Bitcoin Forum
November 12, 2024, 02:00:14 AM *
News: Check out the artwork 1Dq created to commemorate this forum's 15th anniversary
 
   Home   Help Search Login Register More  
Pages: « 1 2 [3] 4 »  All
  Print  
Author Topic: The Bitcoin scripting system is purposefully not Turing-complete - why?  (Read 3439 times)
Peter R
Legendary
*
Offline Offline

Activity: 1162
Merit: 1007



View Profile
May 20, 2014, 04:36:22 AM
 #41


Quote
It is generally accepted that the market value of a commodity and the cost to produce that commodity tend to the same price in a free and competitive market. 

these are not concrete economics ideas, sorry.  This is the gold bug mindset and as the story has been told countless times this ignores the effects of credit, innovation and imho the entire premise of civilization itself.  Is man just a resource extractor?


What does this have to do with gold other than gold being an example of a commodity?  The cost to produce corn tends to the market price of corn, the cost to produce aluminum tends to the market price for aluminum, the cost to produce canola oil tends to the market price for canola oil.  The reason is very simple: if the market price for corn is significantly greater than the cost of production, farmers will plant more corn and less potatoes to earn a greater profit. This action will increase the supply of corn, which will tend to reduce its market price, and eventually bring the profit margin for growing corn in line with other produce that could be produced on the same farm.   

Run Bitcoin Unlimited (www.bitcoinunlimited.info)
Peter R
Legendary
*
Offline Offline

Activity: 1162
Merit: 1007



View Profile
May 20, 2014, 05:07:28 AM
Last edit: May 20, 2014, 06:23:18 AM by Peter R
 #42


Quote
The market value of the Nxt network should reflect the present value of all future (discounted) cash flows.  

more or less true.  It's not just cash flows because the software also supports(and charges for) asset issuance, trades, and other upcoming features like credit and virtual corporations.


When I say cash flows I mean money, not shares of Nxt.  If the software "supports (and charges for) asset issuance, trades, and other upcoming features like credit and virtual corporations," then the Nxt network should earn money for that service and distribute it to its share holders.  

I suppose the way Nxt earns money is by people buying Nxt (with bitcoin for example) for the purpose of consuming their Nxt tokens to buy some service (rather than forging).  This is what gives you the "E" in the P/E ratio.  So if it costs $1 to issue a "virtual corporation" and that $1 gets used up (distributed across the network to forgers), then the network earned $1.  

I did some quick calculations and at a 10% cost of capital (cryptos are risky), the Nxt network would need to consume about $10,400 / day in fees (for transactions, issuing virtual companies and assets, etc.) to support it's current $38,000,000 market cap.

I'm trying to find out how much the Nxt network forgers earn per day on average, but I can't find the information.  Do you have data on the number of transactions per day and total amount of fees paid to forgers?  

Run Bitcoin Unlimited (www.bitcoinunlimited.info)
bluemeanie1
Sr. Member
****
Offline Offline

Activity: 280
Merit: 257


bluemeanie


View Profile WWW
May 20, 2014, 04:49:01 PM
 #43


Quote
It is generally accepted that the market value of a commodity and the cost to produce that commodity tend to the same price in a free and competitive market. 

these are not concrete economics ideas, sorry.  This is the gold bug mindset and as the story has been told countless times this ignores the effects of credit, innovation and imho the entire premise of civilization itself.  Is man just a resource extractor?


What does this have to do with gold other than gold being an example of a commodity?  The cost to produce corn tends to the market price of corn, the cost to produce aluminum tends to the market price for aluminum, the cost to produce canola oil tends to the market price for canola oil.  The reason is very simple: if the market price for corn is significantly greater than the cost of production, farmers will plant more corn and less potatoes to earn a greater profit. This action will increase the supply of corn, which will tend to reduce its market price, and eventually bring the profit margin for growing corn in line with other produce that could be produced on the same farm.   

