Gold collapsing. Bitcoin UP.

[rocks]

Very well said

 

[Mengerian]

It should be noted that the segwit subsidy is not a consensus rule and thus miners are free to discard it.

Not sure what you mean here. The formula "a + b/4 <= 1MB" would become a new consensus rule after the SW soft fork. (where "a" is the size of the "old" blocks, and is the blocksize non-upgraded nodes would see. "b" is the size of the auxiliary "witness" blocks that old nodes don't see.)

The point is though that these calculations should be left to miners and not... well, whatever shenanigans are going on here.

agreed

 

[Richy_T]

We may have crossed wires here. I'm not referring to the rule you stated (which I am not aware exists) but that in terms of prioritization, segwit is prioritzed ahead of regular transactions that have four times the fee attached.

If this is my fault, my apologies.
[doublepost=1454561462][/doublepost]Oh, I see where you are coming from. I don't think that b/4 can be said to be the size of segwit transactions that are included in the block.

The consensus rule is

a+b1 <1mb
where a is regular transactions and b1 is the block part of segwith transactions.

call b1+b2 the total of segwit transactions, b1 the block part, and b2 the non-part block

I believe the status is that under "ideal" conditions, a tends to 0 and b2/b1 could be as high as 4 but is more likely to tend towards 1.6-1.7

However, a is not likely to tend to 0 so we will see much smaller than this.

Unless a not being 0 is already included in the 1.6-1.7 calculation?

 

[cypherdoc]

@Mengerian

the debate has always been about BW for miners and their desire not to get orphaned. data storage is not an issue. so in that sense, i think miners will be penalized by this drive by core dev to push multi sig tx's for LN despite the benefit of being able to prune their sigs once verified. the reason being, that the more multi sigs per block, the greater the move towards 4MB block sizes. that's bad b/c of higher orphaning.

 

[Richy_T]

I just went back and look at ajtowns post again and I think he does a terrible job of attempting to explain things. Very hard to understand (and I've studied quantum physics). Sorry, Cypher.

I'll go and hunt down the actual details of segwit and the discount (if I can find them. Who knows these days?) and see if I can clarify.
[doublepost=1454562515][/doublepost]I have to say to be fair to ajtowns, he may not be actually trying to sell this thing. But I'm fond of the quote attributed to Einstein, "If you can't explain it to a six year old, you don't understand it yourself". This is not rocket science and should be able to be able to be broken down into reasonably logical steps of justification (at least one of which I suspect to be deeply flawed).

 

[Mengerian]

We're defining "a" and "b" differently, so it might get a bit confusing

The consensus rule is

a+b1 <1mb
where a is regular transactions and b1 is the block part of segwith transactions.

This is the current consensus rule, and would be the consensus rule still enforced by non-upgraded nodes after a segwit soft fork.

call b1+b2 the total of segwit transactions, b1 the block part, and b2 the non-part block

After the segwit soft fork, the way segwit is coded (if I understand it correctly), upgraded nodes would enforce a new consensus rule, (a +b1) + (b2/4) <= 1MB

 

[cypherdoc]

@Richy_T

yeah, maybe. but according to Towns:

a=old tx's with sigs + SW tx's w/o sigs
b=SW sigs=witness data

so a+b/4<=1MB

 

[solex]

Don't forget that some of the SW sig is in base block data (cypher & mengerian "a", or richy "b1") so that the old nodes agree that the SW tx are signed.

edit: name switch..

 

[Richy_T]

Mengerian & Cypher, that sounds wrong but I'll check more carefully as there's a lot of misinformation going around.

My understanding was that segwit transactions can be split so that only part of the data goes into the block but that the block limit is still 1mb. a+b/4<=1mb would only be the case of (let's call b=b1+b2) a+(b1+b2)/4<=1mb which would only apply if b2=b1*3 and I don't believe there's a fixed ratio of b1 to b2 (different proportions of segwit data can be stored in the segwit extension depending on the structure of the transaction).

But like I say, I'll look into it and get back to you. I'm fairly sure I'm right but it wouldn't be the first time I've been wrong if I am.
[doublepost=1454568068][/doublepost]I've just thought of an analogy using airline passengers and luggage that I'll have to work on. Though I think the passengers will have to be robots so they can put their legs in the luggage.

