Gold collapsing. Bitcoin UP.

[Mengerian]

Here's a bit which shows some of Greg's manipulation tactics:
"Pruning doesn't remove transactions, not without a radical departure in the security model to "SPV (trust the miners) security" for history."

I am pretty sure I remember that he's the guy who mainly pushed for 'we need all transactions since the beginning for trustlessness' in various threads on BCT.

That there's still a trust root (genesis block, what are valid transactions) and that the 'all transactions since genesis' is just a particular approach, that he conventiently leaves out.

In my opinion, having validated UTXOs and getting history from miners for (lets say a year back) would be enough to prove that all is fine. Because else, the whole network would need to lie to you for a full year.

Why is it that such ideas are completely excluded from Blockstream Core's 'Overton window'?

1. Change protocol so that a hash of the UTXO is put into each block
2. As a new client, download data: UTXO set 1 year ago + 1 year txn data + all headers of course
3. Check that the downloaded data follows the rules and all UTXO hashes match
4. Profit - it is extremely unlikely that the whole network conspires to build a hashpower-longest chain for a full year that lies to you

Yeah, I totally agree with you. It seems like a natural evolution to not have to store every transaction since genesis, to be able to just go back as far as one wants anchored by trusted/validated UTXO sets.

It seems strange to call this a "radical departure in the security model". Having hundreds of millions of dollars worth of proof-of-work on a chain seems like a very high cost to forge history. And in any case, individual nodes could choose whether to prune old blocks, and how far back they want to validate, it's up to them to decide what security trade-off makes the most sense.

I find it fascinating how Bitcoin seems to have the capability to evolve into a real-time transaction flow processing network. The discussion in this thread about weak blocks helped me realize that in the future blocks can become almost a background abstraction that each node relies on as an internal concept to validate security, but are never (or rarely) explicitly handled over the network. The same would go, presumably, for the blockchain as a whole, and historical blocks.

 

[Zangelbert Bingledack]

If opt-in RBF is a bad change, it is unfortunate that this thread is still going basically unchallenged in any effective way:

https://www.reddit.com/r/Bitcoin/comments/3urm8o/optin_rbf_is_misunderstood_ask_questions_about_it/:

Btcdrak and others are loving it, linking to it all over the place because it looks like a massacre of all opposition to opt-in RBF.

 

[rocks]

Yeah, I totally agree with you. It seems like a natural evolution to not have to store every transaction since genesis, to be able to just go back as far as one wants anchored by trusted/validated UTXO sets.

It seems strange to call this a "radical departure in the security model". Having hundreds of millions of dollars worth of proof-of-work on a chain seems like a very high cost to forge history. And in any case, individual nodes could choose whether to prune old blocks, and how far back they want to validate, it's up to them to decide what security trade-off makes the most sense.

I find it fascinating how Bitcoin seems to have the capability to evolve into a real-time transaction flow processing network. The discussion in this thread about weak blocks helped me realize that in the future blocks can become almost a background abstraction that each node relies on as an internal concept to validate security, but are never (or rarely) explicitly handled over the network. The same would go, presumably, for the blockchain as a whole, and historical blocks.

I remember discussing this in the old thread.

Even with UTXO sets the full history would be preserved. The full history would be kept in archival nodes and there would only have to be a few of them that keep the full history secure. The vast majority of full nodes only need the current UTXO set and some amount of recent history. A year is probably plenty, there is no way to hide an illegal transaction that long ( or for any amount of time for that matter).
[doublepost=1448940266][/doublepost]

Btcdrak and others are loving it, linking to it all over the place because it looks like a massacre of all opposition to opt-in RBF.

Well that is easy to do if you silence everyone who disagrees. These are the tactics of tolitarian states.

 

[solex]

@Zangelbert Bingledack

The problem is that Jeff Garzik says the opt-in RBF change is OK, and IMHO he is an independent viewpoint yet close to the BS/Core Dev action. Few of us are able to challenge the minutiae of this change (I can't). However, I can recognise a slippery slope when I see one (my mental image here: an insect entering the attractive funnel of a pitcher plant!). Already there is talk of Full RBF, different flavors of it and even a default for it.

These are technocrats who are at points in a spectrum of willingness to abandon 0-conf because it is not perfect, or fraud-proof. The enormous economic benefit of the existing usage of 0-conf is ignored. When credit cards were invented a major feature was security by ink signature, yet when these cards were implemented over the internet the card companies knew they were abandoning a major security feature, but it was expedient to do so due to the potential economic benefit. And they didn't know how enormous internet retail would become, they just knew that they needed to go forward with an imperfect solution - or lose market share.

 

[cypherdoc]

keep re-reading this post by Andrew Miller that @Justusranvier posted last week to understand why gmax, an supposed expert in distributed systems, fails to understand how Bitcoin works.

tl;dr: Bitcoin depends on probabilities not hard solutions. he can't accept this.

"Random oracles (non-invertible cryptographic hash functions with uniform random outputs).
Hash functions are something we accept with faith, at least for the time being. It has been proven that no hash function behaves perfectly. On the other hand, some simple properties (like collision-resistance) are provably attainable. I doubt anyone would argue that Bitcoin relies on untenable properties of hashes. Nonetheless, this is a reason why their use has not been so widely explored in distributed systems (counter example) - from a very traditional point of view, they are too strong an assumption."

https://bitcointalk.org/index.php?topic=99631.0

 

[cypherdoc]

NBG just keeps going lower and lower. Talk about a knife catcher:

 

[albin]

Double-secret proposals in Hong Kong?!

