Urban75 Home About Offline BrixtonBuzz Contact

Bitcoin discussion and news

proper unintended consequences :D

Injecting some tiny (and cheap) slogan about eg Falun Gong, holocaust denial or what lies behind a superinjunction could make downloading the blockchain illegal in some but not other territories, and injecting a virus signature can cause AV software to have an apoplexy. I didn't know any of that, it almost makes me want to set up a crypto account just to troll the blockchain evangelists.

However, the idea that inserting a 21kB image could cost $380 in electricity is utterly insane. I'm not clear whether that's a one-off cost or knocks on to every downstream transaction, does anyone know?
 
That's really cool and interesting. And funny.

However, the idea that inserting a 21kB image could cost $380 in electricity is utterly insane. I'm not clear whether that's a one-off cost or knocks on to every downstream transaction, does anyone know?
It would only cost you 20 dollars to post. The electrical cost is taken by the miners.
And only the first time. Not each future transaction.
 
It would only cost you 20 dollars to post. The electrical cost is taken by the miners.
That's not my reading of the paper but I'll freely confess a lot of it is beyond my understanding. This blog post is a bit easier to digest and this paper adds a bit of explanation.

SFAICS arbitrary data can only be inserted into the blockchain by miners in the process of verifying transactions. Like all other parts of the blockchain, that data can't subsequently be removed or changed by anyone.

If that's wrong I'd be grateful for an understandable source explaining how the blockchain can be modified (and kept secure) by anyone except during verification of a transaction, and whether there's an audit trail to show who modified the blockchain between transactions.

And only the first time. Not each future transaction.
Again, that's not the way I read it, which is that miners during all subsequent transactions must process all data for full cryptographic security to be maintained because they have no way of telling which bits of data are fake transactions or non-financial injected content.
 
That's not my reading of the paper but I'll freely confess a lot of it is beyond my understanding. This blog post is a bit easier to digest and this paper adds a bit of explanation.

SFAICS arbitrary data can only be inserted into the blockchain by miners in the process of verifying transactions. Like all other parts of the blockchain, that data can't subsequently be removed or changed by anyone.

If that's wrong I'd be grateful for an understandable source explaining how the blockchain can be modified (and kept secure) by anyone except during verification of a transaction, and whether there's an audit trail to show who modified the blockchain between transactions.


Again, that's not the way I read it, which is that miners during all subsequent transactions must process all data for full cryptographic security to be maintained because they have no way of telling which bits of data are fake transactions or non-financial injected content.

The data is just added by the people making the transactions. Instead of putting in a proper address, people are using hex. Nothing strange is happening.
Each new transaction set is only calculated once. Then its part of the block chain.
 
The data is just added by the people making the transactions. Instead of putting in a proper address, people are using hex. Nothing strange is happening.
Each new transaction set is only calculated once. Then its part of the block chain.
A source explaining the answers to the questions I asked?
 
I'm not sure what you want to know, and why.

Why do you think the Blockchain can be modified? Who said that?

Why do you think miners recalculate all the old blocks again, and not just the new one?
the assertions you made in #4056 don't correspond to what I've read, which I've detailed. So I'm asking you to back them up.
 
the assertions you made in #4056 don't correspond to what I've read, which I've detailed. So I'm asking you to back them up.
OK
1)
Here is the cost in fees to send 21kB of data.
Screen Shot 2018-03-21 at 07.58.25.png
2)
I don't really know how to backup the fact that miners only do the mathematical work to add a new block to the chain other than pointing you to this basic description of how the blockchain grows. Block - Bitcoin Wiki
 
OK
1)
Here is the cost in fees to send 21kB of data.
View attachment 130538
2)
I don't really know how to backup the fact that miners only do the mathematical work to add a new block to the chain other than pointing you to this basic description of how the blockchain grows. Block - Bitcoin Wiki
Thankyou.

