Thanks to people who support 1-2 MB blocks - great idea u fools...

[adamstgBit]

Allow Nick Szabo to explain it to you:

Compared to existing financial IT, Satoshi made radical trade-offs in favor of security and against performance. The seemingly wasteful process of mining is the most obvious of these trade-offs, but it also makes others. Among them is that it requires high redundancy in its messaging. Mathematically provable integrity would require full broadcast between all nodes. Bitcoin can’t achieve that, but to even get anywhere close to a good approximation of it requires a very high level of redundancy. So a 1MB block takes vastly more resources than a 1MB web page, for example, because it has to be transmitted, processed and stored with such high redundancy for Bitcoin to achieve its automated integrity.”

I will say it again, that Bitcoin is the only network of its kind that has saturated the level of home-based internet infrastructure to the point that it can now move onto the next stage of its evolution (8MB), while the niche it used to cover can be served by its closest PoW competitor, which has exact same properties except for smaller market cap, while being at the time more convenient to operate at home.

Satoshi has always intended to increase the limit and even provided a small code-snippet demonstrating how to do it.

How about no? You can go ahead and create EvolutionCoin but you can't and won't have Bitcoin.

Haha!

Nothing in the Universe is static. Bitcoin has never been effectively capped either.
Keep it at 1MB forever and it will wither and die or mutate into sidechains that would then tear it apart anyway.

Remember, the true path of Bitcoin isn't ambiguous.
Sometimes things need to change in order to stay the same.

A protocol like Bitcoin will eventually become static and sort of ossify once it is considered good enough. The evolution will come from additional layers.

Bitcoin is the cambrian explosion but its blockchain alone will never support the whole financial system.

i agree that it will be static at some point but not at this stage and not with artificial 1 MB blocks.

i disagree it will never be static, at best in a few years the protocol, and "consensus layer" will be static for years at a time, but the higher level code will always be improving. and it's hard to imagine the protocol NEVER changing in the slightest.

in anycase we are clearly nowhere near this static state, i've said it before and i'll say it again, if bitcoin doesn't improve your opening up the door wide open to competitors having serious advantages, and it won't take all that much for the user base to switch.

[1714616231]

1714616231

[1714616231]

1714616231

[1714616231]

1714616231

[1714616231]

1714616231

[adamstgBit]

Allow Nick Szabo to explain it to you:

Compared to existing financial IT, Satoshi made radical trade-offs in favor of security and against performance. The seemingly wasteful process of mining is the most obvious of these trade-offs, but it also makes others. Among them is that it requires high redundancy in its messaging. Mathematically provable integrity would require full broadcast between all nodes. Bitcoin can’t achieve that, but to even get anywhere close to a good approximation of it requires a very high level of redundancy. So a 1MB block takes vastly more resources than a 1MB web page, for example, because it has to be transmitted, processed and stored with such high redundancy for Bitcoin to achieve its automated integrity.”

Nick Szabo is not saying it's usefull for 1 node to get the same TX data 6 times, he's saying its useful the TX data is redundantly stored on all nodes.

[coalitionfor8mb]

Just found another way to look at it.
Think of electron orbits in the atom, they are discreet!

The 1MB orbit has definitely proven itself stable,
but our atom (of Bitcoin) is now being repeatedly overwhelmed with incoming energy (stress-tests) and needs to find the next stable orbit for its electrons.

I argue, that anything less than 8MB will simply be smashed through (with inertia) and the electrons may just fly away . The same holds true in case we overshoot.

Now, if we target the next orbit just right, we will have a good chance of increasing the overall energy of the atom (Bitcoin's market cap), while keeping the electrons (user-base) in place, though on a higher orbits.

However, in order for that new orbit to become stable, everyone and their dog (and even cat) should know that Bitcoin is now capped at 8MB.
No one (I repeat, no one!) forks with the limit until it's time, of course!

[adamstgBit]

it would be interesting to know where Fidelity would take us if they were to "flip the on switch" are they going to be spitting out 5TPS or 50TPS....

[tl121]

Nick Szabo is not saying it's usefull for 1 node to get the same TX data 6 times, he's saying its useful the TX data is redundantly stored on all nodes.

