Gold collapsing. Bitcoin UP.

[dwaltrip]

@theZerg Reading this reddit post, I had one question


What prevents an attacker from setting loose a large number of nodes and controlling the block size in that fashion? You described that an excessive number of nodes could result in a similar topology as when the network is partitioned by a single node that has a significantly lower block size limit policy.

 

[Justus Ranvier]

What prevents an attacker from setting loose a large number of nodes and controlling the block size in that fashion? You described that an excessive number of nodes could result in a similar topology as when the network is partitioned by a single node that has a significantly lower block size limit policy.

The nodes that matter most are the exchanges, payment processors, and wallet providers.

The effective block size limit is going to be a negotiation between them and the miners.
[doublepost=1450928976][/doublepost]Chris Pacia from OpenBazaar has some LN skepticism:

https://chrispacia.wordpress.com/2015/12/23/lightning-network-skepticism/

 

[Erdogan]

We have 3 packs in the mining market;

1) 1 MB, smaller, undecided, don't care, don't know

2) Voting for larger blocks in the comment

3) Running large block support mining (XT - bring it on!)

In the node segment, 2 packs:

1) 1 MB, smaller, undecided, don't care, don't know

2) Larger blocks (running XT or bitcoinunlimited - bring it on!)

 

[lunar]

@Zangelbert Bingledack great comparison.

The issue is not automatism versus conscious action; it is autonomous action of each individual versus the exclusive action of the government. It is freedom versus government omnipotence.

This existential difference is becoming painfully apparent. Now we even have a handy list, thanks to BlockCore. It's worth considering this is now effectively an actionable register of doubters with which we can individually engage (verbose comment added). These few are the ones holding the rest of the industry hostage, thus attempting tailor made diplomacy might be productive?
Perhaps a quick classification guide might be useful? The way I see it there are 3 overlapping classes.

The Breakers- These are intellectually dishonest people intent of hobbling the network. either to promote personal off chain altcoin solutions or worse big industry saboteurs.- This group is a right off and should just be ignored.

The Bears- These are the pessimists they envisage a myriad of different ways bitcoin can break, obscure attack vectors, often many years in the future. - The general negativity is a vital reality check against rushed decision making. Often the CS pet project people, lazer focused with a tendency not to see the wood for trees. Debate is productive but this group should not be in gatekeeping positions.

The Blockers. Intellectually honest, yet too ego driven, or just fail to notice the doomed hypocrisy of attempting to centrally plan an open source system that was designed specifically to prevent such a thing. This group could use a better understanding of the importance of combining free markets and the game theory incentives that @Mengerian points out above. Not every variable should be nailed down with code. Out of the beautiful chaos emerges a previously unimaginable creation more wondrous than any one individual could possibly have foreseen.

 

[theZerg]

This is coming along very nicely!

Two quick questions:

#1. Is the data in Figure 5 real or simulated? It illustrates the effect very well!

#2. Do you want feedback on this and for us to point out typos? Is there a thread setup anywhere to make comments? (I've been keeping notes as I go through it.)

Data is real of course!! ()

WRT the prior graph, I think the the Bitcoin Relay Network has basically made orphans very rare. I think we need to work to get the P2P network performing on par with the relay net.

Absolutely give feed back!! I have only a few more days until Ledger submission deadline...


[doublepost=1450931835][/doublepost]

@theZerg Reading this reddit post, I had one question


What prevents an attacker from setting loose a large number of nodes and controlling the block size in that fashion? You described that an excessive number of nodes could result in a similar topology as when the network is partitioned by a single node that has a significantly lower block size limit policy.

In theory, nothing. Although certain carefully thought out practices in how the client code connects with other clients make it very hard. But such an attacker could do much worse things to today's Bitcoin network -- he could isolate portions of it to create artificial forks and double spend coins on each fork.

In practice, the Bitcoin Relay network stops this. And also people will think that its suspicious that a lot of hashing power has disappeared, communicate with other people and then manually configure their nodes to route around the attack.

But really this is not BU specific.
[doublepost=1450932298,1450931439][/doublepost]In regards to the theme "the market will find a way" check out my analysis of the GHOST protocol in the "Effect on Network Security" section of http://www.bitcoinunlimited.info/1txn. Basically several researchers previously identified a security issue which is basically "as block sizes increase, the effective hash rate descends because of validation time, which lets an attacker produce a bigger fraction of the blocks which can be used in these negative ways...".

But the drive to KEEP the ASICs hashing introduced the 1-txn blocks which solves this security problem...

 

[Peter R]

@theZerg: OK just making sure because this is different data than the large data set you used for the analysis. Could you keep logging the Relay Network data so that we can get a really nice and smoothed-out version of Fig. 5? One look at that chart already instantly communicates that the empty blocks typically occur quickly after a solved block, but if it was less ragged it would be even better.

"Absolutely give feed back!!"

Ok, I'll set up a thread here and post my review. By the way, I think it's already at the level that it would be admitted into peer review.
[doublepost=1450933976][/doublepost]Decent piece from Emin Gün Sirer:

http://hackingdistributed.com/2015/12/23/bitcoin-fee-market/

 

[Zangelbert Bingledack]

@lunarboy

Very much agree, as long as by "hit list" you just mean people to engage in one-on-one discussion with to see if they can be shown the light

 

[theZerg]

...given all the attention on reddit it SURE would be nice if ppl came over here and saw the BUIP process working... hint hint RBF, subchains. Will take about 20 min to write one up! Just ref existing doc sections that describe impl.