Peter,  is the market for microchips related to the cost of sand?

after all microchips are made out of silicon right?

do you have any more pulp fiction economics to share with us?

-bm

ps.  Grin

Just who IS bluemeanie?    On NXTautoDAC and a Million Stolen NXT

feel like your voice isn't being heard? PM me.   |   stole 1M NXT?
bluemeanie1
Sr. Member
****
Offline Offline

Activity: 280
Merit: 257


bluemeanie


View Profile WWW
May 20, 2014, 05:15:19 PM
 #44


Quote
The market value of the Nxt network should reflect the present value of all future (discounted) cash flows.  

more or less true.  It's not just cash flows because the software also supports(and charges for) asset issuance, trades, and other upcoming features like credit and virtual corporations.


When I say cash flows I mean money, not shares of Nxt.  If the software "supports (and charges for) asset issuance, trades, and other upcoming features like credit and virtual corporations," then the Nxt network should earn money for that service and distribute it to its share holders.  


that's precisely what happens.  The TX fees go to the owners of NXT in proportion to their holdings(roughly).  It's a stochastic process.

I suppose the way Nxt earns money is by people buying Nxt (with bitcoin for example) for the purpose of consuming their Nxt tokens to buy some service (rather than forging).  This is what gives you the "E" in the P/E ratio.  So if it costs $1 to issue a "virtual corporation" and that $1 gets used up (distributed across the network to forgers), then the network earned $1.  

I did some quick calculations and at a 10% cost of capital (cryptos are risky), the Nxt network would need to consume about $10,400 / day in fees (for transactions, issuing virtual companies and assets, etc.) to support it's current $38,000,000 market cap.

you are quite confused sir.  When we speak of P/E in NXT we're talking about how much TX fees you earn per NXT.  What you just mentioned here is APPRECIATION or Capital Gains, that's not P/E. 

-bm

Just who IS bluemeanie?    On NXTautoDAC and a Million Stolen NXT

feel like your voice isn't being heard? PM me.   |   stole 1M NXT?
Cryddit
Legendary
*
Offline Offline

Activity: 924
Merit: 1132


View Profile
May 20, 2014, 08:05:12 PM
 #45

Thanks. But I guess there must be tools that convert high level Turing-complete languages to low-level BTCscript code. Right?

No.

That's more or less the definition of 'Turing-complete' - if you have a Turing-complete language, then anything that can be expressed in it can be translated into any other Turing-complete language. 

The reason they're called that is because Alan Turing proved a result in computability theory that showed all such languages are absolutely equivalent in terms of what they can express.

If you have something that isn't a Turing-complete language, then it is fundamentally not able to express some set of algorithms in the set that Turing-complete languages can express.

IOW, if you're writing in a Turing-complete language, then much of what you write can never be expressed in BTC's script language.  Ever.  At all. 

Peter R
Legendary
*
Offline Offline

Activity: 1162
Merit: 1007



View Profile
May 20, 2014, 08:32:34 PM
 #46


Quote
It is generally accepted that the market value of a commodity and the cost to produce that commodity tend to the same price in a free and competitive market. 

these are not concrete economics ideas, sorry.  This is the gold bug mindset and as the story has been told countless times this ignores the effects of credit, innovation and imho the entire premise of civilization itself.  Is man just a resource extractor?


What does this have to do with gold other than gold being an example of a commodity?  The cost to produce corn tends to the market price of corn, the cost to produce aluminum tends to the market price for aluminum, the cost to produce canola oil tends to the market price for canola oil.  The reason is very simple: if the market price for corn is significantly greater than the cost of production, farmers will plant more corn and less potatoes to earn a greater profit. This action will increase the supply of corn, which will tend to reduce its market price, and eventually bring the profit margin for growing corn in line with other produce that could be produced on the same farm.   

Peter,  is the market for microchips related to the cost of sand?

after all microchips are made out of silicon right?

do you have any more pulp fiction economics to share with us?