It's relatively simple to get the same blocks at all nodes without any node having to receive anything except block headers more than once.  The block header succinctly identifies the block and verifies its content. The bandwidth required to receive block headers is sufficiently low as to be negligible. The block headers can be flooded, yielding censor proof knowledge of a new block, but the transaction data can be held until requested. This will cost an extra round-trip to allow for the block data to be selectively requested on only one connection, but this time will be negligible.  Nodes will need a policy to rate their connections as to reliability and performance so that they get block data in a timely fashion.

It's more complex to efficiently forward transactions in the mempool, because there is no similar fixed succinct representation of multiple transactions.  Some kind of dynamic spanning tree, as in the graphs earlier might work if the goal were just efficiency, but the goal also is to prevent censorship, which makes the problem more difficult.

[RoadTrain]

Allow Nick Szabo to explain it to you:

Compared to existing financial IT, Satoshi made radical trade-offs in favor of security and against performance. The seemingly wasteful process of mining is the most obvious of these trade-offs, but it also makes others. Among them is that it requires high redundancy in its messaging. Mathematically provable integrity would require full broadcast between all nodes. Bitcoin can’t achieve that, but to even get anywhere close to a good approximation of it requires a very high level of redundancy. So a 1MB block takes vastly more resources than a 1MB web page, for example, because it has to be transmitted, processed and stored with such high redundancy for Bitcoin to achieve its automated integrity.”

Nick Szabo is not saying it's usefull for 1 node to get the same TX data 6 times, he's saying its useful the TX data is redundantly stored on all nodes.

What exactly are you referring to as "get the same TX data 6 times"? I keep reading that, but not following.

[brg444]

Allow Nick Szabo to explain it to you:

Compared to existing financial IT, Satoshi made radical trade-offs in favor of security and against performance. The seemingly wasteful process of mining is the most obvious of these trade-offs, but it also makes others. Among them is that it requires high redundancy in its messaging. Mathematically provable integrity would require full broadcast between all nodes. Bitcoin can’t achieve that, but to even get anywhere close to a good approximation of it requires a very high level of redundancy. So a 1MB block takes vastly more resources than a 1MB web page, for example, because it has to be transmitted, processed and stored with such high redundancy for Bitcoin to achieve its automated integrity.”

Nick Szabo is not saying it's usefull for 1 node to get the same TX data 6 times, he's saying its useful the TX data is redundantly stored on all nodes.

What exactly are you referring to as "get the same TX data 6 times"? I keep reading that, but not following.

Don't mind him he really doesn't know what he's talking about...

[brg444]

point is a home node having 60+ peers is nutty

Well... my goal was to share information with whomever asked for it without abusing the offer. Apparently that's crazy!

it is!

you're probably forwarding TX's to all your peers most of whom probably already received those TX datas from the other 60+peers Their connected to.

and that sir is called resilience by redundancy

The only thing that needs redundancy is the blockchain. Broadcasting the same transaction twice is called inefficiencies.

Welp! Should've told Satoshi heh  

Seriously who do you think you noobs are giving engineering lessons to people when you are most clueless about how the system operates.  

Then go on and explain how broadcasting the same transaction twice increase resilience instead of making pointless post. 

Well first you'll have to explain where do you get the idea that transactions are broadcasted twice?

Sorry I should add this precision: Broadcasting a transaction to a node that has already received this transaction.

In other word, inefficiencies.

Are you aware of exactly how this works? The transaction data is not uploaded to the peers unless they request for it, in which case if they do then it signifies they have not "already received this transaction".

[adamstgBit]

Allow Nick Szabo to explain it to you:

Compared to existing financial IT, Satoshi made radical trade-offs in favor of security and against performance. The seemingly wasteful process of mining is the most obvious of these trade-offs, but it also makes others. Among them is that it requires high redundancy in its messaging. Mathematically provable integrity would require full broadcast between all nodes. Bitcoin can’t achieve that, but to even get anywhere close to a good approximation of it requires a very high level of redundancy. So a 1MB block takes vastly more resources than a 1MB web page, for example, because it has to be transmitted, processed and stored with such high redundancy for Bitcoin to achieve its automated integrity.”

Nick Szabo is not saying it's usefull for 1 node to get the same TX data 6 times, he's saying its useful the TX data is redundantly stored on all nodes.

