mimblewimble team mailing list archive

[Ignotus Peverell <igno.peverell@xxxxxxxxxxxxxx>]

I don't necessarily agree with all these points but I won't argue either, given that I've gradually become cooler toward fee burning (pun intended).

Certainly, all the points mentioned have some importance (I think Andrew Poelstra was the first to point out the additional incentive of submitting transactions straight to miners), but what resonates the strongest with me at this point is the difficulty to calculate exact supply. It turns out this is a very undesirable property, especially when it comes to a MimbleWimble implementation.

So I'm okay with finally reverting that decision and having all fees go to miners, as typically done on most blockchains. As I was the biggest proponent of fee burning, I'm guessing few will object.

- Igno

P.S. Note that for us, sending transactions straight to miners isn't necessarily all bad for privacy. It prevents anyone else from observing it, as it gets disaggregated in blocks (and potentially cut-through).

-------- Original Message --------
On 22 January 2018 7:30 PM, Casey Rodarmor <casey@xxxxxxxxxxxx> wrote:

> Sorry, last bullet point should be:
>
> • It incentivizes pooled mining, since users will probably only
> submit transactions to the top N accelerators, if you're not in the
> top N pools, you're incentivized to join one to maximize your fees
>
> On Mon, Jan 22, 2018 at 8:21 PM, Casey Rodarmor casey@xxxxxxxxxxxx wrote:
>
>> On Mon, Jan 22, 2018 at 11:59 AM, Fernando Nieto fernandonm@xxxxxxxxx wrote:
>>
>>> Once blocks are full, even if you only burn part of the fees, what makes
>>> you think users will publish transactions with a fee greater than
>>> MIN_FEE, instead of sending a MIN_FEE transaction to a known miner and
>>> then splitting the rest of the fee (that would otherwise be burnt)
>>> between them?
>>
>> I'm curious what others think of this, but to me this is a really
>> strong argument against burning fees.
>> This would incentivize every miner to run transaction accelerators,
>> which seems bad on a few fronts:
>> • Accelerators are bad for privacy, since they give miners a chance to
>> associate IPs with transactions
>> • Accelerators require a bit of extra infrastructure, which means that
>> miners won't be able to switch frictionlessly to and from mining Grin
>> • It incentivizes pooled mining, since unless will probably only
>> submit transactions to the top N accelerators, so if you're not in the
>> top N pools, you're incentivized to join one to maximize your fees