I'm not sure if you are being intentionally obtuse, or if you actually don't understand:

The market value of a commodity microchips and the cost to produce that commodity those microchips tend to the same price in a free and competitive market.

If the market price for "Commodity Microchip A" was vastly greater than its cost of production, then manufacturers would make more Microchip As and less Microchip Bs to earn a greater profit.  This free-market dynamic acts to bring the cost of production inline with the market price of that commodity. 

This is only true for commodities, however.  Take something like 20nm SHA256 ASICs for bitcoin mining.  The market price for these ASICs is much higher than the cost of production because this technology is not yet commoditized.  Only certain companies have the IP/trade-secrets for this technology, which means they can earn very high profits (far above the cost of production) since they have few competitors.  But what we will see over the next few years is more players developing their own SHA256 ASICs to compete for the large profit margins currently enjoyed by KNCMiner, etc.  This will drive the price of these ASICs down until they eventually become a commodity like a standard 12-bit analog-to-digital converter.

Run Bitcoin Unlimited (www.bitcoinunlimited.info)
jonald_fyookball
Legendary
*
Offline Offline

Activity: 1302
Merit: 1008


Core dev leaves me neg feedback #abuse #political


View Profile
May 20, 2014, 08:43:26 PM
 #47

You're on point as usual Peter.

James D'angelo recently said a SHA-256 chip is a relatively simple chip,
which really surprised me, so it could happen fairly quickly.

telepatheic
Jr. Member
*
Offline Offline

Activity: 56
Merit: 1


View Profile
May 20, 2014, 08:58:35 PM
 #48

James D'angelo recently said a SHA-256 chip is a relatively simple chip,
which really surprised me, so it could happen fairly quickly.

Yes and no, a SHA256 chip is quite simple but each mining chip is actually more complex as it includes many parallel SHA256 engines so a similar number of transistors to a standard CPU. Not only that, but 20nm is smaller than you manufacture most chips except for CPUs. Most micro-controllers and other ASICs are built using a far cheaper 90nm process.

On top of that, heat dissipation is incredibly important with bitcoin miners as they run continuously. This involves some very careful circuit design to distribute the production of heat over the whole surface of the chip and ensure that thermal runaway doesn't occur.
ThePurplePlanet
Full Member
***
Offline Offline

Activity: 144
Merit: 100


View Profile
May 20, 2014, 09:50:34 PM
 #49

to further back up my point, here are some trading stats from the NYSE.

http://www.allcountries.org/uscensus/835_volume_of_trading_on_new_york.html

so in 1999, we had daily highs of several billion trades a day.  Since 2000 this has increased considerably.

now keep in mind that under the current schedule for ie. Color Coins, *creating* a new ticker costs a few cents.  Thus it's reasonable to expect even more than 1 billion transactions a day.  I think you're ignoring the problem because you would rather attention not be going to alternatives like NXT that do have the possibility of supporting this sort of volume ...

I am not ignoring the "problem".  Volume on NYSE is in the billions of shares because fees are insanely low, less than $0.00004 USD (88 satoshis) per share.   Any assets traded on the blockchain would need to compete with bitcoin transactions and fees will probably not be that low.  If NXT is more economical for colored coins then the assets will move there so I don't see a problem either way.


this problem doesn't exist in NXT because the cost of an attack scales perfectly in ratio to the cost of NXT, whereas in Bitcoin it scales to the cost of hashing power.




"to the cost of NXT" ....eeerrr.... which is precisely zero?