What exactly are you referring to as "get the same TX data 6 times"? I keep reading that, but not following.

i blieve the bitcoin networks looks somthing like this


and i'm saying the network's topology is inefficient and as a result when nodes relay TXs a node can get sent the same TX multiple times.

ex. 6 nodes get 1 tx to propagate, all 6 nodes send the tx to all 6 nodes, each node gets the same tx 6 times.


Holliday says he has 60+ peers connected. thats alot of peers for one node

i think at most a node should have 5 peers

*maybe when a node is trying to download the blockchain it make sense to dwl from as many peers are possible. but when propagating a TX at most you should send that tx to 5 peers, who also send it to 5 peers, who also send it to 5 peers, etc... five levels in 3125 msgs have occurred.

if all nodes have 60 peers 3 levels in and 216000 msgs have been sent. waste full? i think so.

bitcoin topology should allow a msg to propagate in an exponential way ( 3 peers forward a TX to 3 peers who forward a TX  to 3 peers...) while trying to avoid a peer receiving the same TX from 60 different peers.

[RoadTrain]

Sorry I should add this precision: Broadcasting a transaction to a node that has already received this transaction.

In other word, inefficiencies.

I'm not sure what you are talking about. Do you mean that if a node sees a new transaction or block, it immediately relays it to other peers?
As far as I'm aware that's not quite true. The only thing a node does when receiving new tx/block is send an inv message containing its hash. After that, a peer can request a tx/block using a getdata message if it doesn't already have it.

A node receives tx only when it asks for it.

[brg444]

and i'm saying the network's topology is inefficient and as a result when nodes relay TXs a node can get sent the same TX multiple times.

ex. 6 nodes get 1 tx to propagate, all 6 nodes send the tx to all 6 nodes, each node gets the same tx 6 times.

That's not how it works 

[adamstgBit]

and i'm saying the network's topology is inefficient and as a result when nodes relay TXs a node can get sent the same TX multiple times.

ex. 6 nodes get 1 tx to propagate, all 6 nodes send the tx to all 6 nodes, each node gets the same tx 6 times.

That's not how it works  

says you. which isn't reassuring.

i'm under the impression the topology is laid out in a pretty random manner and peers blindly forward the TX to all the peers they are connected to.

when Holliday gets a TX to forward, he's forwarding that TX to all 60 peers, which is why he's having bandwidth issues.

[brg444]

and i'm saying the network's topology is inefficient and as a result when nodes relay TXs a node can get sent the same TX multiple times.

ex. 6 nodes get 1 tx to propagate, all 6 nodes send the tx to all 6 nodes, each node gets the same tx 6 times.

That's not how it works  

says you. which isn't reassuring.

i'm under the impression the topology is laid out in a pretty random manner and peers blindly forward the TX to all the peers they are connected to.

when Holliday gets a TX to forward, he's forwarding that TX to all 60 peers, which is why he's having bandwidth issues.

No. Peers don't "blindly forward the TX to all the peers they are connected to".

That's not how it works  

[adamstgBit]

Sorry I should add this precision: Broadcasting a transaction to a node that has already received this transaction.

In other word, inefficiencies.

I'm not sure what you are talking about. Do you mean that if a node sees a new transaction or block, it immediately relays it to other peers?
As far as I'm aware that's not quite true. The only thing a node does when receiving new tx/block is send an inv message containing its hash. After that, a peer can request a tx/block using a getdata message if it doesn't already have it.

A node receives tx only when it asks for it.

yes node magically know when other nodes transact

[adamstgBit]

and i'm saying the network's topology is inefficient and as a result when nodes relay TXs a node can get sent the same TX multiple times.

ex. 6 nodes get 1 tx to propagate, all 6 nodes send the tx to all 6 nodes, each node gets the same tx 6 times.

That's not how it works  

says you. which isn't reassuring.

i'm under the impression the topology is laid out in a pretty random manner and peers blindly forward the TX to all the peers they are connected to.

when Holliday gets a TX to forward, he's forwarding that TX to all 60 peers, which is why he's having bandwidth issues.

No. Peers don't "blindly forward the TX to all the peers they are connected to".

