https://bitcointalk.org/index.php?topic=1349944.0

It is possible to do this at least with LTC to BTC today (but not one person seems to have shown interest in it so I've ceased working on it for now).

My guess is that people are expecting a decentralised platform that currently isn't feasible (i.e. they expect "day-trading" suitable speed or that they somehow think this can work with fiat - which is not going to happen).

Maybe when LN is implemented it might be at least possible to do crypto-to-crypto trading fast (so there is perhaps some hope down the track for the concept).

My understanding is that the Chinese miners are now not going to back Classic so it is thankfully REKT and therefore I feel no need to continue this (now thankfully unnecessary) topic.

I am going to go back to focusing on working in this field rather than trying to herd cats.

Would you like me to create a signed message using the 1ciyam address to prove it?

(and if that isn't enough then perhaps use my GPG sig as well?)

If you are an exchange and you decide to "remain open" then you are gambling (i.e. if you end up on the wrong fork and you have paid out fiat to people then you have just lost that fiat and gained nothing of any value).

I don't think that exchanges are going to be so brave as to want to gamble.

Bitcoin is not a political thing (so trying to compare it to such things is just plain silly) - it is the fact that Gavin is trying to treat it that way now that is the issue.

Miners do not want to face risks of not being able to turn their BTC into "fiat" because it actually costs them a lot in electricity bills to keep mining (which secures the network).

If we end up with too much doubt as to what is going to occur then exchanges will stop allowing fiat to be exchanged and miners will be forced to start shutting down mines (due to the expense of running them).

In order to make the UI "nicer" I am using "unix time stamps" rather than block numbers - so with a bit of work maybe I could convert the time stamp back into a block number and improve the speed that this could be used at (as Bitcoin time stamps can be out by as much as two hours).

But even so - you are going to need to see at least 1 confirmation from the "initiator" (to the P2SH address they need to send to) before you are going to bother (as the "responder") to send to the complimenting P2SH address (as you wouldn't lock up your funds if you haven't seen that).

And presumably the "initiator" won't reveal their secret until at least one confirmation (keeping in mind that 1 confirmation is still somewhat risky).

So the likelihood of an ACCT taking place in less than an hour is not that much (so this is not a solution for day traders).

Note that I didn't use the "get" calls or "dumprivkey" as for the purposes of simplifying the test script I used hard-coded private keys (but in a "live version" I would be basically doing just that).

Yup - I found that for the 4 wallets the simplest approach is to use "-regtest" (as you can generate the coins instantly).

I also use pretty much the same commands you list - but the construction and signing is done with my own software (I was a little disappointed that Bitcoin couldn't do at least the signing but it is a very unusual looking raw tx so perhaps not that surprising).

The CIYAM script I am using for testing this is here: https://github.com/ciyam/ciyam/blob/master/src/bct_acct_test.cin
(some of it is probably a bit mysterious but you can see the Bitcoin RPC calls fairly clearly).

Over two hours and not a single post (apart from my own).

As I suspected - there really isn't any interest at all in a decentralised exchange (using ACCT with CLTV at least).

(thankfully I've saved myself from wasted development time)

There is a version number but your point is?

(it seems you have very little understanding about Bitcoin but think that you should create posts recommending how it should be changed)

It is a bit like someone who doesn't even know what a "piston" is advising people how to "hot rod" their cars. There is plenty of online information about Bitcoin for you to research if you actually want to know how it works.

(but until you've gone through the effort to actually understand it please stop trying to offer your suggestions about how it should be changed)

The reason for that is just up to how it is implemented and yes it could be possible to get rid of that via scripts, etc.

Unfortunately the ACCT/CLTV Bitcoin tx is not something that Bitcoin (or Litecoin) can handle with its raw tx support (so it can't create the tx and nor can it sign it).


I think that is simply the nature of the beast (so it will never be attractive for "day traders").

I think this is a nicely put together visualisation (my only criticism is that it may have actually erred slightly on the side of being too technical but I guess that depends upon the target audience).

Some other points to note are:

- no fees for the usage of the system (beyond the tx fees that you'll need to pay on each blockchain)

- no private keys or secrets to be held in the online application (so no risk about losing anything to a 3rd party)

- method of transfer is ACCT using CLTV (https://bitcointalk.org/index.php?topic=1340621.0 for reference)

Firstly note that I will be deleting posts to do with "block size" (there are plenty of other topics where you can post about that) and any pointless posts.

If you are really interested in using a decentralised BTC to LTC exchange (let's assume just those two for now) then to register your interest you'll need to provide a Bitcoin signed message for an address that has at least 1 BTC or 1000 LTC (other posts that add nothing will be deleted).

I will be checking that said messages are valid (so any that aren't will also be deleted).

Relevant questions about said decentralised exchange are welcome but not silly ones (so don't post "Can it do XYZ coin also?" and just assume that the answer to that is no for now).

The reason I am trying to get a feel for whether or not there is any actual support for this (by people who actually have BTC and/or LTC) is to avoid wasting time on development if there isn't any real interest (I suspect that there is actually very little interest).

Some points to note about the likely way that this will be implemented (assuming there was enough interest):

- there will be a web application (hosted by ciyam.org) for finding and locking into trades (and doing much of the tricky stuff necessary for ACCT using CLTV)

- there will be a special wallet that you will need to install (most likely in a VM image for added security and simplicity of installation) that uses a web UI (via a locally run web server)

- trades would not be realistically able to be performed faster than one every two hours (due to the inaccuracy of time stamps in Bitcoin and its clones and due to the fact you'd want to have at least one confirmation before proceeding at each step)

The problem with "Proof Of Stake" is what is called the "Nothing At Stake" (or NAS) problem (you might want to do some research into that).

So far no real solution to this problem has been found so any blockchain secured by POS would be less secure than the current POW implementation.

The idea of having POS side-chains does make sense and likely will be occurring.

Eventually the block reward will dwindle to zero and then the only BTC that miners will make are the fees.

The next halving is expected in July this year (block reward down to 12.5 BTC) so I think naturally we'll be seeing "more efficient" usage of blocks even then.

Having empty blocks (i.e. blocks that only reward the miner with the block reward) has been an issue in the past but as the reward dwindles and fees become more important there should be a natural incentive to include as many txs as possible (this is a key part of the original design).

No - you can't do this - the person who is sending you (being the exchange) has the private key for those UTXOs (not you the exchange).

So it is within their potential ability to "double-spend" (not you the exchange).

Clear yet?

(why do I get the feeling you don't even know what UTXOs are)

It isn't up to the core devs at all (remember it is about CPUs or more specifically ASICS voting?).

Quite likely the core devs will decide to simply accept 2MB if the fork risk is looking very likely (as they are not the ones trying to wreck things here) but I would pretty much guarantee that with in a month or two after that Gavin will be back with something that has 4MB (with another hard-fork threat).

I am not talking about "faking hashrate" I am talking about a planned attack with real hashrate.

The motivation for such an attack would be financial (i.e. spend your BTC twice or convert it to fiat twice if exchanges end up on different forks).