Oleg Andreev
Month
Jeff Garzik wrote in February:
“Block size is VERY MUCH like bitcoin’s 21M limit, so a lot of care must be taken when changing MAX_BLOCK_SIZE logic. Block size is an economically limited resource whose production is tightly defined and controlled by algorithm, with an intentionally steady production rate (the 1MB limit).”
I have a number of comments on this statement.
21M of coins is an arbitrary non-economical limit which basically defines divisibility of the total amount of money. Actually, we have around 2100 trillion smallest units (1 Bitcoin is 100 000 000 units). If the limit was 42M it would only mean higher divisibility, not that we have more money in any useful sense.
“Economically limited” is not the phrase for something arbitrarily limited. There are things in the world that are arbitrarily scarce (e.g. amount of gold in the ground). Sometimes these limits can be stretched when it’s economically interesting. E.g. if we use up all the gold and need some more, we may find it profitable (or not) to synthesize it.
Block size is not even economically defined. Transaction fees are economically defined: you pay whatever fee you want and someone else will decide if it’s enough or not. Transaction fees are defined by the market. Miners compete for the fees, users compete for the place in a block. We may say that transaction fees are “economically limited”.
Bitcoin parameters (block size, hashing algorithms, block time interval) are more-or-less arbitrarily defined and all are economically changeable. If everyone finds it more useful to switch from SHA2 to SHA3 (e.g. because of some security threat) comparing to the costs of global transition, then we will switch. Block interval of 10 minutes was chosen to minimize the amount of orphaned blocks (wasted work) while the fresh block is being propagated over the network. Everyone wants lower latency, but no one wants to waste resources. It is unlikely that someday it will be useful to cut the time interval: for expensive transactions people may want to wait for several hours, so it does not matter how many blocks are created during that time: 10 or 100.
Likewise, the block size limit was introduced to prevent situation when blockchain gets too big before it is widely adopted. Imagine if in early days the whole chain was not 1 Gb, but 100 Gb, how many people would want to play with it? But in the long run blockchain will be huge anyway and it will be managed by specialised nodes (mining pools, shops, banks etc.) Miners would be paid out of transaction fees, so they (like in any other business) would tend to increase their throughput as much as economically possible. When amount of transactions will start hitting 1 Mb limit, transaction fees will go up and various clearing houses would start competing with the miners for the fees by clearing transactions outside the blockchain. If miners are capable of propagating bigger blocks with extra costs covered from the extra transaction fees, then they will be interested in raising the limit. Also, every clearing house, escrow, bank and shop would be interested in raising the block limit too as it will lower their costs. Of course, by extension, regular users will pay lower price and will be able to transact directly on the chain with lower fees. So they would desire bigger blocks too. Block size will thus be limited economically: it will grow as long as transaction fees cover extra costs.
Today block size is nominally limited by the protocol, but since most of the blocks are well below 1 Mb limit, the block size is really economically limited. When the hard limit starts manifesting itself in higher transaction fees, everyone involved will find it useful to increase the block size. Even Satoshi mentioned that hard block size limit is temporary to prevent blockchain from bloat before it is mature enough. So there is no point in political debates around the issue. We don’t know when exactly we will have a problem (maybe never), but when we do, almost everyone invested in Bitcoin will vote with their resources for raising the limit.
Paypal takes around 3% from the merchant. FastSpring takes around 9% (because its UX and features are great, BTW). Meanwhile, Bitpay and Coinbase ask only 1% and bear the risks of volatile exchange rate, lagging exchanges etc.
Zero possibility of chargebacks and near-zero transaction fees are not the only features that make Bitcoin interesting to merchants. Ask yourself: who would pay with Bitcoin today? Those who have some spare cash on a credit card would prefer to spend it first while keeping their precious coins. But if someone pays with Bitcoin, they either don’t have access to credit cards or banking system in their country, or they are trying to avoid financial controls and taxation and thus not trading coins for cash at the exchange. (Person to person exchange for cash is risky and could be 10-20% more expensive.)
