SPV wallets see https://bitcoin.org/en/choose-your-wallet and look for those labelled "simplified validation".

You also very likely ran into the problem that older versions of Bitcoin Core were slow to download blocks. For example version 0.10.0 released in February 2015 introduces parallel block download, which makes resyncing faster. Numerous other optimizations were made over the last 12-24 months (eg. better detection of stalled downloads). From your description it sounds like you were running a significantly older version of Bitcoin Core, so you were missing out on all these optimizations.

I would like to point out that despite the block chain growing continuously, we are still very far from the point where only large service providers can hold the block chain. The average block size is currently ~400 kB (https://blockchain.info/charts/avg-block-size?showDataPoints...), this means the block chain grows by ~20 GB every year. This is only 1% the size of an average desktop hard drive (2 TB), and would take only 4.5 hours to download (once a year, mind you) assuming a steady common residential Internet download speed of 10 Mbit/s.

Heck, a dedicated computer on a common residential Internet connection could still run as a full node and handle an average block size 100x larger that today (~40 MB per block, 2 TB growth per year = 0.5 Mbit/s continuous download). I would hardly describe something an enthusiast can do as a thing that "only large service providers can do". And keep in mind by the time 100x growths happens, CPUs/HDDs/Internet speeds will be much faster than today. Bitcoin's scalability has already been benchmarked pretty well: https://en.bitcoin.it/wiki/Scalability - no immediate bottlenecks are in sight.

"high fees in its transactions"

Fees are optional, and typically 0.0001 BTC (3 cents).

"up to 15 minutes for the transaction to be approved"

No, transactions are typically "received" instantly within seconds, and take 10-60 minutes on average to "settle" (be confirmed by 1 to 6 blocks). This is a lot faster than credit cards or ACH which both take 24h (or more if this is a weekend) to be received (ie. dollars available in the recipient's account), or bank wires which take days (or end-of-business-day if your are lucky).

comboy• 11 years ago

Hey mrb, I miss your blog posts.

(For those curious: http://blog.zorinaq.com/ )

jafaku4• 11 years ago

"Hi everyone, I just heard about Bitcoin... And I'm here to fix it"

IkmoIkmo4• 11 years ago

Yes, it's called pruning, it's in development and not finished.

The basic idea is that bitcoins are actually units of money that are in your wallet. Rather bitcoin works by chaining transactions (put into blocks, hence the blockchain).

Thus if you give me $1, I don't have $1, I actually have a record that you gave me $1 so that must mean I have $1 to spend. That's the simplistic idea.

That is the reason why instead of storing account numbers with money numbers in them, bitcoin stores transactions.

By now it's probably obvious to you that the vast majority of those transactions don't need to be stored. If I give you $1 and you give it back, repeat 1 million times, we end up with exactly the same amount of money. And all but the last few transactions can be reasonable deleted. (pruning)

As for syncing, syncing up a wallet is much, much faster now that it's headers first. But most users never sync, they either use 3rd party wallets like Coinbase, or they use wallet software that plugs into various nodes and APIs, so that no blockchain has to be present on-disk. The blockchain is really stored mostly by nodes, dedicated servers, businesses, a few enthusiasts.

That doesn't mean someone else manages the wallet by the way, like you mentioned for 'consumer devices', it just means that someone else manages the node. But the node doesn't control your money, your private keys do, and those are just a few bytes of storage. And as there are thousands of nodes and many APIs, trust of nodes isn't any different as when running your own node (which gets and sends data to other nodes, requiring the exact same amount of trust which is zero when it comes to controlling your money).

As for transaction fees being expensive, they're extremely cheap? Pennies to send thousands or millions, and as it's an open protocol anyone can build free centralized payment networks on top of it, like Changetip. Conversion with other currencies is also cheaper than most fiat conversions (e.g. dollar to euro), and in fact some companies already offer it for free (e.g. bitcoin/usd is completely free and pretty much instant with Circle, founded by Jeremy Allaire who founded Coldfusion).

The 15minute thing might be a deal-breaker for some. There are lots of solutions though that can remove that timer, built on top of bitcoin. Sidechains, treechains, payment channels etc. Anyway time will tell if these things work well but there don't seem to be any dealbreakers yet on the horizon.

Procrastes• 11 years ago

There are a couple of good options for working with truncated blockchains to a checkpoint without reducing the security of the blockchain. Likely one or the other with settle out as the common practice, and the blockchain growth problem will be solved.

Power consumption, on the other hand, may not be solvable even in theory, and it also seems likely to drive mining to just large service providers. The current pools aren't a solution in this case because the startup cost to even participate meaningfully will grow too large.

Maybe this could be solvable with a different proof mechanism. A few have been suggested and tried (PPC etc.), but I don't think it's solvable for BTC as it's currently defined. I'm happy to be proven wrong.

imaginenore• 11 years ago

What are you talking about?

Blockchain is now 30GB. A 4TB drive costs around $120. So it literally costs you 90 cents to store that data.

