Who cares, it's not like the default value is a network consensus parameter. Anyone can set it to whatever they like, whether 0kb / 0 txs mined or 1000kb. I doubt that people are running huge server farms and can't pull a cli parameter on their own.

Technically speaking, he is right although all that has happened have created a spin on the events that if we don't include cheap spam in the blockchain then ...bitcoin won't grow... Like bitcoin needs spam growth and not legit tx growth.

There are a lot of actors in the forum, twitters, news etc, proclaiming the fullblockalypse, but the reality of the matter is that there is no fullblockalypse due to the fact that actual blockchain use is low and the rest of the free space is topped off with spam. That's why fees don't rise. If every tx was legitimate, and there was an actual crowding of txs competing for the limited space, fees would have gone through the roof.

Anyway, instead of NACK, he could say "ACK with the condition that default min fees also rise to combat spam".

Yes I am aware but if you do it x4 or x10, you must extract all the block rewards. If you see monero 1-1-1-1-1 it's just 5 block rewards, not actual txs that would be 10txs for 10 minutes. That's because I haven't manually removed from the count the block reward for the miner.

How isn't it about fees?

Let me ask you this: If I take bitcoin-qt source, change a few lines of code regarding the suggested fees in order to hardwire it to pay a fixed 0.1$ fee (~100 satoshi / byte) for a 250byte tx, compile it and market it as an alternative client that works every single time, then what would you say? That now bitcoin works predictably and that the network is fine?


Wallet software functionality is not related to bitcoin the network. One wallet could be ok in calculating fees, another might not.

Actually, one year is close to passing and nothing destructive has happened - contrary to his prediction. So it is conclusively proven that the block size increase he was saying was urgent, in fact wasn't that urgent.

Also... Gavin wrote:


Yeah, I mean I will stop using bitcoin because I don't want to pay 5 cents in fees or because blocks aren't issued like clockwork every 10 minutes (this has nothing to do with the blocksize debate btw). It's obvious users would rather pay tens of dollars to centralized paypal, credit cards and bank wires instead of the much cheaper, faster and decentralized bitcoin.

Not even pro-FUDders would write that stuff.

The situation from 40satoshi upwards is pretty similar.

https://bitcoinfees.21.co/

31-40 satoshi per byte (1-9 blocks wait)
41-50 satoshi per byte (0-2 blocks wait)
51-60 satoshi per byte (0-1 blocks wait)
61-70 satoshi per byte (0 blocks wait)

Guarantees for 0-1-2 blocks or 10-20-30 minutes don't exist anyway, as the next block could be empty. Same for time. Next block could be issued in an hour due to variance, or you could have 3-4 blocks found in 20 minutes.

And as I re-read it, I'm confident it speaks of old coins that were to have priority, even without fees. There were 50kb assigned for "free txs" of old coins and the new rule is "fuck that if the system is running on the limit".


So "if system is running at capacity, don't expect free relay even if your coin is old".

I think that's related to disabling the benefits of old coins that were "eligible" for free txs, over those who paid fees (if you hit the mempool limit).

Can you show me the transactions these other competitive systems are doing?

I actually tried a block explorer with major altcoins, and their blocks are ghost-towns.

Last 5 blocks (tx count):

litecoin
3-1-1-7-9

dash
3-11-7-9-1

monero
2-1-1-1-1

eth
0-2-0-7-0

btc:
702-1420-2196-2384-2764

It is highly unlikely 0.12 has anything to do with it, as it represents just 15% of the network. If a tx is not relayed it will be rebroadcast until relayed. However it will stay in the mempool as unprocessed if it doesn't have the right fees.

Propagation delay maybe of very large blocks (?). Preferring to not include a lot of txs to reduce their orphan statistics - something like that. But that is contingent on not finding better ways to broadcast solved blocks.


BTC emission serves a dual role though: You can also see it as an expansion and distribution of the monetary base.

I don't know if there is a third party that can override the miners.

Even, if, say, devs thought that the miners are acting irresponsibly with the issue of spam, they would have to make a minimum-fee hard fork to counter tx spam. But in order for that to be enforced, the hard fork should be adopted by nearly everyone (consensus) - with the miners being the most critical adopters.

So if the miners feel they are doing ok, why would they adopt the hf code?

How can it not work, when it is working right now?

You don't pay or you pay less => you wait more.
You pay more => you wait less.