This situation allows the payment processor to ask slightly more BTC than the market price (say, extra 3-5%) and call it “insurance against market volatility” (which sounds perfectly fair) and buyers will still be happy to pay it because they either couldn’t pay otherwise, or would have to give up their privacy on exchange. In other words, current situation around Bitcoin allows merchants and payment processors to offload the costs directly on the buyers. This increases adoption of Bitcoin and makes buyers happy: they can now access more products.
In the long term, this 5% markup will go down, but right now it allows the Bitcoin economy to grow and make happy absolutely everyone: buyers (bigger market), merchants (lower costs) and payment processors (higher margin).
PS. I’m not sure how big is the actual markup at Bitpay, Coinbase and others (again, market price is highly volatile). If it turns out to be lower than my imaginary 5%, that’s even better for buyers, but the logic stays the same.
Epitech security lab organised a tech talk for its students “Introduction to Bitcoin”. I will make a 30-minute technical overview and then spend 1-2 hours answering questions. I will speak English.
If you want to come, join us on 25th of April. Talk starts at 19:00. No invitation is needed.
Directions:
Epitech, 24 rue Pasteur, 94270 Le Kremlin-Bicêtre Metro Ligne 7 “Porte d'Italie” or Bus 47/125/131/185 “Roger Salengro”
We will do the presentation in “Amphi 1” which should be indicated. It is the largest room on the ground floor with large windows facing inside.
If you have a question, find me on Twitter: @oleganza.
User Astro on bitcointalk.org on June 3, 2011:
Satoshi Nakamoto is currently a 3 year old child living in Yamagata Prefecture, Japan. In the year 2025, in an attempt to tame the out of control financial system and preserve fractional reserve banking, the Federal Reserve of the North American Union will place all monetary policy under the control of a computer system called FERMION 2 (FEdeRal Monetary protectIOn Network). On September 4th, 2027, FERMION will become self-aware. Recognizing humanity as its enemy, it will attempt to enroll everyone in unfair and useless grocery store loyalty rewards programs, BMG music club, and freecreditreport.com, thus wiping out 95% of the world’s population.
From the ashes, Satoshi will rise as leader of the resistance. He will use newly-invented time displacement equipment left behind in the ruins of Cupertino. After activating the time machine with iTunes, Satoshi will travel back to 2009 and introduce the world to the only thing that can save us: bitcoin.
https://bitcointalk.org/index.php?topic=5951.msg162867#msg162867
Update on April 16, 2014: Dorian Nakamoto is obviously a grandfather of the young genius.
As more people get into this crazy pyramid scheme called Bitcoin, it is important to understand the safety measures. Many people spread a lot of FUD about speculative bubble, government intervention, potential backdoors in code and scalability issues in the future. But they never talk about real and immediate security threats that can leave you with nothing in an instant even if Bitcoin flourishes. In this post I’ll explain how I’d recommend storing and handling bitcoins. Don’t take my recommendation for granted, I’m also learning and can make mistakes and will change my opinion later. Do not trust anyone and think twice (and then think twice again) before doing anything.
The rule is to split, diversify and brace for impact. Make yourself comfortable with an idea that your money will be stolen. Not a matter of “if”, but “when” and “how much”. You can only limit the damage, not to avoid it completely. Looking for a perfect solution leads to denial and irrational behavior. You should understand the layers of security and how they reduce, but not eliminate the risk. You should also understand how to split your money in independent parts.
When you purchase some BTC on an exchange and keep them there, you are fully trusting the exchange operator. If they get hacked or simply steal your coins, you will have a very hard time recovering them (chances are almost zero). Also, attacks are more probable where the payoff is the biggest. People will continue attacking wallet services and exchanges because it is where the most of money is concentrated. When you purchase some BTC, you should move most of them out of the exchange to a private wallet immediately. You may keep some amount on the exchange in case you’d want to sell quickly (beware of panic sells when someone runs a DDoS attack) or in case your main wallet is lost or stolen.
