Gold collapsing. Bitcoin UP.

[cypherdoc]

I don't think it would.

I don't think it would either.

The only time we've ever seen these spam attacks is when we were getting up near full blocks when the attacks became cheap to pull off and provided a target. They jacked the mempool and then didn't have to pay losses as most of the unconfirmed tx's just fell off after 24-48h.

If you listen to Gavin's epicenter Bitcoin interview, you realize that Core dev needs to be trusted. Gmax is not to be trusted. He's lied in his negative rating of me back on BCT by saying I lied or misled him in our PMs about HF. I called him out on reddit about this in reddit the other day and he couldn't provide any evidence to refute.

We're talking about leading a multi billion dollar financial system. It will never allow someone like him to keep his current role.

 

[molecular]

finally got most my nodes back up from that ddos attack. one more to go. what a PITA.

what DDOS attack? (I was on vacation for a couple of days)

I received some abuse reports for my node. At one point the hoster said I was being attacked, next moment they claimed my host was being used for attacking.

Can someone link me to relevant discussion?

 

[Cconvert2G36]

https://www.reddit.com/r/bitcoinxt/comments/3j28mo/udp_flood_ddos_attacks_part_ii/

 

[Erdogan]

I like the mathematics and physics concerning centralization, but let me just state the obvious, which works for almost anything:

With scale comes efficiency, it is easier to get high productivity on a larger scale. A large company can buy an excavator, but a small company can not buy a half excavator.

On the other hand, a small company has less administration. The bigger the company, the more distant all the workers are from the goal of creating profit. In the end, the free market will find the right size.

I think miners have already come to an optimal size distribution. Larger blocks could possibly change that a little on the technical efficiency side, but nothing on the administrative side.

Cypherdoc: Thanks for the invitation back on bct

 

[awemany]

I'm working on V2 to clarify the two points of contention. I'd like to make it public sometime around the ScalingBitcoin conference (September 12).

Awemany: interested in reviewing V2 of the Fee Market paper?

I read and understood your first paper, so yes, I am interested in reading and giving feedback on V2.

Regarding math skills or credentials, I am just a random Internet dude

I like the mathematics and physics concerning centralization, but let me just state the obvious, which works for almost anything:

With scale comes efficiency, it is easier to get high productivity on a larger scale. A large company can buy an excavator, but a small company can not buy a half excavator.

On the other hand, a small company has less administration. The bigger the company, the more distant all the workers are from the goal of creating profit. In the end, the free market will find the right size.

I think miners have already come to an optimal size distribution. Larger blocks could possibly change that a little on the technical efficiency side, but nothing on the administrative side.

Indeed. There are other decentralizing effects, too - one example: Someone in a country with Internet access but a failing currency and strong capital controls might fire up a smallish miner to convert failing fiat -> gas -> electricity -> Bitcoins. [And, yes, I am sometimes a bit worried about long term ecological consequences of Bitcoin]

 

[cypherdoc]

Since this supposed large miner efficiency advantage exists this very moment, then why isn't f2pool continuing to press it's advantage by releasing a series of self constructed multi input single tx 1MB blocks like it created in July?

 

[sickpig]

BitFury just released a white paper on the block size increase:

http://bitfury.com/content/5-white-papers-research/1-bitfury-report-on-block-size-increase/bitfury-report-on-block-size-increase.pdf

 

[cypherdoc]

-355

I was told only Bitcoin was volatile.


-400

that's not good.

 

[Peter R]

Since this supposed large miner efficiency advantage exists this very moment, then why isn't f2pool continuing to press it's advantage by releasing a series of self constructed multi input single tx 1MB blocks like it created in July?

Increasing the block's size has a direct cost to the miner's profitability (there's a greater chance of orphaning), so this isn't worth it. Besides, the other miners could still SPV mine of the block header while validating this block so this wouldn't work for other reasons too.

The advantage I'm talking about is due to the miner not having latency when mining on top of his own blocks. It only applies when the miner solves two (or more) blocks in a row. Like I said before, this is a small theoretical advantage, not unlike other economies of scale.

I read and understood your first paper, so yes, I am interested in reading and giving feedback on V2.

Great! I'll forward it to you when it's ready. Very little has changed, but I've expanded on the Shannon-Hartley stuff a bit to hopefully better explain why the "healthy fee market" result still applies if block solutions are compressed (e.g., using the relay network or, in the future, using IBLTs).

