Topic: about alternate chains vs using bitcoin blockchain (Read 2246 times)

Yeah it is true that Open Transactions has not reached the optimisation stage yet.

It is not fast and has not yet even tried to be fast.

I had no intention though of trying to support high frequency trading, quite the opposite in fact.

The intent is rather that the scale factor markets be used, and I also have all assets denominated in integers which also is intended to favour large scale trades over trivial little trades. If you want to deal with less than whole bitcoins use some smaller value coin, if you want to buy one at a time then hopefully that small a scale retail scale kind of market is more expensive than if you buy things by the ten or hundred or thousand or million or whatever.

It was mostly intended for when nations / civilisations / corps etc wanted to do major international or intercorporation trades at scale, not for the thousands of customers who walk into individual branches to buy a few coins or notes to access directly themselves.

When a branch runs out of some coin or other it'd turn to the scale markets to pick itself up another 100,000 or 10,000 or 1,000 or whatever scale batch to re-stock, that kind of idea.

Or when international trade resulted in two nations each holding more of the other's currency than they liked to they'd do a big trade to get back to a more normal level.

You would actually need and want the overall market to be distributed, so that each broker's customers buy at that broker's server or bank of servers, that broker re-stocks from a server that deals in larger scale blocks of stock or coin or porkbellies or whatever commodity/share/bond/coin/etc, which in turn restocks from one that specialises in even larger scale blocks and so on.

Kind of like if you don't want to buy something by the fleetload go to a place that sells individual shiploads, if you want even punier lot sizes go to a place that deals in container-loads, if that is too rich for your blood find a place that sells by the pallet, etc.

If our server deals in shiploads, we don't want to waste our bandwidth etc etc on children walking in looking to buy a gram...

...Especially if that would result in one of our shiploads being short a gram...

-MarkM-

That's like saying: I have a webserver where I trade with 5 friends and it works well. At scale many things change. Moving 5$ around is not the same as investing live changing money in a stock market. OT makes the same claims as the others, that you can produce stocks and bonds out of thin air. There is a reason legal systems exist, they mostly work to the advantage of market participants. OT is better than master-/bitshares in that is not P2P, which for public market good can't work.

In terms of the impossibility of scalable P2P exchanges one has to think about the timestamp of each message (=transaction) in the system, and how messages are ordered on that basis. In bitcoin A makes a tx to B and C to D, and both tx are independent, so one does not have to enforce time order. the blockchain does not allow complete time order. People forget bitcoin starts with a timestamp server. The fact that time order of messages can be neglected to some degree is very important.

Open Transactions has been working for about a couple of years now, and I have yet to have any accounts in it go out of balance.

It works rather well in fact.

-MarkM-

not sure if that's an accusation or a threat?  either way it seems pretty lame.

and I'm not sure everyone views Mike Hearn as a stellar asset to Bitcoin.  Seems his idea of 'blacklisting' addresses was unanimously rejected.

Google is the center for all the internet spying, Snowden has shown whats going on- game is over.

"Don't be evil."  LOL.

The only one I'm aware of is ripplescam.org, written by TradeFortress (well, see for yourself how he ended up himself...).

Google is not directly involved in Ripple, it's Google Ventures (which are independent from Google itself) who invested in RippleLabs. They also invested in the Bitcoin exchange project "Buttercoin" by the way, also e.g. Mike Hearn, a Bitcoin core developer and a great asset to Bitcoin works at Google. If you do mistrust them, that's your personal opinion, not a universal fact.

Nice try of trying to derail this thread further ("aside from this"... instead of "Sorry, I was wrong"), I'm outta here to not bump your promotion of your website even more.

aside from this there were a number of other criticisms of Ripple floating around.

I mean it's related to Google, whom it is quite clear cannot be trusted with anything, certainly not to create a p2p money technology.

https://github.com/ripple <-- the server is called rippled, it is open source since September last year.

You can just ask RippleLabs about pending patents (I'm unaware of any - and releasing source code before potential patents are granted would also not have been a smart move by them...), the only thing that is protected as far as I know is the "Ripple" brand and logo.

I did.  the mods moved them to 'newbie' section, even though NO ONE has ever suggested anything theoretically wrong with it(quite the opposite).  I get hits every day to the whitepapers on the http://altchain.org web site.