1)
The article in the Register gave the insertion cost for 20kB as $380, but that's based on an interview, so it's unverifiable. I've found other cost estimates, but most low cost, high security insertion methods seem to rely on input script manipulation or fake transactions. Your snip gives little clue as to which insertion method is being used, but methods using multiple output scripts have limited capacity, which caps the overall data to somewhere around 1500B per output script, about half what's implied by your snip. So I'm puzzled. Without knowing the method it's hard to be sure that the following applies
Transactions can deviate from the approved
transaction templates [48] via their output scripts as well as input scripts. In the-
ory, such transactions can carry arbitrarily encoded data chunks. Transactions
using non-standard
output
scripts can carry up to 96
.
72 KiB at comparably low
costs. However, they are inefficient as miners ignore them with high probability.
Yet, non-standard output scripts occasionally enter the blockchain if miners in-
sufficiently check them
S2.1 sorry about the formatting

Which implies one needs to understand the probability of that insertion being permanently stored on the blockchain.

2)
As I said, the blockchain can only be permanently altered by mining, ie storing verification of transactions during the creation of new blocks. Each transaction currently costs 834kWh but I can't find anything discussing the marginal electrical cost of data insertion (or the data overheads involved). An entirely fake transaction (which I think you were implying in #4058) takes that much power at low cost to the injector. A series of fake or low-value-but-unnecessary real transactions used as carriers for data will cause absurd overall network costs, again for low injection costs (well, $30 for 20k is only 'low' in this circumstance, it's nonsensically high in every other context).

As with the rest of Bitcoin, this breaches the principle of 'Polluter Pays'.

Just for context, when we started discussing power consumption on this thread, back in November, each transaction took 236kWh. Now it's 834. Overall daily power consumption has more than doubled in that time.
 
Last edited:
Surely if this was just about drugs bitcoin would have remained little more than a curiosity confined mostly to the dark web?
 
Surely if this was just about drugs bitcoin would have remained little more than a curiosity confined mostly to the dark web?
I don't know the make up of total transactions tbh but the logic of the argument is exactly the same no matter what isn't it? In fact, that logic has been openly offered on here in defence of this monstrosity.
 
Fuck me, that's frightening.

I don't like frightening people, but... in November the country equivalent was Nigeria, number 67 worldwide. Now it's Greece, number 46.

There are around 1/3 fewer daily transactions now than in November.
 
Surely if this was just about drugs bitcoin would have remained little more than a curiosity confined mostly to the dark web?
That was and remains its only use value. The rest is commodity bubble speculation. If you look at the number of transactions per block, it bears no relation at all to the speculation bubble that has ballooned on top of it. That's one reason why this is different from the dotcom bubble. The dotcom bubble was related to a genuinely new way of doing things. This isn't. There is no equivalent to Amazon or Google in there among the thousands of cryptocurrencies. They're all AOL.
 
Every trade requires two transactions too, in practice -- one to obtain bitcoin and one to buy the thing with the bitcoin. So that's 4.4 months' worth of household electricity consumption for every trade.

Buy drugs with bitcoin just three times in a year and you've effectively doubled the amount of electricity you consume.

And that's just today. But the consumption per transaction has trebled in about 4 months. By the end of the year, you'll be doubling your electricity consumption for the year just by buying your drugs this way once per year.
 
That was and remains its only use value. The rest is commodity bubble speculation. If you look at the number of transactions per block, it bears no relation at all to the speculation bubble that has ballooned on top of it. That's one reason why this is different from the dotcom bubble. The dotcom bubble was related to a genuinely new way of doing things. This isn't. There is no equivalent to Amazon or Google in there among the thousands of cryptocurrencies. They're all AOL.

Well yes, currently drugs and other illegal stuff is its only use value. But presumably in amongst all the chancers there are people who believe that it will be so much more than it is, the future of currency exchanging in-itself?

I'm not really that up on it and it clearly is a dodgy pyramid scheme but will it develop into something more? I guess my point was this can't just be about a handful of druggies, right wing nuts and spivs looking for a quick buck regardless of the cost?
 
