I am relatively new to bitcoin and electrum.

One thing I don't get about electrum is why does it need to have a non standard server?

Couldn't the client connect directly to bicoind servers nodes using jsonrpc?

And about security, couldn't the client connect to multiple servers in parallel and ask for things in parallel (sort of a map reduce interaction) to achieve:
- Speed: you just take notice of the N fastest responses that are consistent with each other
- Security: you are NOT trusting one node, but a bunch of them, just like bitcoind but without the blockchain download hassle.

If that would be true, bitcoin would be completely USELESS, but that is NOT the case.

First, the miner has no guarantee that his/her BAD block will be first to get into the chain. For that he needs a lot of processing power today and even more if we ever get to have bitcoins "on the street" for instant shop payments. The cost/benefit ratio assures that mining attacks are not worth unless for huge transactions that nobody will expect to be instantaneous.

Second, more difficult it is to get a second or more consecutive blocks in the chain is WAY difficult, you will need to monopolize >51% processing power.

Third, I wouldn't believe that other peers and miners would accept a malformed block that contains a double spending like 2 transactions saying that output X is used more than once.

Forth, "inventing" transactions it not feasible without the private key of the bitcoin previous output holder. What is the chance that you guess right a key of 32bytes, even if you know it has to conform to some rules to be a key you are talking about 2 exp 100 posibilities or more... good luck with that!


THE ATTACK I was talking about DOES NOT NEED ANY MINERS, it is cheap and can be carried out with 5 or 10 peers. making bitcoin report (and I would say punish) double spending detections would make this attack impossible just by waiting a few seconds, that could longer depending of the size of the transaction so to keep the cost/benefit attack ratio over a safe value.

I repeat the question.
Is double spending reported when detected?

It not... Why not?
It seems too stupid NOT to report it and also not to punish the transaction issuer.

---
1HV1Ycu5d6Ff2jUCp965Ec9dmswwfGxuwi

Thanks to VeeMiner, that pointed me to the paper:

http://people.scs.carleton.ca/~clark/biblio/bitcoin/Karame%202012.pdf

I learned some interesting facts:

1) It is relatively EASY AND CHEAP to cheat a merchant with double spending if he doesn't wait more than a few seconds to validate the transaction with 0 confirmation.

Ah, this is NOT theory, the researchers MADE IT HAPPEN (to themselves) to confirm the predicted attack success probabilities. With just a handful of malicious peers, one connected to the merchant directly and a few helpers (<10 was more than enough)

2) The only really proper and reliable FIX is to add a very simple CHANGE to the protocol, that is:

When the researchers could perform their test attacks, the bitcoin network was accepting any first valid transaction that arrived to them and discarded the others that arrived later SILENTLY. ( I WONDER WHY??? it seems quite a stupid thing to do not TO REPORT this EVENT through the network)

The SOLUTION is as simple as just to REPORT back to the network of peers any detected double spending as soon as two or more transactions that are still not in a block get received for the same inputs. I would say also all those transactions should be penalized by not making it to the next confirmed block at least.

The researchers say that the protocol already contains a transactions alert message that was unused at the moment and could become the double spending alert. So the change would be minor indeed.

I think this is quite IMPORTANT. Supporting such alert messages would expand bitcoin use cases to instant transactions like buying stuff at the supermarket, paying your bread, etc. The merchant could be quite confident that in 10-20s or less there is no double spending alert the probability that the transaction will get confirmed is quite sure.

So the question is...

Does current bitcoin support double-spending alerts?
If so, where can I read about it?
If NOT? why?
What is the reason? (Are you crazy? what are we waiting for?)

AND THANKS again to VeeMiner FOR POINTING me to this paper!!!

Ok, I already completed this paper.

It basically demonstrates both theoretically and empirically that it was possible and CHEAP for them (no mining at all) to make double-spends to a faked vendor (themselves of course). They just made sure the attacker's peer connected to the vendors peer while they maintained some (FEW) helpers at some points of the network.

Those helpers had a direct connection with the attacker daemon so that they could receive the order "to inject" the BAD transaction on the network when A needed it, that is just a few moments before or after A fakes the GOOD transaction to the vendor daemon.

