Topic: Time To Grow Up (Read 2918 times)

Oh, just noticed another line from the lyrics I quoted on the first page.

"The track is on some battling raps, who want some static?"
(the static in question is obviously that of a block size cap)

Hey, Bitcoin! Want some static? Of course you do!

Haha, the (Uni)Verse has some nice sense of humor.
Enjoy!

One more thing, reaching consensus isn't about respecting the differences in opinions or popular voting, it's about finding a single underlying layer of truth and sticking to it. Not everyone will be able to see it or understand it, but if enough of the brightest of us do, the rest will follow.

That's how mathematicians prove their theorems.
Let's look at what isn't the correct solution first:

---

1) Arguments against removing the limit or placing it too high.

The home-based demographic of Bitcoin is important. That's where Bitcoin was born, that's how it gained its value. The large network of nodes enforces the rules of the game by its mass. It creates a solid playground for other participants (miners, merchants, exchanges) to make their moves and flex their muscles. It's like a boxing match, where the home-based nodes are the spectators. Don't forget about the butterfly effect, where a subtle change to the configuration of the system may have a tremendous effect on the outcome.

2) Arguments against keeping the limit unchanged or too low.

Bitcoin gained its value by slowly increasing its effective capacity, while being guarded by a high enough static hard limit. That's the only configuration that managed to survive to this point and the only one that is known to work reliably so far. There are no precedents of a crypto-currency that has been effectively capped on its transaction volume and become successful. Don't forget about the competition either, being arrogant and ignorant about it is the best way to get out of business quick.

3) Arguments against floating/flexible limit.

Letting miners or any other full nodes to meddle with the limit will create the situation, where everyone begins pursuing their own interests, while the idea of consensus would start falling apart. A single static limit (which is easy to implement, easy to enforce, easy to follow) is the only guardian of the consensus that represents the interests of Bitcoin as a whole. If in this contingency situation we put those interests above our own, we will all agree.

---

This reduces the space of correct solutions to a single static hard limit, which is not too high and too low. Now, taking into account the limitation expressed by Chinese miners and the current pre-fork capacity of Bitcoin's closest PoW competitor, the actual figure is not that hard to come by.

The current situation is not caused by mistakes of the devs, their misunderstanding or any other circumstances. It's the test for the community. You see, the truth is simple. Only the best of the best deserve to have Bitcoin and now is the time for us to prove that we do.

Remember, you are not done until you're running the software that represents the idea of consensus as outlined above, so let's get busy. In the game of Chess only when you commit to a move, the clock starts ticking on the other side. I'll see you there.

They require a softfork that would reintroduce advance scripting functions so that the SPV proofs used to move coins between chains can be recognized by the mainchain.

Do sidechains require any modification to the Bitcoin protocol?
If they can work with Bitcoin as it is now, what is the problem with them?

I invite you to re-assess the parameters of the system (including the block size cap) not only based on technological considerations, but from the perspective of incentives structure as well. The fact, that we haven't seen enough consolidation in the space of full nodes, is because the rules of the game clearly indicate that there is a ceiling on the level of bandwidth that the network is willing to tolerate with its current size. Don't sign up to the rules, that you likely won't be able to follow in the future.


Why shoot in the dark? It's like trying to solve the problems of 2020 or 2036 today, without even knowing what they are. Don't underestimate the psychological effect of the limit. That's the only reason we are having this conversation about the change. It's a sync-point (or barrier in pthread terminology) at which we all arrive to see who is running and where we are going. Without these the inertia of the process might build up to the point, that any attempts to hold it within the path, that best represents the goal it was created for, simply won't gain any momentum and instead fall on deaf ears.

There must be a single static limit firmly cemented into our perception of Bitcoin in order for it to have any effect whatsoever, but big enough not to limit Bitcoin's growth within the safety threshold determined by the current state of technology.

Do you run a reachable full node? I did, and these numbers don't surprise me at all.