That's not how it works  

repeating that doesn't make it more reassuring.

[brg444]

and i'm saying the network's topology is inefficient and as a result when nodes relay TXs a node can get sent the same TX multiple times.

ex. 6 nodes get 1 tx to propagate, all 6 nodes send the tx to all 6 nodes, each node gets the same tx 6 times.

That's not how it works  

says you. which isn't reassuring.

i'm under the impression the topology is laid out in a pretty random manner and peers blindly forward the TX to all the peers they are connected to.

when Holliday gets a TX to forward, he's forwarding that TX to all 60 peers, which is why he's having bandwidth issues.

No. Peers don't "blindly forward the TX to all the peers they are connected to".

That's not how it works  

repeating that doesn't make it more reassuring.

The purpose of my exercise is to get you to find the proper information so as to educate yourself.

Start here:
http://lmgtfy.com/?q=bitcoin+gossip+network

BTW I expect some kind of "sorry I was wrong" post next.

[adamstgBit]

and i'm saying the network's topology is inefficient and as a result when nodes relay TXs a node can get sent the same TX multiple times.

ex. 6 nodes get 1 tx to propagate, all 6 nodes send the tx to all 6 nodes, each node gets the same tx 6 times.

That's not how it works  

says you. which isn't reassuring.

i'm under the impression the topology is laid out in a pretty random manner and peers blindly forward the TX to all the peers they are connected to.

when Holliday gets a TX to forward, he's forwarding that TX to all 60 peers, which is why he's having bandwidth issues.

No. Peers don't "blindly forward the TX to all the peers they are connected to".

That's not how it works  

repeating that doesn't make it more reassuring.

The purpose of my exercise is to get you to find the proper information so as to educate yourself.

Start here:
http://lmgtfy.com/?q=bitcoin+gossip+network

BTW I expect some kind of "sorry I was wrong" post next.

http://gavintech.blogspot.ca/2015/09/how-efficient-is-bitcoins-gossip.html

nodes are semi-randomly connected to each other, and information (transactions, blocks, IP or onion addresses of peers, and, very rarely, alert messages) is flooded across the network.

https://en.wikipedia.org/wiki/Gossip_protocol

Due to the replication there is an implicit redundancy of the delivered information.

"sorry I was wrong"

[brg444]

and i'm saying the network's topology is inefficient and as a result when nodes relay TXs a node can get sent the same TX multiple times.

ex. 6 nodes get 1 tx to propagate, all 6 nodes send the tx to all 6 nodes, each node gets the same tx 6 times.

That's not how it works  

says you. which isn't reassuring.

i'm under the impression the topology is laid out in a pretty random manner and peers blindly forward the TX to all the peers they are connected to.

when Holliday gets a TX to forward, he's forwarding that TX to all 60 peers, which is why he's having bandwidth issues.

No. Peers don't "blindly forward the TX to all the peers they are connected to".

That's not how it works  

repeating that doesn't make it more reassuring.

The purpose of my exercise is to get you to find the proper information so as to educate yourself.

Start here:
http://lmgtfy.com/?q=bitcoin+gossip+network

BTW I expect some kind of "sorry I was wrong" post next.

nodes are semi-randomly connected to each other, and information (transactions, blocks, IP or onion addresses of peers, and, very rarely, alert messages) is flooded across the network.

  You noobs really need to get fed every answer heh.

Here:

A node learns about a new transaction, either from a peer or of it's own creation.

It announces to every one of it's peers (8+ typically) with an inventory message that it has learned about a transaction or block with a particular ID.

If the connected peer doesn't already know about the announced transaction, it sends a getdata request back asking for the contents of the transaction. If it already knows about the transaction, no getdata is made in response.

To put into context: if Holliday's node uploads content of a transaction to its peers it is because they DON'T know yet about the transaction. Clear enough?

[adamstgBit]

bottom line here is that if it was laid out properly Holliday wouldn't be uploading more than 2MB every 10mins

something is afoot, that much is clear.

[adamstgBit]

if Holliday's node uploads content of a transaction to its peers it is because they DON'T know yet about the transaction. Clear enough?

what isnt clear is why we are asking Holliday to use so much bandwidth, and why the task isn't evenly spread out