Bitcoin is not the answer for microtransactions; how to rebroadcast unconfirmed?

[BitcoinGirl325]

I was under the impression that Bitcoin would be the answer for sending micro payments over the Internet, but that is apparently not the case at all.

With a .0001 BTC required transaction fee for all transactions under .01 BTC, that amounts to a 6% transaction fee (at today's exchange rates) for a $1 USD transaction. And that equates to a 12% fee for a $.50 USD transaction. These fees are significantly higher than credit card fees, Paypal fees, Western Union fees, and even foreign exchange rate fees.

A few days ago, I didn't realize any of this information (i.e. the required fees), so I tried to send $1 worth of BTC to a friend WITHOUT a transaction fee to show him how "amazing" this new technology is. Turns out to be not very amazing at all, because the transaction never confirmed because of the lack of the transaction fee. Worst of all, there seems to be nothing I can do about it... I can't even add the 6% fee to it now, in order to get it confirmed!

I used the BitWallet app on iOS to send the transaction, and the hash of the transaction is: c238d939b330b682abe1c729300636600254c23c67e1eea4e8daa9abc26dd2c5

So now the transaction is in a permanent limbo state that can't be resolved, and I have effectively paid a 100% transaction fee because the money is in limbo.

Is there anything I can do to rebroadcast this transaction with the appropriate fee, so it actually confirms?

[franky1]

anyone care to try to re-push the tx somehow

01000000019230ce51556db4aa00585cb169d1d09b28abca6eae6ac63c7c1b45645e179fc401000 0008b4830450221008e8a12c169ce28a8e43a5d59c78bf3d7ce392ae5da8ade76d7470430b56482 d202206241d68e400ba465668990d5c319415c28201bbd8da851d9c9814cd6e2b3651a014104572 2135b0e7aca6c5f63b1985b7258d29268692ffd99221a8e6b1af0b1ad943a72ec7633fb6ecfc543 9c1dbf10a490cdfdc2be265a6b086a600fc2263a3c1a86ffffffff02c1900200000000001976a91 4433b6bc44b5e8d37883a0e7038a79ad613fbbc3288ac3b800200000000001976a9145559eef599 61ecb7c0919ba4e66967875863a0f788ac00000000

it seems that blockchain.info's pushtx no longer functions right. instead of re-transmitting unconfirmed tx's, it is saying it already exists. (which it should only say that AFTER confirmation)

[leopard2]

Bitcoin is not suitable for microtransactions just like $100 bills....use a decent altcoin like LTC or DOGE for that purpose and use BTC for larger payments.

People must understand that BTC is a perfect replacement for wire transfers, not micropayments.

As the value of BTC and the size of the blockchain grows this will get worse - for the USD value of BTC, the non-suitabilibity for small payments will not make much difference, in fact I believe that a huge blockchain spammed with small transactions, is bad for BTC's value

[franky1]

Bitcoin is not suitable for microtransactions just like $100 bills....use a decent altcoin like LTC or DOGE for that purpose and use BTC for larger payments.

People must understand that BTC is a perfect replacement for wire transfers, not micropayments.

As the value of BTC and the size of the blockchain grows this will get worse - for the USD value of BTC, the non-suitabilibity for small payments will not make much difference, in fact I believe that a huge blockchain spammed with small transactions, is bad for BTC's value

bitcoin is perfect for microtransactions. but the mining pools are messing with the original concept for their own greed. so its human greed to limit bitcoins potential that is at fault.

[DeathAndTaxes]

With a .0001 BTC required transaction fee for all transactions under .01 BTC, that amounts to a 6% transaction fee (at today's exchange rates) for a $1 USD transaction. And that equates to a 12% fee for a $.50 USD transaction. These fees are significantly higher than credit card fees, Paypal fees, Western Union fees, and even foreign exchange rate fees.

The fees for CC and PayPal are much higher as much as 33% on $1 transaction. Sending $1 by WU isn't even possible but it costs $5 to send $20 (40% in fees).  Not sure why you would think that sending $1 by other means would be cheaper. 