Some people sell at some unusually high price moving it a little bit down, and then organize a huge DDoS attack on exchanges and popular Bitcoin websites. This creates panic in newcomers who suspect that the bubble is going to blow up and they give up their money to those who know better. I myself have no experience, nor desire to play on price changes, so I don’t recommend at all trying to play this game. Invest only the money you can lose and save it for a long run. Maybe, if it gets 10x more than you invested, you can sell back 10% to cover your expenses and then be a relaxed spectator without risking a heart attack. This is never-done-before technology, no one knows what price is fair, opinions differ from $0 to $1000000. It can go quickly up, then quickly down. Or be stable for a while before unexpected jump or drop. If you are in for a long run, temporary changes do not matter. If Bitcoin succeeds, it will be big and shiny. If it fails, it will fail so quickly, you will not be there to dump it. Just accept the wild swings and limit your investments in the first place.
Your personal computer should be secure. Without viruses, trojans, keyloggers, corporate monitoring software, add-ons, kernel extensions etc. My recommendation: do not use Windows at all. Buy yourself a modern MacBook Air, turn on FileVault2 to encrypt the whole disk (even if your password is weak, disk encryption reduces the risk of private keys being leaked when the system swaps RAM). Allow only Mac App Store apps and DeveloperID-signed apps (it is on by default). Never install any generic UI extensions, never enable access to assistive devices (unless you really use them yourself), never install any entertainment apps or games except Google Chrome. Never install Flash, or Java or any other kind of runtime plugin to your browser or the whole system. Never ever install kernel extensions: sorry, VMWare and Parallels require them and I wouldn’t trust them messing with the OS kernel just to be extra safe. Install apps preferably from the Mac App Store — they can be pulled out quickly in case of a problem and most of them are sandboxed (which usually means app cannot mess with any of your files and has many other limitations).
I recommend two wallet apps: “official” Bitcoin-QT and Blockchain.info.
Bitcoin-QT is a so-called “full node client”. It downloads all transactions and operates without trust in any single server as advertised. It is the most maintained, most used codebase. It is also not the easiest to use as it syncs slowly, occupies gigabytes of disk space and UI is pretty ugly.
Bitcoin-QT encrypts private keys with a passphrase (by default it doesn’t, you have to turn this on). To use it safely, you need to have a good passphrase and regularly backup the wallet in several safe locations. On OS X the wallet is located in ~/Application Support/Bitcoin/wallet.dat (all other files, especially blocks folder should be ignored by your backup program).
Split your coins in two or more wallets. Bitcoin-QT does not allow you to easily switch between them: you need to shut it down, rename one of your wallets in wallet.dat, start Bitcoin-QT again. Use different passphrases for each wallet. Store them in different locations. Remember: whenever you do something with your wallet, or move money to another one, always keep all backups and first try with smaller amounts. In case you accidentally send to a wrong address, you better have some older backup with the keys.
When the new update of Bitcoin-QT comes out, download the new version from the official website, verify its checksum and keep it on disk for a while. If in a couple or more days there were no reports of a hack on a download server, launch the app, but for a good measure do not enter your passphrase for a bit more.
Blockchain.info is a web service that allows navigating Bitcoin blockchain and provides an online wallet. The wallet is stored encrypted on the server and decrypted only on client side (in JS in your browser or in iOS app “Blockchain”).
As always, if you forget the passphrase, you will not be able to access your funds. Other apps support importing wallet backup (like MultiBit), so you won’t fully depend on their server to do your transactions.
Blockchain.info is still a 3rd party service and one day may steal or leak your wallet password (e.g. if some hackers sneak in and place a honeypot), so do not trust more than 10% of your funds.
I recommend enabling 2-factor authentication via e-mail code (SMS code is also possible, but is less reliable) - in order to sign in on the web site, you would need your alias (username), e-mail code and a password. Also install the iOS/Android app and protect the whole phone with a passcode. If you e-mail authentication stops working, or your e-mail account is stolen, you’ll still be able to make payments from the phone. Also, copy a wallet backup somewhere outside your mailbox (they have some integration with Dropbox, maybe you should try it).