Interesting facts they proved:
1- Some blocks can take up to 40 minutes to form.
2- Even after 2 seconds of head start of the GOOD tr. vs the BAD one, just 2 helpers used to guarantee the success of the attack most of the time so that the bad transaction is confirmed in the next block and the merchant is scammed (with no company or individual to sue).
And after 1 second the success is sure with JUST 2 helpers!!
Not really that expensive to have two or three modified bitcoind arround, a botnet of a few tenths of processes is easy and cheap to deploy.

3- Bitcoiners suggestion to use a "listening period" is NOT a definitive solution and can be defied by increasing the number of helpers (but still under 10 in most tests).

4- Bitcoiner's suggestion to use observers is only useful while increasing merchant bitcoin deployment costs setup, while a easier solution exists...

The problem here is simple...
Why does bitcoin peers SILENTLY drop double-spending transactions when they detect them?
It is silly they should just REPORT the invalid transactions to their peers (while reject them both in this block)

This is what the paper suggest as the solution just use an alert message (that the protocol already defined) to notify fraudulent transactions.

In this way a merchant, with an typical standard setup and no more hassles, can rely on bitcoin to make "instant" payments:
- In a shop, just make the receipt printer wait 6-12 seconds (configurable by the vendor with a default value considered "safe enough") so that any double spending attack is detected and notified in that time OR print a warning otherwise... so that the merchant can make the client wait for the police...

I prefer this solution to the one I was proposing and that was much more complex.

Anybody knows if these "double-spending report alerts" are implemented TODAY in the protocol?

(If that is the case I would have to admit that bitcoin does INDEED support "instant" transactions securely enough,
if not, why is this not implemented already?
what is the problem for such a small change with so obvious returns to be rejected?
still to much self-complacency?)


AND THANKS to VeeMiner FOR POINTING me to this paper!!!



 

Then this paper should be wrong:
http://people.scs.carleton.ca/~clark/biblio/bitcoin/Karame%202012.pdf

"In this paper, we analyze the security of using
Bitcoin for fast payments, where the time between
the exchange of currency and goods is ... seconds).

We focus on doublespending attacks on fast payments and demonstrate
that these attacks can be mounted at low cost on
currently deployed versions of Bitcoin.

We further show that the measures recommended by Bitcoin developers for the use of Bitcoin in fast transactions are
not always effective in resisting double-spending; we
show that if those recommendations are integrated
in future Bitcoin implementations, double-spending
attacks on Bitcoin will still be possible.

Finally, we leverage on our findings and propose a lightweight
countermeasure that enables the detection of doublespending attacks in fast transactions."

I haven't read it all yet, I am curious about the "lightweight countermeasure" they purpose...

But this proves that the one saying "this is slow" had a point!

So what do you reply now?

According to this:
https://en.bitcoin.it/wiki/Securing_your_wallet

"Since the backup does not have the private keys necessary for authorizing spends of these coins, restoring from the old backup will cause you to lose Bitcoins."

The backup restore operation is inherently UNSAFE!
Is that stupid thing to do? Specially if they don't offer any alternative.

Why would you EVER "want to forget" keys?

The way I see it is:
1) When I backup the tool should create a copy of all keys know to that moment. (what it does now)

2) When I restore the tool really MERGES keys on the current wallet and on the backups wallet. If i do it on an uncorrupted newer version of my same wallet this will be a no-op. Restore=merge, it should JOIN FUNDs without transactions, this merge wallets operation is USEFUL and SAFE while restore is DANGEROUS and should be avoided most times.

3) If you really want to "restore" the tool should warn the user that keys maybe lost from the resulting wallet, and ANYWAY make a backup of the previous wallet state to a separate file BEFORE replacing any keys and letting the user know that file was craeated and can be used if there is a need.

I don't get it... what am I missing here?

Again,

what's wrong with upgrading bitcoin support instantly confirmed decentralized payment in addition to the ones it already supports?

Can anybody explain what problems do they see in that?

I am sorry, but I think everybody here is wrong in a way or the other, even bitcoinologist.

It all depends on the Use Case...