That was and remains its only use value. The rest is commodity bubble speculation. If you look at the number of transactions per block, it bears no relation at all to the speculation bubble that has ballooned on top of it. That's one reason why this is different from the dotcom bubble. The dotcom bubble was related to a genuinely new way of doing things. This isn't. There is no equivalent to Amazon or Google in there among the thousands of cryptocurrencies. They're all AOL.
good post, I agree.

I'd add that the current strand of discussion is about using the Bitcoin blockchain for something other than financial transactions- distributing illegal content, political slogans, RIPs, birthday greetings and all sorts of complete drivel (examples here). This has been around, but rather unknown, for a few years now but as knowledge of the capability grows (eg people on this thread now know), the relatively low costs to the originator means it could become a whole new significant use case.

At it's simplest and cheapest, creating a fake address with human readable characters and sending it say 10 US cents injects those characters, that slogan, into the Bitcoin blockchain forever, for a very low cost. Can be used for more or less anything, build it and they will come.
 
Well yes, currently drugs and other illegal stuff is its only use value. But presumably in amongst all the chancers there are people who believe that it will be so much more than it is, the future of currency exchanging in-itself?

I'm not really that up on it and it clearly is a dodgy pyramid scheme but will it develop into something more? I guess my point was this can't just be about a handful of druggies, right wing nuts and spivs looking for a quick buck regardless of the cost?
The likes of camouflage on here seem to be true believers and in denial about a lot of the wider consequences. The bubble is a result of people looking for a quick buck, and probably in many cases not thinking too hard about anything else.

I do think there is something about the set-up - creating rarity artificially - that has encouraged the bubble. Plus it appears to feed into a general mistrust of and anger towards the established financial order (although btc is dominated by Chinese money nowadays - not so much anger as trying to get around the country's capital controls). There is also an enormous amount of propaganda out there spreading layer upon layer of total bullshit about bitcoin, which gives it a veneer of respectability.
 
I don't like frightening people, but... in November the country equivalent was Nigeria, number 67 worldwide. Now it's Greece, number 46.

There are around 1/3 fewer daily transactions now than in November.
And that's just bitcoin. Add in all the other cryptononsenses and the figure is significantly higher.
 
The likes of camouflage on here seem to be true believers and in denial about a lot of the wider consequences. The bubble is a result of people looking for a quick buck, and probably in many cases not thinking too hard about anything else.

I do think there is something about the set-up - creating rarity artificially - that has encouraged the bubble. Plus it appears to feed into a general mistrust of and anger towards the established financial order (although btc is dominated by Chinese money nowadays - not so much anger as trying to get around the country's capital controls). There is also an enormous amount of propaganda out there spreading layer upon layer of total bullshit about bitcoin, which gives it a veneer of respectability.

Oh fuck off.

I was well excited about this edgy new tech for a while. Then I sobered somewhat because when all said and done- it doesn't matter what you use as money, power is what any arrangement comes down to in the end (and I don't mean megawatt hours). Now I just find the technology really interesting, nothing that's going to fundamentally change social relations or make revolutions (well, apart from it being another in a long line of Liberal revolutions to set capital free) but still interesting. I still find that many of the criticisms on this thread seemed not very... on point as it were with the exception of the energy question, where my position is simply "more optimistic" because exponential energy demand per unit isn't a good feature of any business plan (tech advancement usually means going in the other direction- faster, cheaper, lighter etc- though that still leads the door open to many many more units overall... transactions in this case).

The real difference between me, and the likes of you lbj, is that I have always taken bitcoin seriously. from the first time I heard about it I saw it as full of potential of some sort. You considered it joke for the most part until suddenly you went all alarmed. Even now you want to see bitcoin as a silly joke on the one hand and a serious threat to the economy and the environment on the other. You want to eat your cake and to have it too.

I take bitcoin seriously- I don't doubt that there's the potential for bitcoin to seriously fuck up the world- or for that potential to be taken seriously enough within the bitcoin community for ways to be found for it not to seriously fuck up the world. either way, it's a serious discussion, not a joke. Grok that if you are capable.
 