If 30GB bothers you, use MultiBit or Electrum.

presty• 11 years ago

21 Inc was previously 21E6 https://bitcointalk.org/index.php?topic=334759.0 who was in the news back in 2013

jboggan2• 11 years ago

What percentage of that funding is just buying and holding Bitcoins? Many of these large rounds have that business model baked into the amount they raise. There's much more profit to be had from buying at current prices and selling at a consumer inflection point than there is in actually building a Bitcoin service.

malenm• 11 years ago

I think building Bitcoin services is the only way to make Bitcoin _reach_ that consumer inflection point.

IkmoIkmo4• 11 years ago

It's idiotic to assume an investor will invest in a company to buy coins, rather than coins directly.

Maybe if it was expensive art, exotic animals or something difficult and tricky to buy and store without having specialized knowledge, yes then you'd invest in a company to buy and manage these assets.

But storing bitcoin is storing private keys... a password. Anyone can buy it and store it. Hell you can even buy it on registered exchanges, or as a security from SecondMarket's bitcoin fund.

No. These guys raised money to hire software and hardware engineers, they're looking to build a product.

Does that mean the investors aren't buying bitcoin? No, they do so, but outside the company. It's well known that there are a number of VCs (e.g. Chamath) who have bought a few million worth of bitcoins besides their investments in bitcoin companies.

I don't any evidence that many of the 'large rounds' have this business model. There are some companies that bought bitcoin, but that's for liquidity, not investments. Circle, Coinbase or exchanges all need bitcoin to sell as they're brokers/exchanges. But they didn't buy it as an investment, they could just buy bitcoin as individuals, as could investors.

jafaku4• 11 years ago

It's not idiotic, that's why there is a Bitcoin fund, and soon there will be a Bitcoin ETF.

If being an expert in computer security isn't "specialized knowledge", I don't know what it is.

jboggan2• 11 years ago

I spent 2014 trying to get a Bitcoin derivatives exchange off the ground and had a lot of intimate conversations about business models with large players, I am not speculating here.

Consider that Bitcoin has return characteristics that are highly desirable to a VC's risk profile, yet the restrictions of their limited partnership agreements precludes VCs from directly buying Bitcoin.

loceng2• 11 years ago

How much Bitcoin do the founders, employees, or investors own? Curious to see what real gains they could realize if they succeed in making people see Bitcoin as a legitimate means of payment.

vessenes• 11 years ago

LOTS. 21e6 has been mining for some time, and A16z has been funding and involved in Bitcoin for a number of years now, with a very 'winner take all' approach.

loceng2• 11 years ago

Thanks. What's Coinlab? The website is down.

brighton36• 11 years ago

hopefully, they take me up on the advice I gave Bitpay: http://cointelegraph.com/news/113255/keeping-score-at-the-20... (see "The solution")

Animats• 11 years ago

If they raised over $100M, why do they need Bitcoin? I'd assume they intend to have their own coin. Fewer headaches.

markkat• 11 years ago

It would be very difficult to get another coin to the level of adoption that bitcoin has. Many have tried, and none have come close.

shashu10• 11 years ago

They possibly could if they create a more secure cryptocurrency

kaonashi• 11 years ago

A solution in search of a problem.

swamp40• 11 years ago

Since they're looking for ASIC engineers, I guess they're a mining hardware company.

I've been hoping someone will design a bitcoin cash voucher reader/printer, like you see in the casinos.

(For use in small retail business and vending machines.)

downandout• 11 years ago

With this kind of investment capital on the line, I sure hope that they are focusing their efforts outside of the consumer payments area. I think the fact that consumers must purchase Bitcoin in advance, with no clear advantages that average consumers would appreciate (consumers generally don't care about privacy), makes mass-adoption of Bitcoin for payments incredibly unlikely to ever happen.

The biggest question in my mind is whether or not they have developed a realistic revenue model. The whole point of the Bitcoin protocol is to relinquish control of the blockchain to the network's participants, which makes it hard to monetize. They could license whatever software/hardware they manage to build for various applications of the technology (voting, research, etc), but $116 million seems like a big bet on a company building solutions on top of open source software.

Edit: It appears that they are building both software and hardware, and more interstingly, intend to operate their own datacenters. One of the job descriptions from https://21.co/#jobs :

"21 is seeking a Datacenter Operations Engineer to travel to our datacenters and keep them up and running smoothly. You should like big iron, cutting edge equipment, and firmware. The job description says you will be responsible for "day-to-day operations of 10,000+ server installations". I guess we know where $116M is going.

kordless2• 11 years ago

It may be coincidental, but Qualcomm just announced a massive stock buyback.

My best guess is this is an IoT play targeting embedding cheap cell chips into devices which will carry their own wallets. The cell chips connections to the network can be funded with cryptocurrencies the devices carry, but the ultimate goal is likely enabling software delivery into the devices. Devices pay for the software downloads. Users fund devices using credit cards, etc. Think of it as a fancy licensing engine for the trillions of devices that will be hooked up.

My second guess would be a decentralized cell network, powered by cryptocurrencies.

Either of those is a massive market.

Secretive Bitcoin Startup 21 Reveals Record Funds, Hints at Mass Consumer Play

Related Stories