1) Online monetary transactions: Bitcoin RULES!
Bitcoin is faster than anything else that existed BEFORE (eg. paypal). I think anybody already agrees in that...
...BUT "faster" doesn't mean "instantaneous", bitcoinologist is right about that, instantaneous is "SECONDS" NOT "MINUTES". Here many bitcoiners blinded by lack of auto-critic or excess of love for the tool/currency are changing the meaning of "instantaneous" to avoid admitting that there is room for improvement in that area.


2) Online goods/services transactions: Bitcoin RULES!
Here is where I think bitcoinologist is wrong. Most, if not all, web services and web shops are don't deliver instantaneously. It is perfectly fine taking 10 minutes or an 1 hour to receive an email for a new DNS service that will have to propagate anyway, plane tickets, train tickets, goods that are going to be delivered later in the day, etc.
When the costumer clicks "buy", you just check the transaction when through and reply "Congratulations! your purchase was successful!"... which DOESN'T mean that you are NOT checking its confirmations on the background and can revert or suspend the goods or service delivery if anything wrong is detected 1 hour later.


3) "In person" transactions: Bitcoin is inconvenient most of the time.
Merchants of small goods/services cannot wait 10 minutes for 1 confirmation or 6h for a the proposed confirmations. Bitcoiners replies to this are unsatisfactory and a symptom of too much self-complacency on bitcoin and lack of auto-critic:

 - They say the merchant should risk the money cause it's unlikely that it will be lost. But again, the merchant could disagree, even if it does that with VISA. Cause in VISA at least both the buyer AND the VISA company are identified and the merchant could sue someone, while in bitcoin the money is lost and the buyer might not be identified most of the time. Imagine Carrefour or Walmart, for instance, they probably would never accept bitcoins due to fear of some hacker group coordinating an attack on their shops attacking it to steal lots of goods on a prepared coordinated fashion.

 - Others say that they could use a service provider, but that violates the philosophy that you can can be (if you prefer) your own service provider on all matters.

The facts remains that:

1- Bitcoin mining is what makes bitcoin so useful and successful BUT...

2- But at the same time, making all transactions speed depend on mining speed does not sound right.

3- Bitcoin potential success means scalability problems: most clients mobile phones already and soon PCs wont be able to keep up with the block chain size and growth traffic, so they will have to trust external providers... just like the ripple protocol consensus does.


My proposal, that none has taken a minute to tell me what they see wrong about it is:

Why doesn't bitcoin aim for a decentralized instant (meaning seconds) transactions like this?

A) At some point, a second blockchain is created, it can be linked hashed to a certain block. This new block chain does have very little to zero mining reward. Lets call it "the fast block chain or FTB".

B) The original block chain or OTB is untouched, it works just as it does now, maybe with smaller rewards to allocate the ones in the faster block chain.

C) Transactions can refer to outputs on both block chains.

D) The FTB could work like the OTB but using a mining cycle of 3 seconds instead of 10 minutes. Or, without mining, it could use a 3 seconds consensus algorithm just like in ripple to close a block.

E) Transaction fees on the FTB will be higher on the FTB than on the OTB (but don't need to be expensive) to reward FTB miners/servers.

F) Transaction load on the OTB will decrease, making small transactions that don't pay fees faster or even able to get through at all. (Imagine what would happen on an single unimproved bitcoin chain) to tiny transactions when handling thousands of tps, many won't get through for many, many blocks. Others may never get through cause there is always another more rewarding, even if its days/weeks/years younger.


And I don't buy the go to ripple then... Ripple weak spot for me is their lack of mining. One company created all the XRPs? I don't trust that. Mining is better than that, the problem is that what is good for mining may not be so good for "instantaneous decentralized transaction confirmation".

I think they are wrong in this case.

The guy doesn't stop talking about gold, and nobody says bitcoins are going to compete directly with gold or even other fiat money. Bitcoin will be successful ONLY just by taking away a part of paypal's and Visa markets. Those are their competitors and only with that its future is guaranteed as an exchange medium for some type of transactions. That might help stabilizing its price and allowing it use as the "digital poor's man value keeper".

BUT...