As a heads up future versions of the mainline client will reduce the min fee to relay from 100 bits (.0001 BTC) to 1 bits (.00001 BTC).  Still small txns will never be free.  Low cost yes, but free no.   The fees act as a denial of service prevention mechanism.  

Is there anything I can do to rebroadcast this transaction with the appropriate fee, so it actually confirms?

No but the client will probably try to rebroadcast continually.  A miner may include it in a block but it may not.   It is possible to "delete" a txn and eventually other nodes will forget about it so you can create a new txn but I am not familiar with the wallet you are using.

[Pente]

Dogecoin is more suited for microtransactions. And the OP problem is why competition amongst cryptocurrencies is good.

[Meuh6879]

Example : With android bitcoin wallet, i can only send the minimal 0,06 mBTC (0,02 Euros).
Fees = 0,10 mBTC ...




What is a "micro-payment" ?

[Anders]

bitcoin is perfect for microtransactions. but the mining pools are messing with the original concept for their own greed. so its human greed to limit bitcoins potential that is at fault.

I don't know the Bitcoin details, but wouldn't miners who included lots of micro transactions produce the longest chains and thereby "win" more often?

[franky1]

bitcoin is perfect for microtransactions. but the mining pools are messing with the original concept for their own greed. so its human greed to limit bitcoins potential that is at fault.

I don't know the Bitcoin details, but wouldn't miners who included lots of micro transactions produce the longest chains and thereby "win" more often?

i thought it was the other way round. if 2 miners solve a block. if one block has a transaction that is not in the other block then that extra transaction is orphaned and thus ruining the block.

if only miners all communicated together and agreed to include ALL tx's of a certain timescale. they would all be on the same level. then it would just be purely fastest hasher first.

imagine it that the protocol had it so that whilst miners are mining blockA. they are all receiving transactions ready to build blockB. and there is a unix time limit where the consensus then agrees as the cut off period. this is 10 minutes. so not only will it irradicate orphans. but also help to keep to the 10 minute average, thus keeping difficulty under control.

but no miners would want to do that, would they. agreeing is not in their best interest. competing is.

[shorena]

There was someone yesterday who bought bitcoin from a reseller. The reseller didnt pay a fee, thus it took 5 hours. Its hard to tell how long it takes. Ive seen a week, Ive seen a few hours.

bitcoin is perfect for microtransactions. but the mining pools are messing with the original concept for their own greed. so its human greed to limit bitcoins potential that is at fault.

I don't know the Bitcoin details, but wouldn't miners who included lots of micro transactions produce the longest chains and thereby "win" more often?

No, the amount of TX in a block does not make the block "longer" in terms of the longest chain. Each block is a single link within the blockchain, no matter how many TX it includes (confirmes). The longest chain part is important when it comes to several blocks beeing broadcasted close after eachother, as franky1 allready explained.

[Anders]

There was someone yesterday who bought bitcoin from a reseller. The reseller didnt pay a fee, thus it took 5 hours. Its hard to tell how long it takes. Ive seen a week, Ive seen a few hours.

bitcoin is perfect for microtransactions. but the mining pools are messing with the original concept for their own greed. so its human greed to limit bitcoins potential that is at fault.

I don't know the Bitcoin details, but wouldn't miners who included lots of micro transactions produce the longest chains and thereby "win" more often?

No, the amount of TX in a block does not make the block "longer" in terms of the longest chain. Each block is a single link within the blockchain, no matter how many TX it includes (confirmes). The longest chain part is important when it comes to several blocks beeing broadcasted close after eachother, as franky1 allready explained.

Who then decides how many transactions to include in a block? Can one miner choose to include only one transaction while another miner includes thousands of transactions in the next block?

[odolvlobo]

Who then decides how many transactions to include in a block? Can one miner choose to include only one transaction while another miner includes thousands of transactions in the next block?

