Bitcoin Forum
December 08, 2016, 02:30:10 PM *
News: To be able to use the next phase of the beta forum software, please ensure that your email address is correct/functional.
 
   Home   Help Search Donate Login Register  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 [22]  All
  Print  
Author Topic: Ultimate blockchain compression w/ trust-free lite nodes  (Read 68608 times)
ittayEyal
Newbie
*
Offline Offline

Activity: 4


View Profile
June 24, 2014, 03:01:28 PM
 #421

I'm not sure who misunderstands. I'll try to rephrase.

Here's the core of my question - does the system (all nodes) forget a prefix of the chain at some point?

If a node reads the entire chain (from genesis), it can prune it locally, sure. But how does a new node bootstrap without the entire chain? It needs to trust a snapshot (rolling root, utxo block, whatever it's called). That's my issue.
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
sugarpuff
Jr. Member
*
Offline Offline

Activity: 56


View Profile WWW
June 24, 2014, 03:13:34 PM
 #422

I'm not sure who misunderstands. I'll try to rephrase.

Here's the core of my question - does the system (all nodes) forget a prefix of the chain at some point?

If a node reads the entire chain (from genesis), it can prune it locally, sure. But how does a new node bootstrap without the entire chain? It needs to trust a snapshot (rolling root, utxo block, whatever it's called). That's my issue.

Rolling root and UTXO are very different architecturally, so I'll just answer for UTXO here.

Nobody, whether it's new nodes or old nodes, needs to know the entire histories of coins. The only thing that matters is whether they have the accurate unspent transactions.

New nodes download the entire UTXO meta chain (step #2 in the summary I posted earlier). This chain is protected by PoW. That's it. By knowing the accurate UTXO tree fingerprint, they can safely build the UTXO tree.

ittayEyal
Newbie
*
Offline Offline

Activity: 4


View Profile
June 25, 2014, 12:58:07 AM
 #423

New nodes download the entire UTXO meta chain (step #2 in the summary I posted earlier). This chain is protected by PoW. That's it. By knowing the accurate UTXO tree fingerprint, they can safely build the UTXO tree.

Ok, I'm convinced. As long as the UTXO meta chain is verified by the system, that's fine. I thought initially that it's a separately maintained data structure. If it's integrated in the Blockchain, in the sense that an invalid utxo would cause the entire Blockchain block to be rejected, then it's fine. In this case you're not adding any principles to rely on, and security is essentially intact as far as I see.

instagibbs
Member
**
Offline Offline

Activity: 114


View Profile
June 25, 2014, 01:02:32 AM
 #424

Ok, I'm convinced. As long as the UTXO meta chain is verified by the system, that's fine. I thought initially that it's a separately maintained data structure. If it's integrated in the Blockchain, in the sense that an invalid utxo would cause the entire Blockchain block to be rejected, then it's fine. In this case you're not adding any principles to rely on, and security is essentially intact as far as I see.



I don't think validity of commitment touches validity of Bitcoin block. Bitcoin nodes are allowed to completely ignore this merge-mined chain.
ittayEyal
Newbie
*
Offline Offline

Activity: 4


View Profile
June 25, 2014, 01:33:40 AM
 #425

I don't think validity of commitment touches validity of Bitcoin block. Bitcoin nodes are allowed to completely ignore this merge-mined chain.

Then these Bitcoin nodes (aka Bitcoin) should not truncate the chain. When you truncate up to some point you trust that you have a valid snapshot up to this time. There is no way to verify the UTXO chain (in the sense that there can be missing transactions) once the prefix is gone. So if someone manages to slip an invalid utxo into the blockchain, and this error is discovered after the prefix is gone, it invalidates every node that forgets prefixes.

So it could work as a fast-bootstrapping probably-reliable mechanism for your home PC. Something between SPV and full wallet. But to be done in full nodes the snapshot mechanism has to be incorporated in the blockchain.
jtimon
Legendary
*
Offline Offline

Activity: 1246


View Profile WWW
July 05, 2014, 09:39:25 AM
 #426

I'm not sure who misunderstands. I'll try to rephrase.

Here's the core of my question - does the system (all nodes) forget a prefix of the chain at some point?

If a node reads the entire chain (from genesis), it can prune it locally, sure. But how does a new node bootstrap without the entire chain? It needs to trust a snapshot (rolling root, utxo block, whatever it's called). That's my issue.

