Topic: [POLL] Possible scaling compromise: BIP 141 + BIP 102 (Segwit + 2MB) - page 7. (Read 14376 times)

The 2nd strawman argument that you've made against me.


My argument is that diplomatic tension is rising between the states that adjoin the South China Sea. One of those states (Phillipines) has switched allegiances from one major military superpower (US) to another (China) just recently. The president of The Phillipines has denounced international organisations strongly (threatening to burn the UN assembly building to the ground, for instance), and is being investigated by the International Criminal Court for crimes against humanity.

Separately, the US, China and Japan are all having a stand off between one another over disputed islands right in the middle of the exact same region.

And if that wasn't enough, one highly unpredictable, volatile nuclear armed state (North Korea) is exhibiting increasingly bellicose rhetoric towards ALL parties.

If that doesn't sound like a plausible tinderbox ready to catch fire, I don't know what sort of scenario you would actually consider potentially dangerous. It's a serious threat, and it's only 1 out of many possible serious threats. It's likely there are even threats to the internet's operation that no-one could have imagined, hence my focus on margins of safety and risk analysis.


Now, how is any of that either implausible or a circular argument




If you're actually interested in responsible scaling, maybe you could investigate them too? But I doubt that you are, you've spent far too long defending your entrenched and dogmatic position, when you don't really have any good arguments to support it.



Which part of "explore the possibility" do you not understand?

The orphan rate has been improved vastly now we're at version 0.13 - 0.14. The amount of slack that frees up COULD be used to safely reduce the block interval, but that would have to be proven on a test network before any proposal could even be made.


And maybe you have something to remember; Segwit itself is a blocksize increase, of 4x. A 4x increase is NOT to be underestimated in impact, increasing resource use by a factor of 4 is very large for a peer 2 peer network.

Internet was originally designed to withstand a Nuclear War.

However advancements in technology are such that a Targeted EMP Pulse designed to destroy an entire Country Electronic infrastructure is not out of reason.

So lets say China uses one on the US,
what happens their is no electricity , no internet , hell, Fiat is not even worth a damn, so who will give a shit about BTC.

However if you save your private keys on paper and can escape the aftermath to a friendly country and are not killed on site for being an american.
Then you can renter this other country and use BTC , because the rest of the world was left untouched and continue living and BTC continued normally.

However the US has safeguards in place to prevent an EMP pulse from China or anyone else, even from that little space weapon they think the US does not know about.  




The Odds of BTC dying from being not being able to process a transaction or the price of processing a transaction is a Bigger Direct Threat to BTC that what you are worried about CB.  

Good Book on EMP , is One Second After by William R. Forstchen .


 

FYI:
If only this magical bandwidth issue that keeps you you at night,
the easy solution is this, when large blocks start causing too many orphans problems.
Decrease the size of the blocks and move to a faster block speed.  
ie: https://bitcointalksearch.org/topic/m.18206691

 

Is it more circular than your argument "I think it's plausible, so it's plausible?"


I haven't said that I'm against an alternative to an increase of block size and I think it's good that you mention them. If these alternatives exist and are capable of solving the problem, why aren't they more thoroughly discussed in the scalability debate? (Serious question.)


These two look interesting. What is the combined potential, in transaction capacity per MB, of these two measures? (Is the second one Bitcoin-NG? Why was it not explored?)

That one will be most probably been rejected by most of the community because it modifies the original reward schedule and has its own dangers (more orphans). And it would not solve the problem of a large blockchain (that also applies to the Bitcoin-NG proposal), but could improve the propagation and validation of individual blocks.

Edit: Don't forget that the original intention of the proposals we're discussing here is to achieve approval for Segwit by miners of the "big blockers" fraction to get rid of the stalemate. I for myself, for now, would be perfectly happy with Segwit alone for now and then let the Core devs decide further scaling measures.

It would help your credibility a great deal if you weren't putting massively exaggerated words in my mouth. I specifically did not say "most or all" trans-oceanic cables are at risk of breaking, I gave a specific (and HIGHLY PLAUSIBLE) political scenario where some important cables could be broken. Your arguments are still circular, you proved me wrong by saying "it's improbable, because I say it's improbable"

Again, why are you planning for optimistic scenarios when we're supposed to be treating this like an actual engineering problem, where that approach FREQUENTLY AND DEMONSTRATIVELY LEADS TO FAILURES NO-ONE FORESAW. Where is your safety margin?


Your imagination for transaction rate increases seems strangely limited to the block size only, considering I asked for an ACTUAL RATIONALE.


Why not improve the efficiency of transaction encoding FIRST, which is also a hard fork

Why not explore the possibility of decreasing the block interval target FIRST, which is also a hard fork

Why not explore the possibility of separating tx and coinbase blocks (and making the tx block interval more frequent) FIRST, which is also a hard fork

Why not do everything that isn't the most dangerous option FIRST?


With some combination of ALTERNATIVES, WHICH EXIST, the x3 increase in tx rate from 4MB could achieve the same exact effect that your inherently more dangerous suggestion does. Why not play it safe, like an ACTUAL ENGINEER. It's only a highly important development in human progress, after all

It's very, very difficult to imagine an event like you're pointing out (e.g. where most of all trans-oceanic cables are broken). It's even more difficult to imagine that the Bitcoin transaction volume would stay as high as before the catastrophe. So even if all what you imagine comes true, block size would be almost for sure much smaller (and block propagation temporarily faster). But I don't want to profundize here as I consider that event highly improbable.


If we really want mass adoption (>1 billion users), we would need >12 MB blocks even to enable LN users to open one payment channel per year. And LN needs to enable its users to close the channels when they need it (one on-chain transaction more per user), because otherwise the counterparty could steal the money. Imagine if a hub tries to steal Bitcoins from their users. We would have instantly millions of unconfirmed transactions because of all users wanting to close their channels.

So your "maximum of 4 MB" would mean Bitcoin user base being capped permanently at about 50 million users, even if all the transactions were done in a second layer/off chain manner.

As already said, I don't want 12 MB now, but a gradual plan to move towards possible mass adoption.

Some people just want more on-chain capacity and secondary layer solutions. As I've mentioned earlier, it is quite possible that Bitcoin Core will go in the direction of a dynamic block size or flex caps (unless the individuals silently changed their opinions on this), see here:
https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-December/011865.html

Is it me, or does that sound incredibly reckless? Don't you understand how fragmented and bottlenecked a widespread mesh-net could be if it wasn't very carefully planned and diligently protected? I'm in favour of that kind of solution, but your casual description is not capturing the full extent of the problem, a successful neighbourhood or town level mesh-net is a whole different ball game to covering an entire land mass with interconnected mesh nets. How are we going to use mesh nets to cross the oceans? (which was the crux of my point anyway)

And what is your ACTUAL RATIONALE for wanting 12 MB blocks? I've not seen anything in this thread that isn't a circular argument

@Carlton Banks: Well, then we in South America take over Bitcoin. We've resolved the last conflict on our continent and you can all die, but Bitcoin will survive

Seriously, the Internet is decentralized, there will be a network ready to manage 12 MB blocks. Mesh wi-fi networks also could be a solution. And even if a extremely catastrophic event causes the Bitcoin network to stop processing blocks for a while, then we eventually could start it again.

@Lauda: I have thought about starting a new thread. For now, I'll wait a bit more and see, because I'm reading currently a bit about Bitcoin's consensus and how one could avoid miners given too much power, so my conclusions could be discussed in the new thread. Here, for example, I discussed a possible counter-measure (proposed by another user in the German language forum) in case miners ignore completely the majority's wishes - it shares the "spirit" of UASF.

My slide rule doesn't need batteries.

I don't think it is very fast at hashing though...

america are hating china.. not because there is some chinese guy sleeping on the american peoples door steps. but because america has been watching too much fox news.

while we had the "everyone hate russia" moment for the last decade.. america was actually doing trade deals with russia.. literally giving russia the front door key to the international space station and a taxi licence to charge americans to be taken to the door and open the door for america to access the space station.

you can buy apple iphones, mcdonalds, and thousands of other american things all in russia.. all while fox news was spewing out to hate russia..

bringing politics back on topic:

and now we have core who gave the pools the vote and bypass node consensus. core changed the fee estimation, relay and allowance lines of code..
but good old coindesk (barry silbert: DGC: blockstream) is using the media band wagon to try swaying people to blame china for bitcoin issues...

never accept what has been spoonfed for you by script writers. even if 100 people repeat it.. instead do you own research.
blockstream changed the fee mechanisms, blockstream changed who gets to vote. but blockstream didnt expect 50% undecided resistance and 25% opposition. they thought it was a easy domination with soft strokes of peoples asses.. but now realise they didnt get it soo easy.. so now they wanna ram the ban hammer hard and blame it all on the pools or other nodes for causing a split..
either playing the victim card of the fake first strike defense tactic as a way to stroke asses to pretend they are protecting bitcoin when really its about gaining domination of bitcoin.

if anyone dares reply that they want to avoid consensus and happy to have blockstream control it and no one else can or shoot keep them on their toes... then just reply "centralising is ok"

Not following geoplitics then I take it?

No cataclysms, super-giant meteorites or alien invasions necessary: all it could take would be something like the Philipino president or the North Korean president provoking a serious conflict with the US. That sort of thing happens constantly, the US establishment is always "policing" states that present them with an opportunity to do so.

The danger is that China would step in to help. And that's a very real danger; the South China Sea is strategically prized by the Chinese government, they already have an independent dispute both with the Japanese and the US over the various islands in that area.

And guess what strategic assets traverse the sea bed of the South China Sea and in the Pacific Ocean surrounding the Philipines? And no, it's not shapeshifting lizard creatures from outer space

You think that it is likely that in the coming years, internet capacity goes down ?   Without some big cataclysm ?  Seriously ?

No, you're exaggerating

Good engineering practice involves planning for marginal cases, and what I presented represents a LIKELY possibility, not the HIGHLY UNLIKELY scenario you used. Please use rational arguments, not hyperbole.

That kind of argument is a fallacy (I think it is called "fallacy of catastrophic expectations"), because, even though unlikely, the hypothesis is not impossible, the hypothesis makes that the problem at hand is not an important problem any more.

This is like "one shouldn't put a roof on a house, because when there's a meteor impact like the one that killed the dinosaurs, the roof will not resist".  If there's a meteor impact like the one that killed the dinosaurs, the problem of the solidity of your roof is not a problem worthy of attention any more.

If society is so fucked up, by technical or political reasons, that computing and internet connectivity falls drastically backwards, "crypto currencies" are not something to worry about any more. (trying to survive from the dictatorship or trying to find food, is).

You don't want people discussing the proposal unless it's to sing it's fallacious glories? I am discussing the proposal and it's implications, it appears you just prefer the most optimistic scenario possible, which of course is engineering suicide.

And you have zero argument, only mockery? I see

Suggestion to OP: Close this thread, start a self-moderated one with a more exact approach to the proposal. This is the only way to filter out the noise if you only want to discuss the proposal and its implications.

I'm prepping Enigma 2.0.

So that's what you would you do if such a situation occurred? Complain directly to reality itself?

I know you don't like the argument, that's why you're trying to wish it away. But political developments of all sorts could cause such drastic things to take place, and who knows how long for, and that's not at all implausible. The worst aspect is exactly how uncertain we can be about the severity or duration of such a possibility. But 12MB blocks though, right?

What is the actual basis of your argument for any of this, all you've done is this thread is jump on the linear x=y big blocks bandwagon without providing any reasoning as to what the purpose of the big blocks will be. Don't say scaling, lol

I'm going to invest in a factory that makes 56kbit line modems.

Assuming the Bitcoin network would be able to handle 1.5 MB base + 6 MB weight is a fantasy scenario now? I think you're in the wrong thread.

Why stop there? 1 MB might be way too much as well. I don't like this argument one bit.

My grandma told me to learn to calculate with paper and pensil, because if all pocket calculators would one day fail I could still do my sums with paper and pencil.