I do disagree re. what is the best approach.  You can make code that performs some function, but this doesnt prove it is secure or economically viable.  There is at least one software system that attempts to take your money claiming to be distributed exchange that is anything but.  I want to avoid this perception as a scam, and there are many scams in this particular area.

I will follow up with you in email.

bluemanie, you should open up your own threads, instead of writing about your proposals in several other threads. Also it is obvious to me, that everyone who works on these P2P models has never worked much with orderbooks. because then you realize it does not make sense for the most important markets (stock market, bond market, fx market, commodities market). also, all these empty promises. first right the code, then tell people about what you've done. making wild claims isn't useful. same thing as bitshares & mastershares & OT & etherum. all talk, no code. because it just doesn't work.

there's a fairly detailed discussion on it here: http://www.wilmott.com/messageview.cfm?catid=10&threadid=95845&FTVAR_MSGDBTABLE=&STARTPAGE=4

Confidence Chains is much faster than bitcoin because there is no hashing.  The exchanges should run very fast in most cases, but latency characteristics depend on a number of factors(how the nodes live in the network, number of assets and transactions on the chain etc.).

yes, it will be open source.


ADDED: it's written in Java.  Crypto uses JCE so it's possible to swap in whatever crypto you want(bouncy/spongy castle).  Also possible to set your own EC curve, or even swap out ECC altogether(in favor of some other kind of DSA).  There seems to be much discussion about the security of NIST curves, and in order to avoid any liability regarding use of strong crypto, I leave that up to users.  for now I'm conforming to bitcoin crypto standards.  Also I use the base58 encoding for addresses. 

at one point exporting software in the USA with an EC curve that exceeded some parameter was actually illegal and considered arms export.

Can you explain this statement?   Doesn't Bitcoin already preserve the order of spending by making sure that there is a chain of transactions?   How does the congress of nodes enforce order in transactions?

Also, exchanges need to be lightning quick.  How do you do this on a peer 2 peer network where different peers see different transactions occurring?

Is this open source?

re. Color Coins, there is an not exactly an atmosphere of information accuracy on that list.  There's a clear tendency to misname things, make claims then retract them etc.  That's all I'll say officially because basically I don't have the resources to 'go to court' on this, I'm just working as an open source developer.

the ETA is very soon.  I have most of the basic stuff finished, just need to complete the Kernel which is the code that decides which blocks go where given the transactions the node knows about.  There is definitely more than one way to do this, and there are strategies for optimizing, perhaps even strategies for exploitation(not unlike Bitcoin).  When this is done we will have a Distributed Exchange.

What do I mean by Distributed Exchange?

here we mean a way to exchange assets that has no specific center.  It is comprised of N nodes, and if any given node is disabled(either by legal or technical means) the exchange is not shut down and users access to the ledger(their account) remains intact - similar but not identical to Bitcoin.  Secondly the ORDER of the transactions(such as POST and ACCEPT exchanges) is determined by this congress of nodes.  This sort of functionality is impossible with PoW systems.

Thanks for looking it over, I will announce when it's at point for early adopters ie. technical people.

-bm

Thanks a lot for the long explanation.  Can you refer to me what 'false claims' were made by the Colored Coin folks?

Also thanks for the 2nd opinion on OT.   I was about to invest some time on it,  but I am going to take a leap of faithhere and believe what you say.

I reviewed your confidence chain stuff and the decentralized exchange.  Looks sound, but I still have a few open questions.

However,  what is your ETA on this?  Is this open source?  I've been waiting on the Buttercoin stuff for 6 months with nothing to show. 

unless something has changed, you are only free to fork the CLIENT.  Secondly it is unclear what sorts of patent protections there are surrounding Ripple.

It's not a community project, it's a commercial project.

You can just fire up rippled and will immediately have 100 billion XRP at your disposal. Just like running bitcoind with a non-Satoshi block chain/genesis block however, no other server will accept these.

You are free to fork Ripple, the only requisite (similar to Bitcoin) is that you change the name and logo.