And where are you getting $13? 1.1GB per day (as in paper) amounts to 400GB per year. 500GB hard drive costs about $60 in my country.

AFAIK sidechains use the same nodes as Bitcoin. Note it might have been a bad idea of me to mention sidechains since they are not a proper scaling method.. I'm sure there are a couple of magic tricks they can pull given a little more trust added but they are still limited themselves by the block size.

I don't know about Lightning but sidechains requires nodes just like the bitcoin blockchain.


I guess it could work but that's a big if.

I'm not sure what is this sidechain nodes you refer to... Bloat is dealt with by layers on top with more efficient ways to deal with transactions, see Lightning as an example. The important thing is not to externalize the cost of this bloat to the Bitcoin nodes which are not rewarded for their work.

If merged-mining is enabled miners can effortlessly mine different chains at once (like they do with Namecoin atm)

Look I'm not here to argue that decentralization + having a maximum of full nodes and security is not important. It is and I do care about that a lot just like you do.

But how does that solve any problem about bloating? The bloat will just go elsewhere btw and there will still need people to run sidechain nodes with massive bloat.

Another thing, how do you expect to keep most of the miners on the main chain if most transaction volume happen on sidechains? How do you know if a sidechain will, at some point, become much more profitable to mine than the main chain? You don't, just like myself but risk is very real because big transaction volume = big potential transaction fees. You should think about the consequences of this very probable scenario.

Oh... that... well..it would be a legitimate point if it made any sense at all..unfortunately it doesn't.

Without nodes there's no Bitcoin and without Bitcoin there is no payment channels, sidechains, off-chain services, nothing.

So basically by running a node you are not only supporting Bitcoin but all of its ecosystem. Imagine when nodes get commoditized and every one can set easily set up one and maintain it a little to no cost, the ultimate decentralization!

Of course that can't happen if you bloat the blockchain so much that you'd need specialized hardware/datacenters to run it.

Basically, who will care running a node at all?

Which is?

You didn't address my point at all.

Get rid? No.

We should just be realistic and tell people it isn't conceivable nor desirable to put all their transactions on the main chain.

Bitcoin will scale using payment channels, side chains & off chain services so that we can guarantee the robustness of the foundation layer 

So basically you are suggesting to get rid of all the companies that got into bitcoin for other things than remittance? We should tell people to outright stop investing and innovating on top of that open network because we don't really want it to be THAT open after all because we fear it cannot scale while in fact it perfectly can?

Meanwhile some people are expecting that somehow a massive amount of nodes (thanks to the artificially kept small blockchain) will be run on a network that nobody can really use but a few niche areas.

This makes absolute no sense to me.

How on Earth do 8MB blocks consume 1MB per second traffic?  You would have to make the assumption that each and every full node must serve a legion of leech nodes.

1 GB of space per day works out to a cost of 3.6 cents per day - a whopping $13 for the entire year.  With today's prices.  And assuming all blocks magically 100% filled.

I don't know what your numbers are based on, but take a look at Bitfury's calculations (Table 2). Currently, in the worst case scenario, my PC could handle 8Mb blocks, but I wouldn't bother running a node that consumes 1MB/s of traffic and eats more than 1GB space a day.

Once again,  BIP 101 is not the end-all be-all.  It's a first stab at the problem, and a reasonable one at that, imo.

But let's take the worst case scenario: technology ceases to improve whatsoever, transaction volume skyrockets enough to fill all blocks 100%, and all the miners (irrationally) choose to jam every block completely full.  In that world, a reasonable desktop PC running on a home connection, and with hard drives that today cost about $300 will have zero problems keeping up and storing the data, and operating as full nodes, until at least the year 2023.  In that made-up, dystopian apocalyptic future, we would have to (gasp!) consider a fork... eight years from now.

Is it really reasonable to assume zero technology improvements in the next eight years? Of course not.

The top three reasons are perfect examples of worst possible assumptions, on which no sound engineering (especially of decentralized consensus-critical payment network) can be based.