[UPDATE: 2015-05-10] Bitcoin Core soft-fork "No Forced TX Fee" v0.10.1 available

[ShadowOfHarbringer]

2012-03-25 Update:

NFTF tags for mainline client versions v0.5.3rc1, v0.5.3rc2, v0.5.3rc3, v0.5.3rc4, v0.5.4rc1, v0.6.0rc2, v0.6.0rc3, v0.6.0rc4 released.

https://github.com/ShadowOfHarbringer/bitcoin-nftf/tags

Trunk/master has also been fully updated to the latest mainline version.

[Onichan]

Edit: Never mind it does work. I was just testing a wallet that only had .00051 in it and it refused to send anything without a fee, but if I used a wallet with more in it then it works.

Ok so I finally successfully built the windows version, but for some reason it is still requiring a fee. I have tried both 5.3.1 and 6.0rc4 and command line version without success. I checked the source I have and it does have your wallet modification. Would there be a reason why it wouldn't let me send fee-less? I am using gitian to build it.

[ShadowOfHarbringer]

Edit: Never mind it does work. I was just testing a wallet that only had .00051 in it and it refused to send anything without a fee, but if I used a wallet with more in it then it works.

Ok so I finally successfully built the windows version, but for some reason it is still requiring a fee. I have tried both 5.3.1 and 6.0rc4 and command line version without success. I checked the source I have and it does have your wallet modification. Would there be a reason why it wouldn't let me send fee-less? I am using gitian to build it.

Hmmm... this is indeed peculiar.

However, there can be reasons it won't let you send without fee. For example, i believe the algorithm (in my fork and in official client) doesn't allow sending very small amounts, or amounts that does not have enough confirmations.

I have not removed all safeguards against sending money without fee, just some of them.

If you want to make sure it works as it should, you can build 0.3.21 version of the mainline client and compare the functioning of that with NFTF. If it is the same, then it works.

Unfortunately, i dont't have time to test it now, because I am going on holidays. When I am back (about 8 april), I am going to test it thoroughly.

[Onichan]

It does work as intended. I was just trying to send from a wallet with too little in it. You don't need to test it, but thanks.

[ShadowOfHarbringer]

It does work as intended. I was just trying to send from a wallet with too little in it. You don't need to test it, but thanks.

Glad i could help.

[randomproof]

I've not tried your fork yet, but does it just ignore the suggested fee or does it tell you about the suggested fee and offer the option of ignoring it?

[ShadowOfHarbringer]

I've not tried your fork yet, but does it just ignore the suggested fee or does it tell you about the suggested fee and offer the option of ignoring it?

It just ignores the fee if there is high probability that it won't be necessary.

It is quite safe for everyday usage, if you keep certain rules (eg. not resending money which don't have enough confirmations yet). Sending money which have at least 7 confirmations should be 100% safe, always.

[ShadowOfHarbringer]

2012-05-08 Update:

NFTF - versions 0.6.0.7/0.6.1 & 0.6.2 released.

Fresh tags -are avaiable for download as usual.
https://github.com/ShadowOfHarbringer/bitcoin-nftf/tags

Trunk code has also been merged from mainline client:
https://github.com/ShadowOfHarbringer/bitcoin-nftf/tree/

[gmaxwell]

It is quite safe for everyday usage, if you keep certain rules (eg. not resending money which don't have enough confirmations yet). Sending money which have at least 7 confirmations should be 100% safe, always.

Please do not tell people that it is 100% safe. This is not true and the whole concept of thinking users can tell which money they're resending is wrongheaded, the client has free run to pick from whatever inputs it wants in the wallet. I have helped users unstick their wallets several times as a result of this patch.

I've cautioned you about this in the past— and you responded by pointing out that you were direct about the risks and consequences. Now you violate that by claiming that it's 100% safe.

Yes, it's usually not problematic— after all the overwhelming supermajority of transactions do not need to pay a fee in any case. So yes, in any case where a fee wouldn't have been applied this is safe— but also pointless.  In cases where waiting a few hours would have made it no longer need a fee then this is safe enough (at worst you'll need to leave your client running for a few hours before your transaction goes)... but in cases where weeks would be required your funds will end up unusable for a long span of time unless someone helps you effectively hex edit your wallet.

[ShadowOfHarbringer]

It is quite safe for everyday usage, if you keep certain rules (eg. not resending money which don't have enough confirmations yet). Sending money which have at least 7 confirmations should be 100% safe, always.

Please do not tell people that it is 100% safe. This is not true and the whole concept of thinking users can tell which money they're resending is wrongheaded, the client has free run to pick from whatever inputs it wants in the wallet. I have helped users unstick their wallets several times as a result of this patch.

I've cautioned you about this in the past— and you responded by pointing out that you were direct about the risks and consequences. Now you violate that by claiming that it's 100% safe.

This is a misunderstanding.

I said:

Sending money which have at least 7 confirmations should be 100% safe, always

I i wanted to say that I am certain that it is 100% safe, i would say that "IT IS 100% SAFE" or "I AM CERTAIN THAT IT IS 100% SAFE".

If that is not clear enough, i don't know what is.

[TehZomB]

I just get spammed with errors. Am I missing something?
http://pastebin.com/dpnDMAAj

[ShadowOfHarbringer]

I just get spammed with errors. Am I missing something?
http://pastebin.com/dpnDMAAj

Try compiling the mainline version and tell me if it produces the same error.

[TehZomB]

Yes it does

[ShadowOfHarbringer]

Yes it does

So you will be better off asking about this the core developers (Gavin Andresen etc).

But perhaps it is a common error. Try google first.

[TehZomB]

Will do, thanks

[jevon]

b) the ability to add a transaction fee to a transaction that you've received and that hasn't yet made it into a block…the client would do this by creating a new transaction with that transaction as an input and sending coins back to yourself (less the desired tx fee)…the client would immediately broadcast this new transaction

b) ... The method suggested will not work anyway, as the coins in the pending transaction can only be re-sent by the recipient, who is not the sender.

The sender can add fee by re-sending the change.

It wouldn't be a big deal to make transactions always have change. There's usually change anyway.

I think that's a problem that would fix itself if clients had this feature and people started using it.  If you're a miner and are actively seeking fee bearing transactions, you really shouldn't be rejecting fee-less transactions if they are inputs to fee bearing transactions that meet your fee requirements.

I agree.

[andTo86]

What if there were different tiers of difficulty - if you only include transactions without fees, your difficulty is 'x', if you include transaction fees, it's 'y'. Dedicate one of the 2016 blocks to transactions without fees, so that if someone F's up, they definitely get processed within two weeks. There would be some shifting / weirdness until the difficulty of the 50+fees vs. 50+0 evened out between the two different tactics, but it should eventually police itself such that it's reasonably equally profitable.

[drakahn]

If there were a transaction with a lot of single satoshi inputs, and a 0.0005 fee, would it ever confirm?

[ShadowOfHarbringer]

If there were a transaction with a lot of single satoshi inputs, and a 0.0005 fee, would it ever confirm?

Haven't tried that yet, you are welcome to be the first

[drakahn]

If there were a transaction with a lot of single satoshi inputs, and a 0.0005 fee, would it ever confirm?

Haven't tried that yet, you are welcome to be the first

I may have to try it soon, i have tens of thousands of single satoshi inputs, if it works i could make a satoshi dice single satoshi cleaning client that only see's single satoshi inputs and inputs the size of the fee (0.0005 would be good) and sends with whatever minimal fee will work