 

[Zangelbert Bingledack]

J. Stolfi presents an interesting view on nodes in this thread, saying that non-mining nodes are harmful and should not be encouraged. Of course he's saying that from the point of view that Bitcoin is broken, but he makes provocative points. For example:

 

[Zangelbert Bingledack]

@Peter R

I went ahead and started a thread for proofing @theZerg's very promising paper:

https://bitco.in/forum/threads/proofread-of-thezergs-paper-an-examination-of-single-transaction-blocks.656/

It'd probably be cleaner to have a separate thread for technical commentary.

 

[lunar]

@lunarboy

Very much agree, as long as by "hit list" you just mean people to engage in one-on-one discussion with to see if they can be shown the light

Blimey hadn't even crossed my mind anyone would interpret that any way other than the metephorical manner it was intended. Off to go edit

 

[Zangelbert Bingledack]

@lunarboy

Unfortunately there's no end to people taking things out of context in this debate, and I expect the heat to ramp up extremely if BU takes off.

 

[cypherdoc]

Data is real of course!! ()

WRT the prior graph, I think the the Bitcoin Relay Network has basically made orphans very rare. I think we need to work to get the P2P network performing on par with the relay net.

Absolutely give feed back!! I have only a few more days until Ledger submission deadline...


[doublepost=1450931835][/doublepost]

In theory, nothing. Although certain carefully thought out practices in how the client code connects with other clients make it very hard. But such an attacker could do much worse things to today's Bitcoin network -- he could isolate portions of it to create artificial forks and double spend coins on each fork.

In practice, the Bitcoin Relay network stops this. And also people will think that its suspicious that a lot of hashing power has disappeared, communicate with other people and then manually configure their nodes to route around the attack.

But really this is not BU specific.
[doublepost=1450932298,1450931439][/doublepost]In regards to the theme "the market will find a way" check out my analysis of the GHOST protocol in the "Effect on Network Security" section of http://www.bitcoinunlimited.info/1txn. Basically several researchers previously identified a security issue which is basically "as block sizes increase, the effective hash rate descends because of validation time, which lets an attacker produce a bigger fraction of the blocks which can be used in these negative ways...".

But the drive to KEEP the ASICs hashing introduced the 1-txn blocks which solves this security problem...

What's the reason some miners mine 1 tx vs zero?

 

[luigi1111]

@cypherdoc I think he means 0 (1 = coinbase).

 

[Mengerian]

Sounds like Mike Hearn has lost his enthusiasm for developing BitcoinXT. That's unfortunate. Quote from the BitcoinXT mailing list:

I do not plan on adding any further features to XT

https://groups.google.com/d/msg/bitcoin-xt/_SODoybQ9uA/_OYJVQmnCwAJ

 

[theZerg]

@theZerg: OK just making sure because this is different data than the large data set you used for the analysis. Could you keep logging the Relay Network data so that we can get a really nice and smoothed-out version of Fig. 5? One look at that chart already instantly communicates that the empty blocks typically occur quickly after a solved block, but if it was less ragged it would be even better.

"Absolutely give feed back!!"

Ok, I'll set up a thread here and post my review. By the way, I think it's already at the level that it would be admitted into peer review.
[doublepost=1450933976][/doublepost]Decent piece from Emin Gün Sirer:

http://hackingdistributed.com/2015/12/23/bitcoin-fee-market/

Yes, data is still being logged. I also want to get it to report the size of the mempool ASAP.

Yes, there's always 1 transaction in there (the coinbase). Maybe I should have called it 0 txn but I was going for precision. Some miners do add txns that they know can't be in the prior blocks (private). I was wrestling with the idea that in a severely supply limited environment a market may develop for these private transaction. However, I now think that miners will just compare txns against the pre-validated txns in the prior block so the days of forced 1-txn blocks may be numbered (they can still be used to slow down the average network validation rate, and used before the entire block is downloaded), but the sense of them (mining before validation) will still exist.

 

[Peter R]

"Yes, there's always 1 transaction in there (the coinbase). Maybe I should have called it 0 txn but I was going for precision."

Personally, I would call them "empty blocks" and then define that term to mean blocks that contain only the (special) coinbase transaction. Otherwise, I think many people will have the same initial reaction as @cypherdoc had.

 

[theZerg]

FYI: Peter R read and gave me early feedback on the paper. His help has been invaluable!

EDIT: WRT 1-txn yes but painful to change at this point... and not quite accurate. Maybe I should explain the term earlier.

 

[Zangelbert Bingledack]

@theZerg

Might be more painful to deal with the confusion. The less dust Core can kick up in the ensuing debate, the better.

 

[Aquent]

Sounds like Mike Hearn has lost his enthusiasm for developing BitcoinXT. That's unfortunate. Quote from the BitcoinXT mailing list:

https://groups.google.com/d/msg/bitcoin-xt/_SODoybQ9uA/_OYJVQmnCwAJ

Mike Hearn has unfortunately been a victim of a long campaign of character assassination. The blame is not, of course, solely external to himself, but some individuals seem to have used a snake like strategy of suffocating their victim with slow but constant and relentless squeezing. So I don't blame him for wanting to get out.

Unfortunately for them, in my view, their time is up because they have now shown their true colours.

As such, we need no longer engage them in a combative fashion or give bait to their trolling like Hearn seems to have done. We need no longer attack them because that only gives them legitimacy.

Where we can co-operate we will co-operate, where we can't we must part ways.

We need to focus on the silent majority, not the unpersuadable.