Hyperdeflation, own half the world by headstart - don't you care at all?

[MoonShadow]

AFAIK pruned merkle trees can only be used to verify your own balance, and not anyone else's. That is, you can't mine using only that, and you if you can't mine then you also can't verify a tx someone has sent to you.

You can if your pruning algo follows the specs in the white paper.  Again, as you have already mentioned, you don't know.  You've done some research about the topics at hand, this much I can tell; however you don't completely 'grok' bitcoin.  Don't take that too personally, it took me weeks before I even understood the most obvious functions of the system.  The system works in so many interconnected ways to support itself, that I have insisted that Satoshi was either a team of professionals acting as an amateur or a truly gifted polymath just learning the art of programming.  The truth is, the system is elegant with finely tuned incentives, interactions and security tricks all in play, not some kid genius's after-school project hack.

For these reasons and others, it would be a terrible waste of effort and bandwidth form me to attempt to explain any of it to you here.  You're simply not close enough to it yet.

[DeathAndTaxes]

By reducing the blockchain to the current unspent txOuts, he was able to reduce the total size from ~1GB at the time to 75MB, the majority of which was taken up by the scripts. Without needing scripts for standard tx, that could probably be cut in half, and if unspent txOuts were reduced to account balances it could probably be cut down to 1/4 that. The biggest change that would entail is that txIn would take the full value of the account rather than the full value of each referenced txOut, which would no longer be required for txIns. You would still have to store a few weeks of blockchain on top of that, but that's still < 100MB easily.

You don't see a problem with that? Currently a 51% attack "only" alloww the attacker to
* stop the network
* double spend his own coins (reverse txs he created)

If network is reduced to a flat ledger a 51% attack would allow the attacker to change balances of accounts he doesn't own which massively increases the economic value of 51% attack.  Even the potential is damaging.

Then again 75MB + 50MB per week (say 4 weeks of history retained) = 275 MB which last time I checked > 50MB claimed.  Also that was a back of the napkin estimate.  Someone actually did create a ledger and the size was 10% of full blockchain.  So that would be more like 200MB + 200MB for 4 weeks of history = 400 MB. 

The cost estimate for a pruned blockchain is ~500MB vs 2GB raw so you are saving very little.  Pruned block chain provides most of the space reduction with no loss of security.

AFAIK pruned merkle trees can only be used to verify your own balance, and not anyone else's. That is, you can't mine using only that, and you if you can't mine then you also can't verify a tx someone has sent to you.

Not correct but at least this time you correctly indicated it it your belief not fact.  

Pruned merkle tree allows verification of blocks, verification of your tx, verification of other users tx, and mining new blocks.  There is no functionality that is lost with the exception of tracing coin history beyond the oldest unspent output.

To verify tx (yours or anyone elses) you simply need to verify the prior outputs of the inputs in the tx to be verified.  By definition those would be included in a pruned blockchain.  Mining doesn't require anything beyond verifying txs other than access to prior blocks block header, the block height, and the ability to verifying work created by peers.

It might be a good idea to actually read the white paper.

[ryu-fk]

Why are you guys trying to defend deflation? Of course it's bad it stops people spending, acts as a disincentive for people to sell products. Computers are different as the decrease in price has come about due to technological innovation

realnowhereman sarcastically said "Yeah; I always defer my consumption of food and water because I know I'll be able to get them cheaper next week.". This is a silly point food and water are obviously not going to be affected because they are essential goods with inelastic demand but you will defer the purchase of a car if you know its going to be a lot cheaper next month.

[DeathAndTaxes]

realnowhereman sarcastically said "Yeah; I always defer my consumption of food and water because I know I'll be able to get them cheaper next week.". This is a silly point food and water are obviously not going to be affected because they are essential goods with inelastic demand but you will defer the purchase of a car if you know its going to be a lot cheaper next month.

Already asked and answered.

So you haven't yet bought a HDTV right?  I mean they likely will be cheaper next year (and they were cheaper this year compared to last, and both were cheaper than 2 years ago, etc).

So you are still waiting?  Same thing with cellphones, laptops, gaming consoles, video games, DVD/Blurays, Video on demand, etc.

You haven't bought any of those right?  Price deflation has destroyed all those industries.  I mean just look at Apple.  Its stock is at a 52 week low because people are delaying purchasing an ipad because the KNOW there will be one cheaper/better out in less than a year.

That is just on the consumer side.  Businesses (where ROI% rules) haven't even upgraded to gigabit LANs yet.  I mean the cost is bound to keep falling by 30% or more per port every year.  What kind of idiot company would pay $20 per port today for a managed switch when Moore's law all but ensures they can get the same managed switch for $10 per port next year.  Don't even get me started on the internet.  We all know deflation has killed backbone spending.  Nobody is upgrading carrier grade switches and routers.  CISCO is almost bankrupt with all the backbone grade equipment which goes unsold every year.  The internet backbone is only going to get faster once we see some price inflation in bandwidth costs.

