Comments on: Ultraprune In Plain English

By: Dave Scotese

Dave Scotese — Fri, 30 Nov 2012 22:43:18 +0000

I think it's a double-edged sword, not by happenstance, but inherently. Any tendency in humans to abdicate the responsibility of protecting oneself from people-without-integrity (may I call them "politicians"?) will eventually lead to a win for such exploiters. These wins pile up and lead to, among other things, the impugnment of the system through which the exploitation is done. Think banks and our perception of them, because in days of yore, dishonest goldsmiths produced fake receipts, and eventually governments said "Oh, that's ok!" Banking is still a good idea, but not with the built-in guarantee to fail that fractional reserve introduces.

Hobbes (Thomas) suggested that we travel with weapons because we know everyone is out to get us. Of course, it's not everyone, but a few (and fewer still as our ability to communicate increases, eg with cellphones). I think that encouraging people to trust those they've deemed trustworthy is great, but encouraging them to find someone to trust simply because that makes life easier is horrible because it leads to leviathan.

By: David Perry

David Perry — Sat, 27 Oct 2012 17:56:30 +0000

Thanks, I do strive for accuracy but sometimes my own wild ideas sneak into text about what’s actually happening ;)

By: Pieter Wuille

Pieter Wuille — Sat, 27 Oct 2012 16:18:32 +0000

Yes, I was referring to the prioritization. Looks much more accurate now.

By: David Perry

David Perry — Sat, 27 Oct 2012 15:23:37 +0000

"The idea behind ultraprune is to use an ultra-pruned copy (only unspent transaction outputs in a custom compact format) of the block chain for validation (as opposed to a transaction index into the block chain). It still keeps all blocks around for serving them to other nodes, for rescanning, and for reorganisations. As such, it is still a full node. So, despite the name, it does not implement any actual pruning yet, though pruning would be trivial to implement now. This would have profound effects on the network though, so may still need some discussion first." Source: You. https://bitcointalk.org/index.php?topic=119525.0 Edit: Ah wait I see what you're talking about, it's the prioritization comment. I can fix that. Edit: Fixed.

By: Pieter Wuille

Pieter Wuille — Sat, 27 Oct 2012 13:46:00 +0000

Thanks, nice article! A few things I find confusing though: the logo at the top of the pos represents a merkle tree. A merkle tree is necessary for the transaction pruning mechanism Satoshi described in his paper, and is essential for SPV nodes, but it has nothing to do with ultraprune really. Also the #2 idea presented, while being correct, is completely independent from ultraprune. All it does is replace the block chain index by a pruned copy of it. A full node still requires processing the entire history (though doing so will happen much faster).

By: Paul

Paul — Sat, 27 Oct 2012 07:53:31 +0000

I guess this relates to your more recent "stats" posting too but surely a wider adoption that would heavily tilt the balance of the user-base into the "technophobe" camp would also mean that a centralised storage option is preferable to most participants?

In general, when it comes protecting things of value on their computers, the majority of people are somewhat careless. Whether it's just laziness and/or just a lack of trust in their own technical ability, cloud storage makes more sense for most. It's quite possible (if not probable) that a more general Bitcoin audience would prefer to trust a third party to maintain their Wallet for them.

Do you imagine a future compromise whereby hosted wallets can also better contribute to the integrity of the whole network?

By: David Perry

David Perry — Fri, 26 Oct 2012 18:01:21 +0000

I can't see any reason why you wouldn't be able to, with some modifications to the client of course. The consequences of you, an individual doing this, are fairly trivial – you just won't have copies of those blocks to send to anyone else requesting them. The consequences of everyone or even a large minority doing this is that those blocks become significantly less available to the entire network, increasing the chance of data loss, increasing the time taken to download the whole block chain and potentially creating gaps in the data that could be quite problematic.

Basically, your client is free to keep or discard whatever data you like, but discarding data reduces your client's usefulness to the network and making a widespread practice of discarding data decreases the usefulness of each network node on average.

By: Hugo

Hugo — Fri, 26 Oct 2012 07:38:38 +0000

Would it be possible to actually delete useless nodes in the "complete" database? I mean, the nodes that are important are only the nodes that hold a balance and nodes that are in the middle of nodes that hold a balance. What are the dangers of deleting some transactions that have happened but are now completely irrelevant?

By: cypherdoc

cypherdoc — Fri, 26 Oct 2012 01:39:15 +0000

great explanation. you've found your niche!

By: Ultraprune In Plain English | Bitcoin News Bits - CoinBits.com

Ultraprune In Plain English | Bitcoin News Bits - CoinBits.com — Thu, 25 Oct 2012 22:12:31 +0000

[...] by enmaku link [...]