Rolling root and UTXO are very different architecturally, so I'll just answer for UTXO here.

Nobody, whether it's new nodes or old nodes, needs to know the entire histories of coins. The only thing that matters is whether they have the accurate unspent transactions.

New nodes download the entire UTXO meta chain (step #2 in the summary I posted earlier). This chain is protected by PoW. That's it. By knowing the accurate UTXO tree fingerprint, they can safely build the UTXO tree.

No offense, but I'm not sure you understand "UTXO" well enough to explain it to others. For starters, UTXO is nothing new, this post is about "committed utxo".
As maaku explained, miners need to know that they're mining on top of the valid chain, that's the only way they can know they have the accurate UTXO.

I don't think validity of commitment touches validity of Bitcoin block. Bitcoin nodes are allowed to completely ignore this merge-mined chain.

There's two ways to implement this: as a merged mined chain or as a softfork. The soft fork would invalidate blocks with an invalid utxo root, with the merged mined chain their validity is independent. I prefer the second option, the mini-chain just seems like a mechanism to deploy it, but until you make the soft fork this is not that useful.

Then these Bitcoin nodes (aka Bitcoin) should not truncate the chain. When you truncate up to some point you trust that you have a valid snapshot up to this time. There is no way to verify the UTXO chain (in the sense that there can be missing transactions) once the prefix is gone. So if someone manages to slip an invalid utxo into the blockchain, and this error is discovered after the prefix is gone, it invalidates every node that forgets prefixes.

So it could work as a fast-bootstrapping probably-reliable mechanism for your home PC. Something between SPV and full wallet. But to be done in full nodes the snapshot mechanism has to be incorporated in the blockchain.

Not sure what you mean by "prefix" here, it seems you mean the past chain.
With committed utxo SPV nodes can be much more secure, but full nodes still need to fully validate the chain (otherwise they would have SPV security, not full node security).

2 different forms of free-money: Freicoin (free of basic interest because it's perishable), Mutual credit (no interest because it's abundant)
sugarpuff
Jr. Member
*
Offline Offline

Activity: 56


View Profile WWW
July 05, 2014, 05:31:30 PM
 #427

No offense, but I'm not sure you understand "UTXO" well enough to explain it to others. For starters, UTXO is nothing new, this post is about "committed utxo". As maaku explained, miners need to know that they're mining on top of the valid chain, that's the only way they can know they have the accurate UTXO.

maaku confirmed that I understood UTXO's SPV+ mode.

jtimon
Legendary
*
Offline Offline

Activity: 1246


View Profile WWW
July 06, 2014, 11:56:50 AM
 #428

maaku confirmed that I understood UTXO's SPV+ mode.

Maybe you understand it, I don't know, but some of the terms you were using are unusual and I thought some people could get confused, that's all.

2 different forms of free-money: Freicoin (free of basic interest because it's perishable), Mutual credit (no interest because it's abundant)
maaku
Legendary
*
expert
Offline Offline

Activity: 905


View Profile
July 06, 2014, 12:01:17 PM
 #429

No offense, but I'm not sure you understand "UTXO" well enough to explain it to others. For starters, UTXO is nothing new, this post is about "committed utxo". As maaku explained, miners need to know that they're mining on top of the valid chain, that's the only way they can know they have the accurate UTXO.

maaku confirmed that I understood UTXO's SPV+ mode.

Jorge is correct: miners have to know they are building on top of a valid chain, and to do that they need to validate the entire block chain history. The post you linked to endorsed a statement which included this caveat, which your most recent explanation does not.

I'm an independent developer working on bitcoin-core, making my living off community donations.
If you like my work, please consider donating yourself: 13snZ4ZyCzaL7358SmgvHGC9AxskqumNxP
paulkernfeld
Newbie
*
Offline Offline

Activity: 10

Nanopayments Enthusiast


View Profile WWW
January 09, 2016, 02:09:03 PM
 #430

Is there a plan to deal with parties that want to store arbitrary data in the blockchain? Right now, OP_RETURN is used as a way to prevent non-financial data from bloating UTXO trees. However, if someone wants to store arbitrary data in the blockchain, ultimate blockchain c*mpression might encourage them to store this data by using fake addresses, because that way they would get much faster lookups and they could basically use this to build an efficient key-value store. Is this just accepted as something that will inevitably happen?
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 [22]  All
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!