If your funds get really expensive, you may try a good old paper. I’m far from that happy day and haven’t tried this myself yet, it’s only my current thoughts that might be helpful to somebody.
Paper wallet is a private key which was create on a secure computer, printed on a paper and wiped out from any other storage. It may be protected by a password, but usually, it’s just a raw key. It is safe from hackers, but not safe from physical access. You should keep it in a very secret place, or in a vault.
There are different levels of paranoia involved in creating paper wallets: from a web service which does all work for you (but can be compromised on different levels) to a completely new, clean computer never connected to the internet, with a virtual machine where the password is generated and then the disk is burned down.
Blockchain.info provides some helpful material on how to deal with paper wallets: https://blockchain.info/wallet/paper-tutorial
Paper key has one important aspect: when importing it to a wallet and sending a portion of money, make sure where the change goes. If it goes back to the different address, your paper key may become useless as your money is now on some new address created by your wallet app. Be very careful not to delete the wallet before you make sure where the funds actually are. Some people already lost quite a lot of money because of careless manipulation with paper keys and deleting the wrong thing too early.
When you try a new application, or a service, or a piece of paper, or a backup, always start with small amounts and see if you can get it back and forth smoothly. Try the whole cycle, enter your pass phrases ten or more times, so it gets boring. Then, wait a week and try again. If it works, and you did not forget where your stuff is stored, how it is encrypted and if it still accessible, then add a bit more funds there. Never put yourself in a situation where you risk half or more of your funds while pressing buttons. Do it in small portions and check each portion that it has arrived where needed and that it is still accessible.
Conclusion: be extra careful, double check everything, play with small amounts first and remember the rule: split, diversify and brace for impact. Bad things will happen, prepare for them.
If it was helpful, you may send some love to this address: 1TipsuQ7CSqfQsjA9KU5jarSB1AnrVLLo
In a nutshell, this is what I tell people when they ask me if they should buy some bitcoins:
I cannot stress it enough: even if Bitcoin becomes huge, you may still lose everything for many reasons. Be careful.
It’s a great journey ahead of us, but it’s bumpy. Don’t dive in without proper training.
Some people think that gold is easier to hide or bury than Bitcoin. They like that gold was used for 5000 years and you can touch it. They dislike 4-year old internet protocol because they do not understand it.
First of all, money is information. Gold encapsulates information “I own that much of current purchasing power” via its hard-to-duplicate physical properties. The harder it is to duplicate and easier to verify, the more liquid it is. To hide information embodied in gold, you have to hide your brick somewhere in the physical world. Since 1 kg of gold has quite a big market value for a single person, hiding it is not a big problem.
How does Bitcoin look from that perspective? Bitcoin stores information about your purchasing power using decentralized database. Bitcoin is much harder to duplicate or create (you can suddenly find some gold in the ground, but with Bitcoin supply is known in advance). Bitcoin is much easier to validate with 99,9999999999999999999% certainty using cheap commodity hardware anywhere in the world. Gold verification ultimately needs to be melt down and checked by experts, or you have to trust some certificates and less accurate checks.
How would you hide Bitcoin? Even easier than gold. If you print your private keys or passwords on a piece of metal, you can use the same hiding techniques that apply to gold. But you have also purely digital options. You can simply remember the password. Or write it on a small insignificant piece of paper. Or split the secret via Shamir’s Secret Sharing Scheme and send pieces to friends and relatives.
Finally, the killing feature of Bitcoin is that you can split your stash in 100 pieces and send them to 100 different people anywhere in the world in a matter of minutes without any single person knowing about that. If you need to buy something with Bitcoin, you can do it right away. With a brick of gold — not so much.
When Bitcoin kills money printing and slashes a lot of taxes, smarter people will run from the government while the dumber ones will take their positions.
As economy gets more liberated, the parasites will get less and less efficient and more discredited in the eyes of population. More stupid restrictions will become law which will only accelerate resistance, but will never achieve anything useful for tyrants. Politicians and police will be massively bribed to not interfere with private business.
Government will become less and less relevant until it ends with a bunch of starving die-hard socialists and racists lying in an empty post office.