Now the discussion is starting to get interesting....
On 03/21/2018 07:11 AM, Luke Kenneth Casson Leighton wrote:
[...]
*if* people agree - world-wide - to slow down on the mining *then* the hashrate - world-wide - will slow down. however the only way that's actually realistically likely to happen is when the "reward" drops sufficiently (halves every 18 months) for the financial incentive to lower.
*then* when the financial incentive lowers, the number of people mining will also lower (non-financially-viable equipment switched off) and the hashrate will correspondingly drop.
the problem with that assumption is that the people mining will only be motivated by profit. if they're genuinely interested in going over the 50% share in order to corrupt / control the blocks then that doesn't happen, and bitcoin goes to hell in a handbasket.
Exactly, this is something most people fail to understand about mining. There's nothing inherent to the algorithm or the protocol that makes Bitcoin so power-hungry. If the entire network agreed to all shut off the ASICs and run everything on low-powered SBCs running on photovoltaic cells only (for example), the network *itself* would continue to function just fine. The difficulty would adjust itself (eventually, up to two weeks for BTC) to match the processing power available to it. That's the design.
When I first came across Bitcoin, the idea was to let your wallet solo-mine all the time in the background. Pools didn't even exist yet when I first tried mining (on an ASUS netbook in rural Peru, no less), or at least I wasn't aware of them. At the time, I compared it to something like SETI@home, which looks for signals from space-brothers as a screen-saver, basically. Everyone mines a little bit, and everyone gets a little bit of the reward. One CPU, one vote. That was the original intent.
The problem is greed. Some clever bastard decided "Why limit myself to just a single background process?" That's when dedicated mining rigs started, and the arms race began, even before GPUs came into the picture.
i've been thinking about this, beyond what monero does (which plans to hard-fork to increase the random-access memory usage). monero is not possible to do on a custom ASIC because it deliberately requires large amounts of memory (6gb, 8gb). if you were to make a custom ASIC it would *be* a GPU... therefore you might as well buy... off-the-shelf GPU Cards.
beyond that i think you need to specify and agree hard limits about mining capabilities that, if exceeded, result in PENALTIES not REWARDS. such mining contracts would need to be stored *in the blockchain* rather than being hard-coded. miners would be required to sign up to the mining contract (in the blockchain) in order to participate, along-side a declaration of running some CPU benchmark tests which indicate the computing capacity that they are declaring that they intend to use.
if they go OUTSIDE of those parameters, by responding too fast on a block (which can be verified by other miners re-running the same algorithms that they did), they get PENALISED *not* REWARDED.
Yep, the Monero devs are very much dedicated to "egalitarian mining", which I think gets closer to Satoshi's original intent,[1][2] and makes something like javascript browser-based mining possible, which then opens up an entire new set of ethical concerns, even if it's a tiny step towards fair hashrate distribution. [3][4]
What are the other options? POS coins take very little energy, but "the rich get richer", and the more coin you have, the more you stake, etc. Not very egalitarian either. Or some insta-premine like Ripple, where I just make a bajillion tokens out of thin air and then start selling them for a tenth of a cent each? Neither of those seem like better distribution schemes to me. So, how do we get newly mined coins into as many hands as possible?
Definitely an interesting conundrum.
- krt
---
[1] https://getmonero.org/2018/02/11/PoW-change-and-key-reuse.html [2] https://cointelegraph.com/news/bitmain-announces-new-monero-mining-antminer-... [3] https://arxiv.org/pdf/1803.02887.pdf [4] https://www.theregister.co.uk/2018/02/27/ethical_coinhive/