it’s just a curiosity, bitcoin has a limited block size because this ensures that the block chain don’t grow too fast, mass adoption is meant to happen in the layer 2
thinking of that, what happens if you limit the block size to smaller and smaller number? how low can you go? can you have just a few transactions per day with all the rest in the layer 2?
imagine a blockchain so small that everyone with a wallet in a smartphone has a copy, i would also imagine super high fees etc but as long as you can have a functional layer 2 this maybe not as bad as it sounds
Grin blockchain scales mostly with the number of users, not with the number of transactions since range proofs do no need to be kept for spend outputs, only kernel is kept
Full node runs on mobile phone (Grin++ for Android)
If the adoption of Grin would lead to congestion in the future, Grin can start using LN for further scaling on layer 2.
You are wrong in assuming fees need to be high with low transaction volume. Bitcoins mining reward and as such security gets lower with each halving, hence high fees are needed to pay for security in the future. Grin has a linear emission/ constant mining reward. As such we can assume the loss rate will create a soft cap to the total number of active coins. In practice this means the security will stay the same and will proportionally become equal to the loss rate on the long term.
For example, if the loss rate would be 0.5% of all coins yearly, this would mean the security/mining fees will approximate this loss rate on the long term (200 years or so) and will become equal to this loss rate of 0.5% of the total number of active coins per year. Hence, fees will never need to pay for securing the blockchain, they will only serve to protect against spamming the mempool and blockchain bloat.
Grin is roughly as runnable on a mobile phone as Bitcoin. While the past may be 4 times smaller for the same transactions, it requires more operations to verify a block due to rangeproofs.
I’ve seen this being said a few times, but I don’t think that’s correct. Grin, like Bitcoin, scales with the number of transactions because it leaves a kernel per transaction. It’s just that the size of historical transaction whose outputs have been spent is smaller.
To weaken the statement I added the word ‘mostly’, because the kernel is rather small. I included the assumption that at some point there will be a way to remove the need to download all kernels, e.g. compacting, but yes technically at the moment the blockchain scales with something like 7x UTXO +1x spend_TXO
Yes i’m kinda describing Grin, but to the extreme. when i mean small, i was thinking like limiting to 1 block per day with 24 tx or something like 876KB per year, you could store the blockchain in a fridge even if there is mass adoption
all the transactions are meant to happen on the layer 2 or 3 and the first layer just for opening and closing channels
what problems will there be and are those solvable, for example open and closing huge amounts of channels in a single tx
@natoshisakamoto It’s possible, but a bit less flexible since you would only be able to open or close ones a day. I mean if the objective is only to limit the blockchain size, there are other ways. E.g. having sort of a horizon, so users only need to download this horizon snapshot to check against
i wonder if something like this make sense for mimblewimble, can somebody that knows more take a look at this?
they are claiming some pretty “incredible” stuff, like each user only needing to store their own proof transactions and the “spine” of the blockchain, also a POW where each user mines their own transaction
@natoshisakamoto You mean read up on Minima? Here is their whitepaper.
The weakness is that it requires a lot of users to be safe. See their whitepaper.
“This type of system means that if there are no transactions then there can be no blocks. Minima will need to reach a critical mass of transactions to sustain a secure blockchain. The transaction rate will need to be very high. Since all the security of Minima is derived from the number of transactions, the more transactions there are, the more secure the network. There will be a bootstrap period until Minima processes enough transactions to secure the network adequately”
All the transactions are made use a small amount of Proof of work. To me this reads a a very easy to attack chain since you only need to generate a lot of nodes and do a small amount of Proof of Work to attack the chain. And what if few are making transactions? I would guess at that point it would be even easier to attack the chain - confirmed, see above. Furthermore, this was used initially in IOATA, it was only meant as spam protection using some proof of work for transactions, not to replace the real proof of work for blocks. So here it appears they want to use it as ‘real proof of work’, which it was never intended for.
what about the blockchain part, seems like a interesting ideia, it seems that each node also stores the last 2 months of blocks but i don’t know where it could go wrong