Taxes are NOT going to disappear and they are going to be paid in fiat currencies. No question. So fiat money will keep on being king over gold and bitcoins. I don't think bitcoin will ever be the unit of count in the near future, the dolar or some other fiat will be doing that.

People will be using gold the way they were using it (as an insurance) before bitcoin, maybe a little bit less, and way less than most austrians say we should. And there are reasons for that...

Today gold is rare and strange to most people, this builds some extra barriers to its use on top of the ones the governments/States created:
How do I recognize gold? I am not used to handle it.
How do I know my gold is pure 24 carat and not some yellow jewellery type 18?
How do I hoard it? shall I keep it hidden at home?
Most people (in "stable" countries) is not used to that with fiat money and with gold is even worse, even small quantities are very valuable and too easy to steal.
Shall I rent a bank security box? how much is that a month? how long till they get to keep all my gold?
OR Should I invest in companies or gold ETFs instead?... Warren Buffet will tell you YES!, even if they are not gold related, companies produce stuff and choosing the right ones will grow wealth (for you and everyone else), while gold just tries to keep old wealth from fading, "it doesn't do anything".
...
All those are REAL barriers of entrance to most people using gold instead of fiat money.

Due to that and the forced uses of fiat money you end up in a situation so that you cannot use gold to buy goods or services, neither in physical shops neither online.

Gold usefulness TODAY is not that great.

And even worse, even if you could use gold on shops, with current prices it's too valuable to be useful. Will you buy bread with a coin the size of a microchip? you'll need tweezers and very good aim!
Even with silver that is starting to be difficult.

AND gold CAN'T, by definition, be used online directly. You will most likely need to use gold-tickets or promises in paper or in digital form while the gold is deposited somewhere that it will not move around too often (it would be too expensive). That is you need to trust a company/bank, pay them a rent for the gold monthly, etc.

Bitcoin, on the other hand, has obvious advantages over fiat money and is more practical for many of todays uses that gold. Its strengths shine most on online trade and the ones to be worried are, as I said, Visa, Paypal and other current fiat money based online payment services.

Is bitcoin a bubble?
Well I can tell you what I believe its going to happen.
The current price, "bubble" or not, can't keep on going up like this. Sooner than later it will collapse violently, there is no question of that.

The real question is, how deep will it collapse? will it drop to zero?
I don't think so, whatever naysayers say, bitcoin solves real problems and it is useful per se, there will be always people behind it for some types of transactions.
I bet it won't go even below last peak price on 2011, or it won't stay so low for too long.
Yes, many people will get nervous and lose lots of money and flee to never come back. But many other will stay, cause bitcoin is useful.

Bitcoin current volatility is for sure not a good thing.

But the only thing that I think could kill-off bitcoin for good is an alternative cryptocurrency with better properties like, for instance, having decentralized instant payments confirmed in seconds, having a way to represent personal debts and webs of trusts and move them around, etc.

Yeah right, try to buy in a supermarket or a restaurant. You can't wait 1h in a queue!

It won't happen, even if bitcoin success grows those use cases will still be out of its reach...unless it updates itself.

I was trying to discuss a fix, nobody discuses my design proposal, instead only say they don't need it...
ok fine!

It wouldn't so strange that tomorrow a new cryptocurrency arrives that gives you all that bitcoin does (it may start as a clone) and also gives you instant assured decentralized payments and your bitcoin suddenly is not the best available option.

Think a mixtere of ripple and bitcoin. That can happen inside bitcoin or ouside of it.
Which one do you prefer?

I personally think it's better inside than having to change cryptocurrency all over again.

It is because I find bitcoin interesting that I am here.

A bit of auto-critic won't hurt the community, but the absolute lack of it may...

It is not, for its current user base and use cases... but it might be soon if it is more successful...
and it is most certainly "NOT instantaneous", saying that would be a lie.

So basically everybody here says Bitcoin as of today is SO perfect and doesn't need ANY improvement.
Not even if, being so perfect, it gets much more successful than it is now!

Very interesting...

I am know remembering that phrase from Mr Wolf at Pulp Fuction:
"let's not start sucking each other's d... quite yet."

Corrections, is 6x0.5Gb is 3GB an hour and that is 72GB a day, still BIG!

