# Last edited on 2015-08-10 21:02:07 by stolfilocal
# NOT POSTED

Consider requiring a minimum transaction fee of 0.3 mBTC (~9 US cents) for each transaction output, in the consensus rules.  Possibly with some extra fee per kB only for extra-long outputs our extra-long inputs (but not for many inputs).  Consider also imposing a minimum output value equal to this minimum fee. 

Consequences:

+ It would cut down frivolous transactions, such as advertising spam, that currently exist only because someone else is paying for them. 

+ The reduction in traffic will, quite probably, delay saturation of the network by a couple of years at least, without any of the alleged harmful effects of increasing the block size limit;

+ In fact, it will quite probably reduce the mean block size for a couple of years, hopefully reversing some of those negative effects;

+ By charging for outputs, not for inputs, it encourages the consolidation of UTXOs, and discourages their reckless creation.

+ By charging a fixed fee for each output, it makes the fees understandable and predictable to clients.

+ While still small compared to the true cost of the network, it restores some sanity to the economics of the service, by requiring those who use the services fo the miners to pay (even if partly) for them.

+ At current prices, the proposed fee is large enough to encourage miners to promptly process all unconfirmed transactions

+ The proposed fee is not so large that it will make bitcoin uncompetitive for e-commerce;

+ If the price rises or falls by a factor of 5,  the fee will still be reasonable, so it does not need to be changed 

+ By being fixed in the consensus rules, it removes uncertainty and complexity from all client interfaces.  

+ It makes RBF and CPFP unnecessary, so that clients would not have to remain in line. 

+ Since clients choose the outputs, but have little choince about inputs, charging for outputs is more consistent with common business sense (price is proportinal to value of service to the client, not to internal costs that are not under his control).

+ A fixed fee provides a weak negative feedback tending to stabilise the price.

+ The drop in traffic may harm bitcoin's image by showing that usage and adoption are much less than what is commonly believed.

+ The traffic will become more reliable as a measure of actual usage.

+ The min fee and min output will make spam attacks more expensive hence less likely.  

+ The reduced traffic will also make spam attacks more difficult and their backlogs will clear much faster.

+ Changing the min fee and min output parameters will reuire hard forks and reaching a consensus on the best value.

+