[ryu-fk]

Technology is a unique sector though, deflation can destroy other sectors like the car industry. If deflation is present in an economy people will be reluctant to buy now due to their future price expectations. Yes some individuals will continue to buy now, but take the whole economy into consideration and overall demand will be down which causes increased unemployment, firms will be unwilling to expand when future prices are going to be dropping. Modems, routers and Ipads are not the whole economy!
Deflation is more dangerous than inflation. A low stable rate of inflation is the best situation.

The price deflation in technology areas are due to innovation and cheaper production costs which is different from an economy in deflation!

[DeathAndTaxes]

Technology is a unique sector though, deflation can destroy other sectors like the car industry.

So companies waste millions of dollars buying IT products when they could just wait?  Of course not.  They buy new equipment because despite the fact that prices will be lower next year they gain MORE by making that purchase today.

Likewise say you decide you will wait a year to buy a car.  Guess what?  Prices are 5% lower.  So you should buy now right?   You saved 5%. But wait they will be 5% lower next year so you should wait?  Next year they will be 5% lower?  So you should wait.  Next year they will be 5% lower.  So you should wait?  End result you will never buy a car?  You should just wait 10,000 years and car will be 0.001% of the price today.   Just like a company will never upgrade to gigabit ethernet?  Someday gigabit ethernet will be ebay salvage so they should just upgrade then.

Even with falling prices there is an equilibrium where a new (or newer) vehicles is cheaper than waiting and using an older vehicle.  If Bitcoin base grew large enough eventually the price deflation would decline to a small % as the monetary base remains fixed and demand for currency grows at a slow rate. 

The price deflation in technology areas are due to innovation and cheaper production costs which is different from an economy in deflation!

Production costs for vehicle manufacturing adjusted for inflation are lower today than at any point in the past.  The idea that the tech sector is magical is silly.  All industries use technology to boost productivity. 

some individuals will continue to buy now, but take the whole economy into consideration and overall demand will be down which causes increased unemployment, firms will be unwilling to expand when future prices are going to be dropping. Modems, routers and Ipads are not the whole economy!

So users will continue to buy ipads, laptops, cellphones, video games, movies, online services, bandwidth, storage, etc TODAY knowing all those products will be massively cheaper in a year.  We aren't talking about 5% price deflation we are talking about 20%, 30%, sometimes 50% price deflation.  So despite knowing they could save as much as 50% they DON'T delay consumption. However those same consumers WON'T buy a car because it will be only 5% cheaper next year. 

[MoonShadow]

Why are you guys trying to defend deflation? Of course it's bad it stops people spending, acts as a disincentive for people to sell products. Computers are different as the decrease in price has come about due to technological innovation

realnowhereman sarcastically said "Yeah; I always defer my consumption of food and water because I know I'll be able to get them cheaper next week.". This is a silly point food and water are obviously not going to be affected because they are essential goods with inelastic demand but you will defer the purchase of a car if you know its going to be a lot cheaper next month.

Computers are not different, just an extreme example.  Deflation is neither good nor bad unto itself, but think about the consumer.  Is it bad for the consumer if the price of a gallon of milk drops 5% a year?  Of course not!  Is it bad for the farmer?  Maybe, but his own costs are likely dropping by 5% per year also.

[Haplo]

You don't see a problem with that? Currently a 51% attack "only" alloww the attacker to
* stop the network
* double spend his own coins (reverse txs he created)

If network is reduced to a flat ledger a 51% attack would allow the attacker to change balances of accounts he doesn't own which massively increases the economic value of 51% attack.  Even the potential is damaging.

Obviously you didn't read very far. Ledgers would be validated in the same way as blocks, via hashing power with periodic ledger hashes. They could additionally be verified with proof of stake, and you could reduce the maximum reorg down to 24h of blocks, or possibly even less with additional measures.

Then again 75MB + 50MB per week (say 4 weeks of history retained) = 275 MB which last time I checked > 50MB claimed.  Also that was a back of the napkin estimate.  Someone actually did create a ledger and the size was 10% of full blockchain.  So that would be more like 200MB + 200MB for 4 weeks of history = 400 MB. 

The cost estimate for a pruned blockchain is ~500MB vs 2GB raw so you are saving very little.  Pruned block chain provides most of the space reduction with no loss of security.

If you had read more than one post, you'd know that he got it down to 7% of the total blockchain size, not 10%. 50MB per week suggests a 5GB blockchain by the end of 2012. Either the blockchain is more than doubling in size every year now, which is in excess of moore's law, or you're full of crap on that figure.

AFAIK pruned merkle trees can only be used to verify your own balance, and not anyone else's. That is, you can't mine using only that, and you if you can't mine then you also can't verify a tx someone has sent to you.

Not correct but at least this time you correctly indicated it it your belief not fact.  