What stops you as an early adopter of NXT once you sell out  to join an alternate chain of NXT and reverse all transactions at no cost? So you propose  a solution in which late adopters always depend on earliest adopters good faith to keep the longest chain ownest?
bluemeanie1
Sr. Member
****
Offline Offline

Activity: 280
Merit: 257


bluemeanie


View Profile WWW
May 20, 2014, 10:42:41 PM
 #50

to further back up my point, here are some trading stats from the NYSE.

http://www.allcountries.org/uscensus/835_volume_of_trading_on_new_york.html

so in 1999, we had daily highs of several billion trades a day.  Since 2000 this has increased considerably.

now keep in mind that under the current schedule for ie. Color Coins, *creating* a new ticker costs a few cents.  Thus it's reasonable to expect even more than 1 billion transactions a day.  I think you're ignoring the problem because you would rather attention not be going to alternatives like NXT that do have the possibility of supporting this sort of volume ...

I am not ignoring the "problem".  Volume on NYSE is in the billions of shares because fees are insanely low, less than $0.00004 USD (88 satoshis) per share.   Any assets traded on the blockchain would need to compete with bitcoin transactions and fees will probably not be that low.  If NXT is more economical for colored coins then the assets will move there so I don't see a problem either way.


this problem doesn't exist in NXT because the cost of an attack scales perfectly in ratio to the cost of NXT, whereas in Bitcoin it scales to the cost of hashing power.




"to the cost of NXT" ....eeerrr.... which is precisely zero?

What stops you as an early adopter of NXT once you sell out  to join an alternate chain of NXT and reverse all transactions at no cost? So you propose  a solution in which late adopters always depend on earliest adopters good faith to keep the longest chain ownest?

how exactly do you figure the cost of NXT is ZERO?

the cost of NXT is not zero and as more value becomes vested in the chain, the trading value of NXT increases(the detente scenario).  This is compounded to the fact that NXT also generates TX fees.

It's been discussed already, if someone were to try and bite off a significant chunk of NXT the price would multiply.  To get a stake capable of an attack?  Not even sure if that would even be possible at this point.

-bm

Just who IS bluemeanie?    On NXTautoDAC and a Million Stolen NXT

feel like your voice isn't being heard? PM me.   |   stole 1M NXT?
telepatheic
Jr. Member
*
Offline Offline

Activity: 56
Merit: 1


View Profile
May 20, 2014, 10:48:29 PM
 #51


It's been discussed already, if someone were to try and bite off a significant chunk of NXT the price would multiply.  To get a stake capable of an attack?  Not even sure if that would even be possible at this point.

Except for the fact that I have demonstrated here that NXT is trivial to attack with only a small stake and a small amount of computational power.
bluemeanie1
Sr. Member
****
Offline Offline

Activity: 280
Merit: 257


bluemeanie


View Profile WWW
May 20, 2014, 10:56:09 PM
 #52


It's been discussed already, if someone were to try and bite off a significant chunk of NXT the price would multiply.  To get a stake capable of an attack?  Not even sure if that would even be possible at this point.

Except for the fact that I have demonstrated here that NXT is trivial to attack with only a small stake and a small amount of computational power.

you haven't demonstrated that.

-bm

Just who IS bluemeanie?    On NXTautoDAC and a Million Stolen NXT

feel like your voice isn't being heard? PM me.   |   stole 1M NXT?
ThePurplePlanet
Full Member
***
Offline Offline

Activity: 144
Merit: 100


View Profile
May 20, 2014, 11:08:28 PM
Last edit: May 20, 2014, 11:26:50 PM by ThePurplePlanet
 #53


It's been discussed already, if someone were to try and bite off a significant chunk of NXT the price would multiply.  To get a stake capable of an attack?  Not even sure if that would even be possible at this point.

Except for the fact that I have demonstrated here that NXT is trivial to attack with only a small stake and a small amount of computational power.

you haven't demonstrated that.

-bm

What do you mean? What are these arguments brought here?

Once people sell out they can join a StakeUndo service like http://www.bitundo.com/ and produce their own parallel chain. Say currently 1% of coins is used to secure the network people owning at some point in the past 2% will do it.  Once such pool is set up I ll be happy to join... I will want my stake back.. It costs me nothing anyway.

