Topic: Combine txouts that have the same content (Read 598 times)
legendary
Activity: 1862
Merit: 1011
Reverse engineer from time to time
I believe the Hard fork part alone is the biggest reason why something like this should be very carefully considered.
I think I read a thread about this some time ago and with concerns about block sizes and growing transaction volume, it might be worth thinking about this again:
What about introducing a transaction type that anyone can issue (even if the inputs of that transaction can't actually be spent by that person) under the following conditions:
* There is more than 1 input
* All inputs have the exact same script as the other inputs and as the outputs they are referencing
* The resulting single output has the exact same script as all inputs
* OP_RETURN outputs are not used as input
* The transaction pays no miner fee (the sum of all input Satoshis = Output value)
Advantages:
+ "defragment" the UTXO set by collapsing identical transaction scripts into a single one
+ not limited to an "address" system (which people wrongly assume Bitcoin is based upon, since popular websites display it like this), works on the lowest common denominator on the protocol level
+ easy to verify that scripts are identical
+ maybe the only way for decades to spend some of the dust transactions out there
Disadvantages:
- It is not be enough to know what outputs you could have spent in the past to sign a transaction offline, you need to check online which output(s) you can spend right now
- Takes up even more space in blocks and requires miner cooperation to include these for free
- Might have unforseen consequences(?) for some more esoteric transaction output scripts
- Is a hard fork
A potential modification could be that this kind of transaction can only take place after the potential inputs have aged for some time, so people can't immediately spam a second transaction as soon as you recieve a pubkeyhash transaction. On the other hand, miners might anyways take care of the economies of this, as a reduction in UTXO size is in their main interest.
What about introducing a transaction type that anyone can issue (even if the inputs of that transaction can't actually be spent by that person) under the following conditions:
* There is more than 1 input
* All inputs have the exact same script as the other inputs and as the outputs they are referencing
* The resulting single output has the exact same script as all inputs
* OP_RETURN outputs are not used as input
* The transaction pays no miner fee (the sum of all input Satoshis = Output value)
Advantages:
+ "defragment" the UTXO set by collapsing identical transaction scripts into a single one
+ not limited to an "address" system (which people wrongly assume Bitcoin is based upon, since popular websites display it like this), works on the lowest common denominator on the protocol level
+ easy to verify that scripts are identical
+ maybe the only way for decades to spend some of the dust transactions out there
Disadvantages:
- It is not be enough to know what outputs you could have spent in the past to sign a transaction offline, you need to check online which output(s) you can spend right now
- Takes up even more space in blocks and requires miner cooperation to include these for free
- Might have unforseen consequences(?) for some more esoteric transaction output scripts
- Is a hard fork
A potential modification could be that this kind of transaction can only take place after the potential inputs have aged for some time, so people can't immediately spam a second transaction as soon as you recieve a pubkeyhash transaction. On the other hand, miners might anyways take care of the economies of this, as a reduction in UTXO size is in their main interest.
Jump to: