Topic: Please do not change MAX_BLOCK_SIZE - page 10. (Read 13023 times)

hey retrep could you take a look at post #191 and tell me if i understand this correctly.

Bitcoin is designed to be a peer to peer digital cash. Read the Bitcoin white paper. Nakamoto wrote that BTC nodes would eventually use large amounts of bandwidth and be run by specialists, so that's what it was designed for.

$30 transaction fees, due to a 1 MB block limit, would make regular transactions impossible. It would convert Bitcoin into a high powered money that only BTC-banks could handle. I don't want a BTC-bank to hold my private keys and be the third party intermediary to my transactions. It's bad enough BTC has to rely on centralized exchanges. Forcing people to store their BTC at banks and process their transactions through them for straight BTC transactions would make the problem of centralization exponentially worse.

Remember it's still very early in this game. Even taking the abnormally high hashing rate into account blocks are barely pushing 150kB average; don't forget that at the conference there were just a dozen vendors, mostly representing internal Bitcoin businesses or investors. For all we know we'll find out that this whole payments stuff never takes off the way people keep hoping, and where it does take off is certainly going to continue to be the "bad actors" Peter Vanesse says he wants to crush.

The Bitcoin Foundation is not Bitcoin. In terms of technology all they can do is pay people to write code; open-source projects can and do fork if they have too. At the same time from what I heard at the conference the legal and lobbying strategy of the Bitcoin Foundation looks to be very much in our interests: I had a really good discussion at the conference with the foundations' chief lawyer Patrick Murck and their strategy of lobbying to make virtual currency trades unregulated. It may or may not work, but it's not going to do us any harm.

It was quite funny at the speakers dinner at the conference how I wound up sitting next to Mike Hearn... and a bunch of finance and investor types. We, well, minus Mike, talked about where Bitcoin was headed and what it's role in the world will be. At that table, and indeed at the whole conference, the finance crowd seemed to really get how what makes Bitcoin special is decentralization and freedom from authority, and how Bitcoin's future lay as a reserve currency rather than an inefficient and inconvenient payments system. You don't remake the worlds financial system by focusing on penny bets and ad-impression scams first and foremost.

Of course, I doubt the high finance crowd spend much time on troll talk...   

Bitcoin is designed to be distributed peer to peer currency first and foremost.  Certainly more than any importance of "microtransactions"

If allowing microtransactions means that you need a T3 line and terabytes of harddrive space to run a node, then your microtransactions will have to go, because otherwise it simply consolidates power and destroys the entire point of the network, which becomes datacenter-to-datacenter-to-bank-to-government-to-datacenter

You can also see that satashi fully intended to be able to use everything over Tor, which again is impossible with bloated block sizes

Blocksize should scale such that the average peer with average internet (broadband, roughly speaking ~50-100kb/sec upload/many MB/sec download currently) and average computation and storage resources can always participate as a full node

Right now I'm running a full node on a VPS because the traffic is a bit rough on the home connection. For people who are invested enough in Bitcoin and want to see it stay decentralized, this might present a reasonable compromise. I keep my wallet in an SPV client and sponsor this node to help keep the network strong.

People already do this for tor nodes and have been doing it for a while. I don't see much reason why people who have benefited financially from the bitcoin network and want to see it continue to succeed would be very averse to taking on this responsibility. It is not something everyone needs to do, should need to do, or even should do. It is one very practical way of adapting to the changing realities of bitcoin though.

Bitcoin Megastore, are you aware of this project for a software solution which goes a good way towards alleviating node bandwidth loading?
https://bitcointalksearch.org/topic/m.2135237

Sage advice which I will take.

Before I go, I'll just like to say screw Gavin and the Bitcoin Foundation horse he rode in on...or at least rides into the future.  I'm sorry to not be more hopeful, but at this point I see Bitcoin to be not much more than a cash cow to be milked.  I'm counting on the army of mouth-breathing think-they-are-libertarians here to continue to help me line my nest with $USD and they have not let me down yet.

I do have confidence in the 'free market' and have confidence that it will eventually produce something the people of world actually need vis-a-vis distributed crypto-currencies.  What we definitely don't need is some corporate controlled one-world currency solution which is exactly where this thing is headed to be blunt about it.

I hope I'm wrong and you are right that there is still some hope for Bitcoin proper, but I'm not betting on it.  Were it not for Garzik and Maxwell I would have zero hope.  I will participate in what efforts I might be able to in terms of experimentation and such as they will likely produce useful data and code no matter what happens.

Don't give up that quick. Gavin represents the United States-centric, FinCEN fearing above-ground side of Bitcoin; the world is a lot bigger and a lot more diverse than that. I suspect Peter Vanesse and his "we'll help regulators crush bad actors" antics are a much smaller part of Bitcoin than you might think.

Speaking of, reminds me of an email I got yesterday:


When I first heard from this guy he sincerely wanted to help out Bitcoin by moving his timestamping needs to my OpenTimestamps service to try to avoid UTXO bloat, and now he's pissed off enough to be adding to the problem as a political statement. Of course, that's a tormail address, and he's been using my service through Tor according to my logs; I suspect he doesn't share Peter Vanesse's thoughts on Bitcoin...

Having said that, nothing wrong with taking a break from the forums - easy to get yourself exhausted by it.


I will say though, Gavin's proposed 75% threshold is kinda funny - right now it would take just three or four pools to meet it.

I really gotta make my decentralizing mining project with pooled-solo mode happen: https://bitcointalksearch.org/topic/decentralizing-mining-with-pooled-solo-mining-221164

There is always rationing.  The question is how.

In a mature system with a market, I'd let the market figure it out.  That's what market are for, and that's what markets do.

But we don't have a mature system, and we don't have a market.  Not only do we not know what a transaction really costs, we don't have any way to find out.  You are arguing as if I didn't believe in markets.  I do, but I only believe in real markets.  I have absolutely zero faith in the ability of a market that does not and can not exist to provide useful information.

In the absence of useful information, it is prudent to be careful.  Since backpressure on the transaction volume does not reasonably exist yet, and whatever actions we take will have serious long term consequences for the entire world, we should not just remove the limit, nor create dynamic rules that are too easy.

When we do need to increase the limit, I would propose the following rules:  Block max size increases iff at the time of difficulty change, the sum of the size of the last 2016 blocks is > (1814 * block_max_size).  If size increase is indicated, block_max_size+=(block_max_size>>4).  I'll leave the implications as an exercise for the reader.  1814 and 4 are magic numbers, they could be changed, but I suggest they not be any smaller than specified.

sounds like a good idea, thanks for explaining.
I'm gonna vote NO, at least for the next 12 months.
however I do have some mining shares and I wish you did too

My, how clever and original you are!

I'm try very hard to avoid the tech section in order keep in order to keep it marginally useful.  I'm out now.  I'll sell you some uBTC on the Google 'send money' thingy in a year or two.  So, see ya in hell.

No, absolutely not. The process for a hard fork looks like:

+ Get rough consensus that the change is necessary.
+ Write the code.
+ Get it reviewed and thoroughly tested.
+ Release software that will support it when X% of hashing power agrees

... where X is a super-majority (like 75% or more).  If 70% of hashing power disagrees, then it doesn't happen. Miners will express support by producing block.version=3 blocks (just like they are now producing block.version=2 blocks that MUST include the chain height in the coinbase transaction).

It is possible the X% threshold will never happen if 1MB is plenty big enough. It is possible it will only happen when transaction fees start going up and pressure increases on pools to make their blocks bigger (or maybe merchants tired of paying high fees figure out they'll save money by mining or operating pools themselves, will get X% of hashing power, and will increase the block size).

Again, I spent a lot of time at the conference talking with people about the block size issue, and there is definitely consensus that 1MB just won't be big enough eventually. That has nothing to do with microtransactions, normal growth in "macrotransactions" will bump up against the limit in a year or three.

So you are OK with like 70% of the haspower going to a fork that did not accept your change?
I mean, you do not seem to be concerned about it at all. Like you just assumed "whatever I decide, will be the right choice".
People may just not agree with you. Especially people who run mining pools and basically only their vote counts.
But not disagree with you because they are mean - maybe, like me, they disagree with you from good reasons.
Do you even have any proofs that increasing this number is necessary?

Try taking your meds first before thinking. You might at least generate something coherent.

Ya know what?  I think you've check-mated us decentralized folks.  A big block can eat a little block, and the SPV's are pawns who have no choice but to upgrade.

I'll have to think about it a bit more, but I think the you and your employer are going to win this one.  So, carry on and make me rich.  My BTC will flow, but they will be to had only dearly.

This + 1000.  It would be wonderful if developers raised the hard cap and miners at least initially didn't make larger blocks.  The cap should only be a security mechanism.  I disagree that no cap is necessary but the cap should be high enough that the economic decisions of rational miners are setting a "defacto" cap lower than the hard security limit cap.

Bitcoin is really getting old, when some political decisions have to be made

Which "bigger powers" and which "people" ?

Is there some secret cabal out there I don't know about?

If you mean "Peter Todd has convinced some big mining pool operators not to increase the size of the blocks they create" -- then great!  That's the free market at work, big mining pools should be free to create blocks that are as large or as small as they like, and to accept or reject other's blocks for whatever reason they like.

exactly. you cannot really say what is a cost of a transaction until you let the fee system to work and set it at some level.
then you can try to optimize it, with the block size and current house technology.
but not before.
it's just crazy - mad scientist is less crazy than it