Last edited:
Thankyou.

1)
The article in the Register gave the insertion cost for 20kB as $380, but that's based on an interview, so it's unverifiable. I've found other cost estimates, but most low cost, high security insertion methods seem to rely on input script manipulation or fake transactions. Your snip gives little clue as to which insertion method is being used, but methods using multiple output scripts have limited capacity, which caps the overall data to somewhere around 1500B per output script, about half what's implied by your snip. So I'm puzzled. Without knowing the method it's hard to be sure that the following applies

S2.1 sorry about the formatting

Which implies one needs to understand the probability of that insertion being permanently stored on the blockchain.

2)
As I said, the blockchain can only be permanently altered by mining, ie storing verification of transactions during the creation of new blocks. Each transaction currently costs 834kWh but I can't find anything discussing the marginal electrical cost of data insertion (or the data overheads involved). An entirely fake transaction (which I think you were implying in #4058) takes that much power at low cost to the injector. A series of fake or low-value-but-unnecessary real transactions used as carriers for data will cause absurd overall network costs, again for low injection costs (well, $30 for 20k is only 'low' in this circumstance, it's nonsensically high in every other context).

As with the rest of Bitcoin, this breaches the principle of 'Polluter Pays'.

Just for context, when we started discussing power consumption on this thread, back in November, each transaction took 236kWh. Now it's 834. Overall daily power consumption has more than doubled in that time.

1)
Cool. 2.1 of that document is a great summary of what the options are.

Transaction fees fluctuate a lot. Chances are that it was that much a couple of months ago.

Adding 21kB as a "standard transaction" would be a single transaction from one address to loads of destination addresses. e.g - one address paying out to a few thousand addresses.
If you pay fees as I mentioned above you'd get this on the blockchain in two hours - no problem.
(you don't use real destination addresses though - you use the data you want to publish)

Adding 21kB as a "non-standard transaction" is an attempt to save cost by being a bit hacky, and hoping no one checks what you are doing.

When a miner solves a block and adds it to the chain, they receive a reward of some bitcoin. They also have the option of adding in peoples transactions to the block and keeping the fees they paid.
They don't have to add any transactions, if they don't want. But obviously they do for the fees and to keep the system running.
The miners should have a system in place that picks the biggest fee payers first, which are also the smallest in size.
This "non-standard transaction" technique is hoping that miners aren't very thorough with the above, and publish their transaction and extra data. But if they are validating that the transactions are sensible, then the miner will just ignore the gibberish transaction.

2)
yeah - the system is fucked. The power consumption increases to meet/combat the excessive number of miners. Its gone a bit crackers.
It's not sustainable at all. I'm interested to see if it finds a level that works.

Side question - You say "transaction currently costs 834kWh" - As far as I know that might well be right. But are you sure its per transaction and not per block?

I'm not implying a fake transaction. None of them are fake as far as I am aware. I am saying that either the transaction payment is legitimate but destination address is not a real address (in which case the transaction is still added to the block and the money waits in limbo for someone with the correct key to pick it up). Or people make a legitimate transaction and cram some stuff in the blank space that they've stretched out on the transaction.

Either way the power consumption is not much different from a normal transaction. The price doesn't effect the power. Similarly - price doesn't affect fees.
 
Every trade requires two transactions too, in practice -- one to obtain bitcoin and one to buy the thing with the bitcoin. So that's 4.4 months' worth of household electricity consumption for every trade.

Buy drugs with bitcoin just three times in a year and you've effectively doubled the amount of electricity you consume.

And that's just today. But the consumption per transaction has trebled in about 4 months. By the end of the year, you'll be doubling your electricity consumption for the year just by buying your drugs this way once per year.

Save the world - bulk buy your drugs
 
Re cryptografitti - yes bollocks, but the ability to stamp transactions " Property of the Islamic republic of Iran" could add some spice to the game
 
Back
Top Bottom