Thinking this a bit further .... One with a small stake can buy and sell from  different exchanges (transfer from exchanges to exchanges to different accounts by mixing) and acquire stake history of much larger proportions than his current stake ..... Can acquire additional stake history by running other services like mixing, lending stake power and running sevices like casino or stake vaults..... This can give a person or a group of people enormous stake history to attack without even owning much at the first place....  Is this a joke?
Cryddit
Legendary
*
Offline Offline

Activity: 924
Merit: 1132


View Profile
May 20, 2014, 11:08:38 PM
 #54

Really?  

Seriously, you are going to insist on baiting someone into demonstrating this attack before you believe in it?  

You just like to watch the world burn, huh?
bluemeanie1
Sr. Member
****
Offline Offline

Activity: 280
Merit: 257


bluemeanie


View Profile WWW
May 20, 2014, 11:10:48 PM
 #55

so far what Im seeing here are a series of broken arguments, invalid premises, personal profit motives, and standard web forum showboating.

-bm

Just who IS bluemeanie?    On NXTautoDAC and a Million Stolen NXT

feel like your voice isn't being heard? PM me.   |   stole 1M NXT?
telepatheic
Jr. Member
*
Offline Offline

Activity: 56
Merit: 1


View Profile
May 20, 2014, 11:13:59 PM
 #56


It's been discussed already, if someone were to try and bite off a significant chunk of NXT the price would multiply.  To get a stake capable of an attack?  Not even sure if that would even be possible at this point.

Except for the fact that I have demonstrated here that NXT is trivial to attack with only a small stake and a small amount of computational power.

you haven't demonstrated that.

-bm

Ok so I haven't produced a real world demonstration, but I have essentially given you the pseudocode. All I need to do now is get an NXT stake and figure out how to compile a custom NXT client. No one has given me any reason why this wouldn't work.
bluemeanie1
Sr. Member
****
Offline Offline

Activity: 280
Merit: 257


bluemeanie


View Profile WWW
May 20, 2014, 11:16:33 PM
 #57


It's been discussed already, if someone were to try and bite off a significant chunk of NXT the price would multiply.  To get a stake capable of an attack?  Not even sure if that would even be possible at this point.

Except for the fact that I have demonstrated here that NXT is trivial to attack with only a small stake and a small amount of computational power.

you haven't demonstrated that.

-bm

Ok so I haven't produced a real world demonstration, but I have essentially given you the pseudocode. All I need to do now is get an NXT stake and figure out how to compile a custom NXT client. No one has given me any reason why this wouldn't work.

maybe no one cares what your theories are?   Grin

it's a conclusion that is perfectly plausible but perhaps not so rewarding personally.

-bm

Just who IS bluemeanie?    On NXTautoDAC and a Million Stolen NXT

feel like your voice isn't being heard? PM me.   |   stole 1M NXT?
telepatheic
Jr. Member
*
Offline Offline

Activity: 56
Merit: 1


View Profile
May 20, 2014, 11:24:54 PM
 #58

maybe no one cares what your theories are?   Grin

it's a conclusion that is perfectly plausible but perhaps not so rewarding personally.

Maybe not! I'm interested in these attack models, whether it is NXT, mastercoin, bitcoin, maidsafe, bytecoin, peercoin... They all have very interesting attack surfaces (and many bugs). Other people can engage in the debate or not, it is up to them.

I would love it if more people actually looked into the implementations of these protocols, actually went bug hunting and tried to understand how it all works at the byte level. All of these coins are experimental. The idea is to break them and then fix them not to try to defend the existing code at all cost.
bluemeanie1
Sr. Member
****
Offline Offline

Activity: 280
Merit: 257


bluemeanie


View Profile WWW
May 20, 2014, 11:26:42 PM
 #59

maybe no one cares what your theories are?   Grin

it's a conclusion that is perfectly plausible but perhaps not so rewarding personally.