Ok, this is the kind of response I was looking for, reasoning about design and its consequences... And replying to the post before this. I would be stupid to start to code something or even to purpose a new design BEFORE arguing with others more experienced in Bitcoin its possible pitfalls.

For a computer system 10 minutes is an awful lot of time, even 2 seconds is an awful lot of time. Computers lives in milliseconds, nanoseconds...

If the problem of that "instant transactions block chain" is just that mining would not be worth it, then we could just remove mining prizes altogether on that block chain. The incentive on that chain will only be transaction fees, the prices of "instant security" and for mining you have the parent traditional block chain that will work just as it does now.

As for the last part "Unconfirmed transactions are nearly instant./to pull off a double-spend is nearly impossible" doesn't change the fact that the restaurant or any merchant or receiver will probably prefer to have confirmation before letting the costumer go. The system should aim to provide that confirmation.

Another question, what is wrong with ripple consensus? (for "instant" transactions)

And, have you considered the scalability problems on 10 minute block transactions if bitcoin ever reaches VISA traffic?
Sooner than later bitcoin-qt will be made obsolete, at least trying to keep the block chain locally on a PC.
TODAY is ALREADY not feasible to have it on your mobile phone, you need to trust someone else service.

2000tps*512bytes= 0.97 megabytes per second (from here https://en.bitcoin.it/wiki/Scalability)
In 10 minutes that means that the block will be 0.5GByte, in an hour that means 3.5GB (more or less)... if your wallet was of for 24h then it has to download 84GB for each day offline!
You need a new HD every week! (with todays technology)

These are real problems to take into account.
(They can probably be fixed more or less easily, but first we need to accept we have them.)

Thanks, but coinbase doesn't seem for me.

I live in Spain, and they don't seem to support Euros exchange or European/Spanish banks.

Please, reply to this:

Are you (all) saying that WE cannot have a decentralized system THAT confirms my transactions in 12-18 seconds instead of 1hour.
Interesting...
Why is that?
Why is 10 minutes such a magical timeframe?

Why 10, and not 5 or 20?

Nobody is saying that Bitcoin isn't a VERY GOOD idea...

It really is!!

What I say is that it could be fixed to be an EVEN better idea without losing ANY of its current uses and properties.

What you have to understand is that for something to be "a good money" it needs to be able to be used ON ANY KIND of transactions, not just one or a few of them.

Instant transactions (meaning in seconds) are too important NOT to support!

Otherwise another currency that can do what BTC can and ALSO supports transacting in half a minute will replace bitcoin in many uses and will devaluate your BTCs.

It's just like food stamps, if they can ony buy food there are less liquid, less useful that those stinky dollars or euros that don't have that limitation.

I know many think of bitcoin just as a niche currency & payment system, but to succeed even in being that it has to be tackle the decentralised instant payment problem or at least be honest about it and concede defeat, saying it is not really that instantaneous, just faster than traditional fiat based payment payments.

Still 10minutes is way too long for some transactions.

I am sorry??
So you can waste 1hour of your time to buy bread of several hours on the supermarket checkout queue?
Good for you, you probably don't have anything more interesting to do or maybe you just don't need to work for a living!
But most people can't wait that long and it is not consumerism, is just like money always worked, many transactions need to be instantaneous.
PERIOD


Are you saying that WE cannot have a decentralized system THAT confirms my transactions in 12-18seconds instead of 1hour.
Interesting...
Why is that?
Why is 10 minutes such a magical timeframe?

Why 10, and not 5 or 20?


Nobody complained about that in this thread. I did not question bitcoin's security.
I just challenged its speed.


Good for you, I am 1 week into trying to get some TBCs and it will probably take me a month more to be able to buy some. It is too difficult and SLOW and there are many untrustworthy and greedy intermediaries. Just like in the fiat world or even worse.


Good for you again!
That use case is good for Bitcoin right now, and nobody here is trying to change those use cases.

What I am saying here is that either bitcoin should remove the word "instant" from their publicity or they should find a way to really allow "instant DECENTRALIZED payments" when they are needed.

I think that can be done with little redesign and cheap fees, more expensive than the traditional ones on the mining blockchain, but cheaper than those of Paypals, and VISAs.