The miner decides which transactions to include. Miners will include all transactions with fees, of course. The number of transactions doesn't increase the cost of mining significantly, so most miners are generous and will also include transactions with no fees.

[Anders]

Who then decides how many transactions to include in a block? Can one miner choose to include only one transaction while another miner includes thousands of transactions in the next block?

The miner decides which transactions to include. Miners will include all transactions with fees, of course. The number of transactions doesn't increase the cost of mining significantly, so most miners are generous and will also include transactions with no fees.

Can't a miner start earlier with the mining if fewer transactions are included in the block? Otherwise the miner has to wait for transactions to come in, or? Obviously that's not how it works but how is the timing of the next block determined?

EDIT: The miner will of course already have many transactions waiting when the mining of the next block starts. That explains it I think.

[cbeast]

Dogecoin is more suited for microtransactions. And the OP problem is why competition amongst cryptocurrencies is good.

Dogecoin is where Bitcoin was in 2010. If it sees any success, it will have the same issues.

[commandrix]

I use Bitcoin for any major transaction I want to make and the relevant altcoin for smaller transactions like tipping, so I've never run across this problem.

[Lauda]

Dogecoin is more suited for microtransactions. And the OP problem is why competition amongst cryptocurrencies is good.

Dogecoin is where Bitcoin was in 2010. If it sees any success, it will have the same issues.

Indeed, only a few could make it. Most would have this issue.
Who sends $1 often enough to care about it?

[LiteCoinGuy]

microtransactions = wait for payment-channels. we are early in the game. be patient.

[bg002h]

Micropayment has been a holy grail of transaction research for over 30 years. The DoD funded important research on the 90's to go with this internet thing they made. Bitcoin wallets aren't made for micropayments and the network is only really set up for sequential micropayments between two parties...

If you are really interested in the micropayment topic, see mike hearn et al's work: https://bitcointalk.org/index.php?topic=244656.0

[franky1]

Who then decides how many transactions to include in a block? Can one miner choose to include only one transaction while another miner includes thousands of transactions in the next block?

The miner decides which transactions to include. Miners will include all transactions with fees, of course. The number of transactions doesn't increase the cost of mining significantly, so most miners are generous and will also include transactions with no fees.

Can't a miner start earlier with the mining if fewer transactions are included in the block? Otherwise the miner has to wait for transactions to come in, or? Obviously that's not how it works but how is the timing of the next block determined?

while mining block A.. people are transacting and filling up the mempool ready for blockB. as soon as block A is solved the transactions of blockB are locked in. its not based on a fixed time event or consensus, just purely as soon as last block is confirmed it uses that data with the mempool tx's to make the next block..

..

now imagine the protocol could be changed to a unix time lock of 10 minutes. and al miners used the same filled mempool without discrimination, so that even if a block is mined in 8 minutes it has to wait another 2 minutes to get all TX's of that period to then start hashing the next block.

not only will it fix the 10 minute average per block as each solution will be found in that time scale. meaning that difficulty does not jump up. but also as i said in previous posts idea. ALL TX's get locked in fairly and equally.

then its just a game of who solves the block first, and with a fixed difficulty thanks to the time lock keeping the average block the same by delaying the start of the next block to the 10 minute window.

.. well its a theory that some altcoin maker can try atleast.

[Anders]

microtransactions = wait for payment-channels. we are early in the game. be patient.

I read:

"This article describes how to use payment channels, a way to set up a pending transfer of value from one wallet to another such that the amount that will be transferred is incrementable at high speed and by very small amounts. Whilst this does not allow you to send micropayments at high speed to different recipients each time, many applications can fit within this framework - typically anything that involves micro-billing for a metered service." -- https://bitcoinj.github.io/working-with-micropayments

It would be good if something similar can be done for many-to-many microtransactions. Such as payment channel servers that collect lots of small transactions and convert them to larger chunks of real Bitcoin transactions. That would be powerful if it can be done in a trustless way.