[IDEA] - Bitcoin-Powered Database

[CIYAM]

I am very much against the "web of trust" as it is only a continuation of elitism.

Am not quite sure why having public DB records of trust votes == elitism. The idea I had in mind was simply a system of publishing specific rating tx's but nothing to do with how they are interpreted (so different apps could take the raw data and draw very different conclusions depending on how they want to do that).

[1715350819]

1715350819

[1715350819]

1715350819

[1715350819]

1715350819

[ripper234]

Am not quite sure why having public DB records of trust votes == elitism. The idea I had in mind was simply a system of publishing specific rating tx's but nothing to do with how they are interpreted (so different apps could take the raw data and draw very different conclusions depending on how they want to do that).

+1

[cbeast]

I am very much against the "web of trust" as it is only a continuation of elitism.

Am not quite sure why having public DB records of trust votes == elitism. The idea I had in mind was simply a system of publishing specific rating tx's but nothing to do with how they are interpreted (so different apps could take the raw data and draw very different conclusions depending on how they want to do that).

How would your trust system be measured? Would you pin the trust system to USD?

[casascius]

They definitely DO need the whole block chain, otherwise they might include an invalid/double spent transaction and loose a block, because this block won't be accepted by anyone.

This "reliable" set of unspent transactions can only be produced by someone you trust (Bitcoin = 0 trust to others!) or yourself - and this person needs to have the whole block chain to be reliable in filtering out bad transactions.

They actually do not need the whole block chain. They merely need a reliable set of unspent tx outputs, and your challenge concerns the term "reliable", not the term "whole block chain".  what you are saying is that the only way to know you have the accurate set of outputs is to download a large amount of extra data so your computer can say "yes this is it", and you are saying something different than what I am saying.

If I could download a digest of the first 200000 blocks - a digest that was produced by a published open source utility and there was a wide public consensus that the digest file with hash X is what the utility produces no matter who runs it therefore is a faithful record of all unspent txouts of blocks 1-200000, with lots of verifiable GPG signatures from respected people attesting to it, I would have no problem trusting that versus spending days downloading a huge file just to know I got the right one. And I definitely would be able to mine with it.

[CIYAM]

How would your trust system be measured? Would you pin the trust system to USD?

I am only proposing trust tx's to be in a/the blockchain - the measurements of them would be entirely up to software apps that read the blockchain (and different apps could have very different ways of measuring it).

[Vitalik Buterin]

How would this work fundamentally better, than, say, Bitcoinica publishing a nightly encrypted backup (a dump) that anybody can download (either via http or torrent), and also publishing a PGP signature to commit to its hash on a particular date?

This would be a whole lot simpler and less complicated and trivial for stakeholders to participate in the solution.

If there were a need to demonstrate that prior revisions aren't changing and that only thing happening is new records being added, the backups could simply be an archive file, where the backup of any night n contains identical copies of all the files that were present in the backup of night n-1, plus a new file representing that day's activity.

Intermediate solution: publish nightlies and store just their hashes in the blockchain.

Seems like it solves the problems with both approaches to me.

[Xenland]

Aren't their SQL database tables that have a "write and read" option?

This is a good point your bringing up.... Having hashing power to verify SQL databases... Publicly?

[ripper234]

Intermediate solution: publish nightlies and store just their hashes in the blockchain.

Seems like it solves the problems with both approaches to me.

Almost. It still assumes all copies of the nightlies won't be destroyed by an attacker (meaning, that someone will actually download the nightlies for safe keeping). This is quite reasonable in practice.

[ripper234]

Aren't their SQL database tables that have a "write and read" option?

This is a good point your bringing up.... Having hashing power to verify SQL databases... Publicly?

I'm not sure I follow. Can you rephrase your question/s?

[CIYAM]

Publishing DB backups (minus sensitive information) is a step in the right direction but if we were to build a blockchain tx approach then it could provide a common protocol that could be used by different rating applications so that in the end users could pick and choose which specific ratings they care about (whilst also being certain that the data hasn't been altered after initial publication).

EDIT: Ooops I forgot which thread I was posting this in - the above is relevant to a rating system rather than a generalised DB.

[Xenland]

I geez I should have clicked the link in the OP before i wrote my reply down i was just in a hurry dont mind me, I'll just keep following closly anyways this is interesting subject

[DeathAndTaxes]

They definitely DO need the whole block chain, otherwise they might include an invalid/double spent transaction and loose a block, because this block won't be accepted by anyone.

This "reliable" set of unspent transactions can only be produced by someone you trust (Bitcoin = 0 trust to others!) or yourself - and this person needs to have the whole block chain to be reliable in filtering out bad transactions.

Not true.  Due to the Merkle tree structure spent tx can be pruned away and without any trust a miner can reconstruct the entire blockchain back to the geneis block.  The only missing parts are the spent tx and dead ends.

You do not need the full blockchain to be a miner.  You DO need a full set of all unpsent outputs and all block headers.  They aren't the same thing.

[Sukrim]

With Cascasius' approach you'd still need to audit the code + trust that the people signing the file etc. are trustworthy and really who they claim to be. Yes, it might be an acceptable risk, but having the block chain yourself and doing the calculations and eventual pruning yourself is still what you need to be ultimately sure. For that you need to have all 200k blocks (pre-pruned or not).

D&T: Yes, you don't need 100% of all transactions, "only" 100% of all currently unspent transactions (something that's not always easy to find out) + block headers. Eventually it might also be nice to have then still stored/hashed in merkle trees, so you can hand at least pruned blocks to the P2P network. It can be estimated that in the future, clearly spent transactions will be pruned as fast as possible, things like this database though might start to bloat these unspent TX. On the other hand, store and delete operations would then be possible, leading to a write being a delete + a new store.