 

[cypherdoc]

flip through these: http://coin.dance/charts

volume mirrors this type of activity:
http://www.ft.com/cms/s/0/7157cd8e-94f5-11e5-bd82-c1fb87bef7af.html#axzz3t23fh5gI

 

[rocks]

Double-secret proposals in Hong Kong?!

So... We should keep waiting and holding our breath?

Let me guess they have new proposals around LN or SC or something else that won't be ready in years solving a problem we don't have.

 

[awemany]

[...]
My guess is that we bubble at some point between now and the halving. Possibly a price surge timed for a scaling solution announced by Core (designed to head off complete mutiny in the ecosystem for XT or BU in the coming weeks - or at least appease industry for a year or two until LN can be pushed onto new users).

[...]

I would really hate to see people accept a faulty compromise with core (such as the '2-4-8'). It will also set us back in terms of decentralizing governance and I think that will reflect in the price. I appears that BIP101 is a compromise on the lower acceptable end (in terms of max. block size) for most already.
[doublepost=1448953173,1448952242][/doublepost]@Mengerian, @rocks

That make so much sense!

I am glad that others still remember that we have lots of options. Isn't it really sad that this part of the whole discussion got squeezed out by the Gregs steamrolling around, assuming Bitcoin is just the tarmac for their LN aspirations?

I mean, in the end operating a full node can realistically get down to constant storage and O(n) bandwidth and probably some O(n log^x(n)) CPU time.

Greg appears to have socially engineered the whole discussion so that 'all txn since genesis' (or non-working, still purely academic ZK proofs) is the only reliable way to run a full node. I have seen this sentiment repeated in pretty shallow ways on reddit a lot of times. Often and when pressed, people who are 'worried about blocksize' switch from being worried about bandwidth to being worried about storage if they start to lack arguments.

Don't get me wrong, I think there's value in keeping the whole history, and I also think some nodes will always do that (even if we would have 8GiB today...).

I am also not sure that we'll ever need that with still-to-be-made advances in storage. I tried to see where we could go in a reddit post on that topic.

But I think something that perpetually grows is making people feel uneasy, from a psychological perspective.

With FTTH, I am quite confident that bandwidth will never be a true issue for Bitcoin's decentralization.

I still remember the gold old times™ with idea threads on BCT that were in an optimistic and cooperative mode of open problem solving. Due to the hijacking of the Gregs and Adams and thermos cans, we lost a lot of inertia and productivity on that front.

 

[Justus Ranvier]

I still remember the gold old times™ with idea threads on BCT that were in an optimistic and cooperative mode of open problem solving. Due to the hijacking of the Gregs and Adams and thermos cans, we lost a lot of inertia and productivity on that front.

Hijacking may be too strong a term.

There are two groups in Bitcoin pursuing two irreconcilable goals:

* Everybody in the world should be able to transact on the blockchain.
* Everyone in the world should be able to fully and independently validate the blockchain.

No matter what happens, somebody is not going to get all of what they want.

 

[awemany]

@Justus Ranvier

That's a good perspective. The second 'requirement' got added later, though.

 

[awemany]

So what the fuck is going on here:
https://www.reddit.com/r/bitcoinxt/comments/3uyyfx/i_didnt_think_this_day_would_come_but_ive_been/

Is this some concerted psyops thing?

 

[cypherdoc]

You can see here we've made absolutely no progress in 10 months and more of arguing:

 

[awemany]

10months? More like 2+ years. I made my reddit account and went from lurker to active discussion participant solely because of the blocksize issue...

At least LukeJr is quite an interesting character. I almost like him

 

[cypherdoc]

So what the fuck is going on here:
https://www.reddit.com/r/bitcoinxt/comments/3uyyfx/i_didnt_think_this_day_would_come_but_ive_been/

Is this some concerted psyops thing?

Skimmed the thread. Any progress?

 

[Zarathustra]

So what the fuck is going on here:
https://www.reddit.com/r/bitcoinxt/comments/3uyyfx/i_didnt_think_this_day_would_come_but_ive_been/

Is this some concerted psyops thing?

Pseudo-libertarians and Idiots everywhere? Sad news:

 

[awemany]

Me and maaku7:


I have this constant deja vu feeling...

 

[Zangelbert Bingledack]

Nothing irks me quite like equivocation on terms. It's the very mechanism by which cognitive dissonance happens, and a favorite way for authority figures to cow their audiences.

"We have consensus on the change because all the devs agreed in our cloistered mailing list discussion which we carefully avoided bringing to the larger non-Core community of devs, or among whatever devs we specify are required to agree on a case by case basis after the fact, but consensus is the way Bitcoin code works and has a very clear definition."

Two completely different meanings of the word consensus, slid between like they meant the same thing. Even in the most charitable interpretation it's a sign of sloppy thinking.

 

[Impulse]

Double-secret proposals in Hong Kong?!

Color me skeptical. "Clever" means "complicated", so there is no chance that these are ready to go solutions, more likely is that they will push us further into the weeds. More debate, more discussion, a likely stalling tactic. More ammo against moving forward with the simple and already implemented solution. I have little faith that anything good will come from this. I sincerely hope that I am wrong.