Pruned merkle tree allows verification of blocks, verification of your tx, verification of other users tx, and mining new blocks.  There is no functionality that is lost with the exception of tracing coin history beyond the oldest unspent output.

To verify tx (yours or anyone elses) you simply need to verify the prior outputs of the inputs in the tx to be verified.  By definition those would be included in a pruned blockchain.  Mining doesn't require anything beyond verifying txs other than access to prior blocks block header, the block height, and the ability to verifying work created by peers.

It might be a good idea to actually read the white paper.

You know I put off reading the whitepaper primarily because I assumed it would be a full dissertation. I was disappointed.

However now I understand the way merkle trees work for the most part. Here's a comparison:

-When you prune merkle trees, you still have to keep the block header, all of the merkle tree hashes, and the full tx of any tx that have any unspent txOuts, including all of the txIns, all of the scripts for the txIns, all of the txOuts (including spent txOuts) plus all of the scripts for the txOuts. Verifying tx still requires either a full pruned blockchain or that you trust centralized nodes to provide you with the correct merkle tree entries that you request.

-For the linked ledger system, you throw away all but the unspent txOuts plus the scripts for the unspent txOuts. All block headers older than a certain time can be discarded along with all of the merkle hashes and tx, and each client only needs to keep 4-5 weeks of blockchain and 3 ledgers (1 for an old seeding period, 1 backup, 1 current), of which 2 weeks of blockchain and 2 ledgers can be compressed since they're seldom if ever needed.

-If all unspent txOuts which have the same address are added together and condensed, that would reduce the size of a ledger by ~50% or more. This would also require using the full value of an acct as txIn, but that also reduces the size of txIns on the blockchain as a result.

2000MB * .035 = 70MB Per ledger
70MB * 2 = 140MB of ledgers that can be compressed
33MB (est weekly block size with reduced txIns) * 3 = 99MB uncompressed blockchain (maximum, varies)
66MB compressible blockchain (2wks historical for seeding)

70 + 99 = 169MB uncompressed
140 + 66 = 206MB compressible

I think your 500MB estimate for pruned merkle trees is overly optimistic. I'll believe it when I see it. Assuming you can compress data to 66% of original using zip or similar, with ledgers you'd only keep a maximum of about 305MB, and only 215MB for a lite client.

That's assuming that 6 months of total blockchain are retained over the mining network for seeding so that clients can do a "full verification" on first start, and is still ~1/2 to 1/3 the size of a fully pruned complete blockchain, and ~10% of an unpruned one, even with backups.

[MoonShadow]

You don't see a problem with that? Currently a 51% attack "only" alloww the attacker to
* stop the network
* double spend his own coins (reverse txs he created)

If network is reduced to a flat ledger a 51% attack would allow the attacker to change balances of accounts he doesn't own which massively increases the economic value of 51% attack.  Even the potential is damaging.

Obviously you didn't read very far. Ledgers would be validated in the same way as blocks, via hashing power with periodic ledger hashes. They could additionally be verified with proof of stake, and you could reduce the maximum reorg down to 24h of blocks, or possibly even less with additional measures.

If you think that you could do better, do it.

[realnowhereman]

realnowhereman sarcastically said "Yeah; I always defer my consumption of food and water because I know I'll be able to get them cheaper next week.". This is a silly point food and water are obviously not going to be affected because they are essential goods with inelastic demand but you will defer the purchase of a car if you know its going to be a lot cheaper next month.

You obviously didn't read much further in the thread.  I already answered that point (I even directly covered the car example).

Food and water are just an easy example.  All products are the same.

Inflation and deflation are irrelevant; all they do is move the threshold of action to a different price point.  I happily agree that a sudden change in inflation or deflation rate would be hard for an economy to adjust to quickly; but at any absolute level there will always be a point when $UTILITY_OF_PRODUCT > $SAVINGS_AVAILABLE_FROM_DEFLATION.  At that point (which will be chosen by each individual) they act.

Computers are not special; they are simply a convenient example -- they are experiencing price reduction on such a huge scale that they are even outstripping the massive inflation built into our present economies.  The behaviour of the buyers is not special though: they defer purchase of a computer until the utility they perceive they will gain from the computer is greater than that saving.  e.g. they will value access to the internet, ability to store/retrieve work, access to huge software libraries more than they value the net present value of the cash in their pockets.  How is that different for any product with any level of deflation in the economy?

To answer your car example (again): you say that you would defer your car purchase for a month because it will be cheaper then.  What you don't say is that for that month you have no car.  Whatever utility you personally get from that car, you have to do without it for the month in order to save the money -- that has a value too.

If I may be so bold; I suggest a reading of Bastiat's Essays on Political Economy (http://www.gutenberg.org/files/15962/15962-h/15962-h.htm).  Particularly the Capital and Interest section; which wonderfully demonstrates why utility has value.