I will grant that there may be some perceived advantages to this.  But what this trust model ignores are the risks and long term costs of PoW systems.  I've already explained the drawbacks many times, they are in the record.  No one will be convinced of anything until they observe the effects themselves, so it's not very much worth spending time establishing these theoretical concepts over and over again.  I was supportive of Color Coins in the beginning, and I still like the basic concept- but the people involved with the project currently I dont associate with.  They've repeatedly made unreliable and false claims and there is no attempts to rectify this or even acknowledge it.  Thus you really have no idea what you're listening to with these people.  Note also that IBM is involved with the Color Coins model.  I think they see the potential to sell more hardware, but that isn't based on their own official statements.
 
just keep in mind that last time I explained why Open Transactions doesn't live up to it's claims I was attacked by Chris Odom.  I spend several months working with this character along with a number of other people.  I have the records to prove this.  AMOF, he *stole* the monetas.net domain from me.  This board just isn't an environment where you can get facts easily.

Given that, statements made about Open Transactions are largely untrue.  It is not decentralized.  It is a client-server system that uses crypto in a fairly straightforward way.  It also has an unintegrated Chaumian E-cash component that was added by way of Ben Lauries Lucre library.  This component does not work with the other parts of Open Transactions.  Mind you, OT has NO REAL WORLD USERS.  I don't take this project or Chris seriously.  Also Chaumian cash, while theoretically sound I don't feel is very relevant today.  The problem with it is that it's not decentralized, thus can be shut down.  OT exhumed Chaumian cash, made the suggestion that they changed or improved it in some way(he didnt), and presented it as a novel new idea.  I also sometimes wonder if David Chaum was involved with the creation of Bitcoin, but we can't know for sure.

Regarding Ripple,  It's also somewhat deceptive.  They made a few tricky moves regarding their open source status.  It's a commercial product and their business model was to sell and market XRPs(similar to Mastercoin).  At one point I asked: If Ripple is open source, why can't I just issue a new set of XRPs?  XRP2.0?  Soon afterwards the project seemed to have faded out of existence.  Ripple also has virtually NOTHING to do with the original ideas associated with this name.  It's all part of the Googleplex mega-complex, which I will never(and probably can never) be a part of due to my outspoken opinions on the surveillance machine they are running under the guise of 'entrepreneurship' and 'startups'.  Thanks to St. Snowden I can now say these things without being branded a mentally insane schizophrenic(which some people in these circles have already tried to do).

Confidence Chains does a lot of things.  Turns out that to support a decentralized ledger you need to build a few things that let you do lots of other fun stuff.  It supports many financial functions, have look at the link in my sig.  If you have an questions you can post them to the forum.  I'll answer them personally.  thx, -bm

Thanks for the comments guys.

@bluemeanie1

Yes, you make a good point regarding colored coins, you still have to trust the issuer. I haven't looked into confidence chains, will have a read!

@frictionlesscoin

Yep, people won't bother, as long the coin is secure. If alt-coins are beneficial to the maintainers of the network: ie, the miners then they'll be on board (or at least I think so). The donated based coin, creates much larger blocks in the future.

The problem that I want to solve is the creation of fairly distributed currencies with better security than the current alternative blockchain approach. If we have a million altcoins in the future with millions of blockchains, there's going to a very large long-tail of poorly secured coins. This is one such proposal. I'll think about it some more.

Of course most all colored coins require trusting the issuer!    The main advantage of using a decentralized block chain as opposed to the conventional way is that the colored coins can be exchanged in multiple exchanges and everyone can be assured that what I exchange in one is the same as the one exchange in another.  That is a very powerful use case!

Mining has almost nothing to do with colored coins.  Sure you can devise some kind of mining scheme,  but for what purpose?  To create another kind of trust free coin on top of the already trust free network?

Are you referring to Mastercoin? well that is a completely pointless exercise.   

I can see where you are going with the confidence chains and I do agree that a decentralized exchange would be better served on a network and protocol different from bitcoin.   However, can you explain to me how your proposal compares to Ripple and Open Transactions?

'Atomic swap' is over rated.  We already have crypto-coin exchanges that already swap between multiple block chains.  

From the perspective of the user,  they would care less if their currency rides on top of the bitcoin block chain or not.  All they care is that it is relatively secure,  with merged mine coins like iXcoin,  its 40% of the bitcoin has rate, so it is going to be good enough for almost all use cases.

In fact, I would argue that people will not feel to kindly about polluting the bitcoin block chain with alternative currencies.