Maybe not! I'm interested in these attack models, whether it is NXT, mastercoin, bitcoin, maidsafe, bytecoin, peercoin... They all have very interesting attack surfaces (and many bugs). Other people can engage in the debate or not, it is up to them.

I would love it if more people actually looked into the implementations of these protocols, actually went bug hunting and tried to understand how it all works at the byte level. All of these coins are experimental. The idea is to break them and then fix them not to try to defend the existing code at all cost.

I did link the core NXT devs your postings and they thought they were LOL.

have a nice day!



-bm

Just who IS bluemeanie?    On NXTautoDAC and a Million Stolen NXT

feel like your voice isn't being heard? PM me.   |   stole 1M NXT?
bluemeanie1
Sr. Member
****
Offline Offline

Activity: 280
Merit: 257


bluemeanie


View Profile WWW
May 21, 2014, 12:02:07 AM
 #60


Quote
It is generally accepted that the market value of a commodity and the cost to produce that commodity tend to the same price in a free and competitive market. 

these are not concrete economics ideas, sorry.  This is the gold bug mindset and as the story has been told countless times this ignores the effects of credit, innovation and imho the entire premise of civilization itself.  Is man just a resource extractor?


What does this have to do with gold other than gold being an example of a commodity?  The cost to produce corn tends to the market price of corn, the cost to produce aluminum tends to the market price for aluminum, the cost to produce canola oil tends to the market price for canola oil.  The reason is very simple: if the market price for corn is significantly greater than the cost of production, farmers will plant more corn and less potatoes to earn a greater profit. This action will increase the supply of corn, which will tend to reduce its market price, and eventually bring the profit margin for growing corn in line with other produce that could be produced on the same farm.   

Peter,  is the market for microchips related to the cost of sand?

after all microchips are made out of silicon right?

do you have any more pulp fiction economics to share with us?


I'm not sure if you are being intentionally obtuse, or if you actually don't understand:

The market value of a commodity microchips and the cost to produce that commodity those microchips tend to the same price in a free and competitive market.

If the market price for "Commodity Microchip A" was vastly greater than its cost of production, then manufacturers would make more Microchip As and less Microchip Bs to earn a greater profit.  This free-market dynamic acts to bring the cost of production inline with the market price of that commodity. 

This is only true for commodities, however.  Take something like 20nm SHA256 ASICs for bitcoin mining.  The market price for these ASICs is much higher than the cost of production because this technology is not yet commoditized.  Only certain companies have the IP/trade-secrets for this technology, which means they can earn very high profits (far above the cost of production) since they have few competitors.  But what we will see over the next few years is more players developing their own SHA256 ASICs to compete for the large profit margins currently enjoyed by KNCMiner, etc.  This will drive the price of these ASICs down until they eventually become a commodity like a standard 12-bit analog-to-digital converter.


what is a 'commodity' Peter?

'co' = together

'mod' = measure

something that is communally measured.  For instance we made milk a *commodity* when we developed ways to measure the quality and quantity of milk.

I dont know it seems you're obsessed somehow with the price of hashing gear.  I don't really care because I think it's irrelevant and most people who seem to be present in these arguments are those who are invested in it somehow.  I figure that sometime in the past few years someone capitalized on both Bitcoin, the engineering muscle that defines/develops/justifies it, and the gigahash hardware.  Such an investment would be very valuable because they could offer under-the-table insurance for the chain(and it *just so happens* that such offerings popped up around this time).  Actually not so long ago there was an IBM exec hovering around the Color Coin project, presumably he saw an opportunity for hardware sales.

things like NXT threaten such a profit nexus because it invalidates their entire premise of worth.

it's very sad. Sad Sad Sad

-bm

Just who IS bluemeanie?    On NXTautoDAC and a Million Stolen NXT

feel like your voice isn't being heard? PM me.   |   stole 1M NXT?
Pages: « 1 2 [3] 4 »  All
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!