In the meantime the "less" and "more" are in the 0-2 cents to 4-6 cents range - which is practically near-zero cost.

It will generally be the case that spam pays as little as possible and legit txs pay more - unless the wallet misinformed the user. So legit txs would have no problem going through.

Fees don't add capacity. They introduce antispamming disincentives to make use of existing capacity in a rational way.

Bitcoin is a p2p protocol that can be subject to numerous p2p attacks, including blockchain bloat attacks, mempool bloat etc - provided the fees are low and that there are co-operating miners that accept near-zero-cost txs for inclusion.


First: Why? What will happen if the mempool is full? We've seen stress scenarios, we've seen what happened when private key lists were handed out - the network managed it pretty well. In the pre-0.12 scenario, let's say some nodes crashed due to lack of memory or something. Now this is fixed with mempool management parameters so you don't need to worry about that.

Second: If you can get txs relayed for near-zero-cost, then what's to say that you won't fill the mempool easily, even with 10mb blocks, by broadcasting 10 times the volume of current spam txs?


0.0001 (fixed) is 0.04$ per tx and it almost certainly will get you in within the first few blocks.
0.0001 per kb is in reality 0.000025 / 0.01$ for a 250byte tx so you'll wait until you bump it up.

There is nothing "very good" about it. It's way too low. Near zero cost fees attract near zero cost abuse. Plus it is not futureproof (the network will not be secure at this level, even if you have 2-10-30mb blocks). Plus it does not replace mining subsidy as it should be doing. Plus it does not give any serious motive to miners to mine the txs when the fees of 1mb block don't even get them 1% of the block reward.


The fee market works. If someone's wallet is crap in dealing with fees, then they should use external sites to determine what to pay. If they made a mistake, they should bump their fee up with RBF.


In the presence of a rational mining market, the miners.

Yeah, from one perspective that is true. Still ebay and paypal make a somewhat arbitrary choice that while the risk is there, it will be entirely shifted to the seller's side. The buyer is placed on a pedestal where he can do no wrong, allowing him to scam a seller with any excuse possible.


Fee estimation breakdowns are a user-interface issue, not a block-size issue. If a wallet fails to properly estimate fees, then an external site should be used.

In any case, guaranteed first priority is fairly constant at levels around 50 satoshi per byte in the last couple of months. With RBF it should be possible to bump up fees if one has made an error of inserting too-low a fee.

True. Still ebay and paypal dominate the online selling market, so not much choice out there if you want to hit a large pool of potential buyers. It's like paying the pimp, lol.

At some point -we can always hope it'll happen- perhaps we'll be able to do the same through decentralized markets and bitcoins or altcoins, that actually get decent volume in terms of listings and transactions.


You can stick various types of stamps on an envelope and depending the accumulated value of the stamps you can ask for first priority, second priority, express, etc.

If, on the other hand, you don't buy stamps that add up to the proper value of even the basic service, your mail will be rejected.

The legitimacy of the intent of the sender is not in question in that case but his level of intelligence is in question - if he expects to conduct his business by not paying what he should.

In bitcoin it's actually better because you have hope that even at 0.4cents or 1 cents, you can get confirmed after a long wait - or you might even get added for free (!).

Now, if people don't know how to set fees, or what fees are appropriate, that may be related to ignoring how BTC is used, or some design failure of the user-interface which is inadequate in warning the user about whether he needs 1c-2c-5c etc and how long it will take for his confirmation. Still, I think with 0.12 you can RBF and bump up your fee and be on your way even if you got it wrong the first time.

What it definitely isn't: a block-size issue.

I'm waiting for ebay to start accepting Bitcoins

(insert meme here "still waiting" / skeleton).

The thing is you don't want to be convinced, or change your mind. That's ok - most people are inflexible regarding their belief systems.

The other day I got 250$ in paypal. From sender to recipient's bank account (=my bank account) I lost 20$ in various fees and conversions - and it took 4 days btw because there was a weekend in-between and a "random security check" by paypal that got me +1 day delay.

How is a 3-5 cents fee for an almost instant tx, or 1-2 cents for a few hours tx, "milking the rich" when bitcoin is probably the cheapest payment method compared to banks, visas and paypals?

Sure, why not... but you've already made your mind that block size is all that matters - irregardless of the fact that people are paying near-zero cost fees (whether the low priority 0-1-2 cents guys or the higher priority 3-4-5-6 cents guys), so what difference does it make anyway?