 

[albin]

Wait so I had the impression that the non-witness portion always would be capped at 1MB regardless, the total size of the witness would be capped, and the 1/4 coefficient had to do with discounting witness data size in terms of fee/kb for miner tx selection logic? Is this not correct?

 

[go1111111]

Pretty drunk but felt I had to answer. How'd I do?

A good reply. I felt there were a few more things in the guy's post that should be challenged though.

 

[YarkoL]

Ok despite my small understanding I decided to address Ilogy's second post.

edit: note that it is a response to his reply

 

[cypherdoc]

@YarkoL

Excellent response.

That's it, imo, Bitcoin should be able to be both a payment network and a settlement network because it is money of, by, and for the Internet. And since one day we should have Internet everywhere in the world, why shouldn't Bitcoin be allowed to permeate every corner of the Earth if it can? Physical gold has done that so that sets the bar for what Bitcoin has to compete with; especially since it has none of the physical attributes that makes physical gold valuable. In fact that bar is higher since Bitcoin cannot generate that longing for beauty, weight, volume, & luster that would drive a thief to snatch it off your hands or neck. . We have an obligation to at least try to take it global for that would be the best definition of decentralized that I can think of. Bitcoin is just information in the end so theoretically it should be able to accomplish these goals. Whoever thought we'd be where we are with the Internet even a decade ago? Its this marriage of the two concepts that escape small blockists: the merger of payment network and settlement network. But it's also the marriage of the physical and the virtual that they miss. Learn to think big.

EDIT

 

[cypherdoc]

A good reply. I felt there were a few more things in the guy's post that should be challenged though.

Thanks for hitting that guy right between the eyes on several of his false assumptions.

 

[awemany]

The most amazing thing is that people believe that Adam, Greg and so forth won't drop Bitcoin like a hot potato as soon as it is profitable for them to do so.

Greg said they have time-locked coins to think long-term for Bitcoin. But AFAIR he never said how much that is. For all we know, he could have locked 1mBTC per employee.

And at some point, the returns of a company that succeeds in redefining and eventual hollowing out of Bitcoin will be higher than what is expected from a measly time-locked BTC stash...

It is my impression that this turning-point in motivation happened a while ago.

Along these lines, and regarding the rename: I was rather thinking about the parallels to a well-known company of mercenaries that changed names twice...

 

[cypherdoc]

Don't forget that some of the SW sig is in base block data (cypher & mengerian "a", or richy "b1") so that the old nodes agree that the SW tx are signed.

edit: name switch..

i don't think that's right (hey, i sure was wrong). remember that these new SW tx's are ANYONE_CAN_SPEND tx's that pwuille in his presentation said old nodes can't verify have been signed. they are merely data messages (the tx) w/o the sigs (which are stored in the new witness merkle block).
[doublepost=1454599070][/doublepost]ok. why do you think gold and Bitcoin are rallying?:

[doublepost=1454600089,1454599025][/doublepost]Bitcoin respecting the bottom of the triangle. Lotsa bulls have been shed the last several months b/c of Blockstream, guaranteed. The Ship is lighter:

 

[Richy_T]

Wait so I had the impression that the non-witness portion always would be capped at 1MB regardless, the total size of the witness would be capped, and the 1/4 coefficient had to do with discounting witness data size in terms of fee/kb for miner tx selection logic? Is this not correct?

That's my understanding.

 

[lunar]

Just a hypothetical. What do you suppose would happen? If say a month after the Classic release, it was looking like the 750/1000 rule would not pass, but say 60% of the hashpower just started mining on a block >1MB anyway.

How quickly do you think this conflict would resolve? Unlimited would follow longest chain and i'd imagine a patch for classic would be issued fairly rapidly. (or would classic/XT follow the longest chain despite not fulfilling the 750/1000 block rule?)

 

[cypherdoc]

that 60% would have to be using a certain implementation to accomplish a >1MB block that would be known well beforehand. it doesn't just happen.

 

[lunar]

I admit it's not likely and would be quite chaotic, but what is to stop the pools that make 60% just issuing a joint statement? Perhaps 'We firmly believe in satoshis vision of main chain scaling, as such we will be mining a >1MB block on 01Apr2016 please be ready.'

April fools day might not be the wisest ;-)