Regarding math skills or credentials, I am just a random Internet dude

As am I. And as Greg says, "wearing a beer hat and arguing from the sidelines."

 

[cypherdoc]

yes, i'm just being rhetorical.

first, of course there should always be some latency.

second, Governing Dynamics.

third, jurisdictions like GFC ensure a degree of built in purposeful latency.

only now is tx growth catching up with our technical capabilities, ie bandwidth, but still is lagging behind. the system surely can handle larger blocks today as my full node stats suggest. the only reason China has been able to build up a dominant market share to date is b/c block sizes have been small. it's time the system outgrows them to leverage today's and tomorrows bandwidth improvements elsewhere.

 

[Inca]

I think the technical issues regarding bitcoin scaling up 10x where we are today are trivial.

What is more concerning is the continuing lack of a decent US exchange with sufficient order depth to prevent price manipulation like we see on finex and okcoin regularly.

I still believe this situation will sort itself but massive price volatility which shakes the price so violently keeps away tomorrow's bitcoin users...

 

[cypherdoc]

BitFury just released a white paper on the block size increase:

http://bitfury.com/content/5-white-papers-research/1-bitfury-report-on-block-size-increase/bitfury-report-on-block-size-increase.pdf

still working my way thru the paper but i still ask, why does everyone think that miners can't already control the size of their blocks and esp if there is no limit:


also, why shouldn't miners, large or small, be free to harvest the tx fees from a spam attack in a no limit situation? if they want to take the risk of an orphan, let them. if they don't get orphaned, great, they just made a bunch of coin processing the spam tx's and clearing the mempools of everyone resulting in growth of the system's tx processing. at some point, an equilibrium will result as orphaning increases in frequency or #full nodes start dropping out.

don't forget that the core dev narrative before the block size debate blew up was that the 1MB didn't have to be changed b/c the large miners had an incentive to keep blocks small for propagation purposes so as to prevent orphaning b/c rewards are more important than fees at this time.

now that everyone has been shown to want an increased block size, Wuille et al flip the argument around and now say that large miners will want to attack with large blocks to force out small miners. no, Governing Dynamics, and the desire to construct small efficient blocks targeting rewards hasn't changed.

 

[cypherdoc]

eventually investors are going to notice that Bitcoin has stopped going down for quite a while now. then, it will go UP:

investors are rightfully anticipating another QE. Janet has no balls:

ouch.

-421

 

[Peter R]

Gmax: "We can't increase the block size limit because they'll be too many orphans."

Other: "But orphans are what drive the fee market -- naturally limiting the size of a block a rational miner will produce."

Gmax: "Orphaning is negligible. We can't expect orphans to drive fees."

 

[cypherdoc]

-465


oh my -510.

Bitcoin holding beautifully.

i know it's shit out there in community but keep your heads up. price looks stable.


i wish Gavin had just become lead dev with a 101 + Core version.

 

[Zarathustra]

Gmax: "We can't increase the block size limit because they'll be too many orphans."

Other: "But orphans are what drive the fee market -- naturally limiting the size of a block a rational miner will produce."

Gmax: "Orphaning is negligible. We can't expect orphans to drive fees."

The delaying tactics prevent txs increase prolongs delaying tactics prevent txs increase prolongs delaying tactics ...
Awesome strategy by the small blockers.

 

[trizma]

glad this thread is up and running even after being censored in an unjust manner in the other forum.
will be following

 

[lunar]

Very good blog post, constructively critiquing BIP 100

https://www.bridge21inc.com/blog/seven-reasons-a-vote-for-bip100-is-a-bad-idea/

 

[cypherdoc]

MrMadden is well versed in Bitcoin and has been around for a long time. You can't fool him.

 

[solex]

i wish Gavin had just become lead dev with a 101 + Core version.

Yep. We are all unlucky because the first two lead devs for Bitcoin made an error of judgment with the 1MB.

Satoshi could have achieved his anti-DoS aim in 2010 by making the block limit conditional
e.g. If block-height < 200000 then max_block_size = 1MB
This would have required consensus to renew it, not remove it.

Gavin did not anticipate how entrenched some community views would be for "no change" to Bitcoin regardless whether a change is for the long-term good. So he allowed 3.5 years to pass until he gave up the lead role, with this massive handicap still in the software.

Does it look likely that Wladimir will be the lead dev to rescue the situation?
No. Unfortunately not.