Bitcoin Forum

if this idea helps, I'm glad, if not, moderator can trash this topic - no hurt feelings.

regarding dust in the blockchain, I propose an enhancement to the wallet code that automatically submits non-fee transactions sending all 30-day old dust to 1) another address in that wallet that isn't dusty/empty, or to 2) the bitcoin foundation donation address.  There are around 810 thousand addresses with dust currently.  These non-fee transactions would eventually empty those dusty addresses...that then can be pruned out of the blockchain.

People can make any tx they want however you can't force a miner to include that tx.

So it is both "solved" and not solved.  There is no central authority which could force a resolution.  If people want to make dust consolidation tx they can and if miners want to mine them they can.

Did you jump right to CHOICE #2 and miss CHOICE #1?

I agree, approval is core. To account for that, how about offering the user an "opt-in" option to turn the "feature" on for their wallet, plus make the "option #2 address" be a fill-in field.  Dust could be defined as 1/100th of a normal Tx-fee...so that as BTCs go up in value, the amount deemed "dusty" would stay insignificant in value.  The wallet could popup a tip every month or so suggesting the user should take advantage of the "dust cleanup opt-in" option as a way to keep the blockchain cleaner/faster for everyone.  Tips can be "turned off", too.
Would that be helpful?

A better solution would be to change the default relaying rules and coin selection policies to favour (in the form of lower minimum fees) transactions which shrink the UTXO set.

Then clients could be configured to perform dust collection by automatically including dust outputs as inputs to other transactions, assuming the user doesn't mind the privacy loss this represents.

I like this idea.  I create dust collecting transactions for my wallets myself manually, and while they do get mined at a reasonable time even without transaction fees at the moment, I'm not sure whether this will also be the case in a year or so.  Believing that this dust removal ultimately helps bitcoin, though, I think it would be great if this kind of behaviour could be encouraged for those who know about the privacy implications and are fine with them (personally I use mixers anyway when it comes to transactions that are linked to my identity).

Deep misunderstanding of how the blockchain works detected.

One address or one thousand, the number of addresses doesn't matter. The blockchain is not aware of addresses, addresses are a wallet construct. What matters is the number of txouts. Sweeping dust is fine and good, but addresses don't come into it.

thanks for the feedback on the misunderstanding - I'll dig into the docs so I don't waste your time.

Choice 2 is a pretty bad idea.  In fact, if the devs included it, they could reasonably be accused of theft.

The rules are:

     It is smaller than 10 thousand bytes.
    All outputs are 0.01 BTC or larger.
    Its priority is large enough (see the Technical Info section below)

So, the wallet would need to have at least 0.01 in dust.

I would suggest changing "All outputs are 0.01BTC or larger" to "All outputs are 0.01BTC or larger, or it has 1 output and at least 5 inputs".

Tidying up dust would be encouraged.

The priority of a transaction is

1 BTC: 1 day
0.1 BTC: 10 days
0.001 BTC: 1000 days

Transactions that merge dust into a single coin aren't given much incentive.

On problem is that if you make tidying up dust easier, you also decrease the incentive to not create dust in the first place.