CoinSpark released

3 views
Skip to first unread message

Gideon Greenspan

unread,
Sep 30, 2014, 1:12:22 PM9/30/14
to bitc...@googlegroups.com
Hi all,

Just a heads up that we've released the full CoinSpark solution, allowing web-backed assets to be traded over the bitcoin blockchain, and held in a lightweight desktop wallet that's available for Windows, Mac and Linux. Contracts are hosted on your own website and notarised on the blockchain.

CoinSpark is in the vein of colored coins but there are some key differences which enable low-cost transactions, lightweight wallets and scalability. We use bitcoin transactions with OP_RETURN metadata to transfer assets and don't color coins directly. Each asset travels completely independently over the blockchain, with multiple assets being able to sit in a single transaction output. And assets are auto-discoverable via a single CoinSpark address.

The site has an online asset creation form, a lightweight (SPV) wallet based on a fork of MultiBit, an open source asset tracking server and comprehensive libraries in 5 (well, 4 but soon 5!) languages with extensive developer documentation. Any and all feedback is welcome.


Thanks and good luck!

Gideon

Michael Sullivan

unread,
Sep 30, 2014, 1:33:47 PM9/30/14
to bitc...@googlegroups.com
Wow this is very cool. 
I had thought of inverting the CC asset authenticity by bring the domain names more front and center and it seems that this is just such an implementation as I had envisioned. 

Going to delve in a little.

Great work,
Sull
--
You received this message because you are subscribed to the Google Groups "Colored Coins" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bitcoinX+u...@googlegroups.com.
To post to this group, send email to bitc...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Tim Swanson

unread,
Sep 30, 2014, 2:18:36 PM9/30/14
to bitc...@googlegroups.com
Very neat, congrats on the launch.

--
You received this message because you are subscribed to the Google Groups "Colored Coins" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bitcoinX+u...@googlegroups.com.
To post to this group, send email to bitc...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Ricardo Filipe

unread,
Sep 30, 2014, 2:40:15 PM9/30/14
to bitc...@googlegroups.com
So if i understand correctly, you require the asset issuer to have
tracker servers in order to prevent double spending of assets? instead
of letting the amount of satoshi's transferred do that for itself...
I am missing the advantages of this approach, when satoshis are
infinitely devisable, if you argue for scalability and lower cost.

Gideon Greenspan

unread,
Sep 30, 2014, 2:50:57 PM9/30/14
to bitc...@googlegroups.com
Hi Ricardo,

No, double spending is prevented by the bitcoin transaction model, just like regular colored coins.

The tracking servers are there purely as a computational aide for lightweight wallets, so they can find out how much of the asset is sitting in particular transaction outputs, without having to store the blockchain or retrieve a large number of transactions. It's a way to solve the problem that all colored coins schemes (and MasterCoin / Counterparty) have grappled with - that the bitcoin network itself cannot confirm asset quantities, so users are either forced into web wallets, or have to do a lot of heavy processing on the client side.

Thanks,

Gideon

Flavien

unread,
Sep 30, 2014, 3:22:11 PM9/30/14
to bitc...@googlegroups.com
Congrats on the launch Gideon.

I see that with your scheme, backtracking isn't possible (or you always have to backtrack the whole blockchain back to block 0), and this is why issuers are required to run a tracking server. The thing I have a hard time with is that if the issuer is required to run a web service on their servers to help clients finding out their own balances, why do they even need the blockchain? They could run a centralized system with a REST interface to allow people to transfer quantities (essentially what open transactions is doing), in that case, they have faster and cheaper transactions than by using a blockchain, with the same constraints (i.e. operating a server).

> I had thought of inverting the CC asset authenticity by bring the domain names more front and center and it seems that this is just such an implementation as I had envisioned.

Ricardo Filipe

unread,
Sep 30, 2014, 3:43:26 PM9/30/14
to bitc...@googlegroups.com
2014-09-30 19:50 GMT+01:00 Gideon Greenspan <goo...@sigsoftware.com>:
> Hi Ricardo,
>
> No, double spending is prevented by the bitcoin transaction model, just like
> regular colored coins.
>

the thing is regular colored coins use satoshis to represent the
amount of the asset... would you mind explaining to me how blockchain
transaction validation is sufficient to prevent double spending
amounts of an asset represented by the OP_RETURN metadata?

Thanks

Flavien

unread,
Sep 30, 2014, 3:50:23 PM9/30/14
to bitc...@googlegroups.com
Ricardo: the satoshi amount has nothing to do with double spend prevention. Double spend prevention is binary: either an output is spent or it's not. All you need in order to benefit from it is that assets must be tied to outputs, then you are guaranteed that assets stored on a particular output can't be spent twice. The quantity of that asset is irrelevant.

Ricardo Filipe

unread,
Sep 30, 2014, 3:52:34 PM9/30/14
to bitc...@googlegroups.com
got it, thank you flavien. i was tying double-spending to amounts
instead of outputs, that was my misunderstanding.

Gideon Greenspan

unread,
Sep 30, 2014, 11:24:12 PM9/30/14
to bitc...@googlegroups.com
Hi Flavien,

Thanks for your response.

Backtracking is certainly possible in theory, it's not just something that we're building the product around. The reason is that backtracking is not compatible with making a system like this simultaneously scalable and usable. If a new asset comes in to my wallet and it's had a large number of previous transactions, my wallet is going to have to work extremely hard to verify the quantity of that asset. That's going to consume a lot of time, and also a lot of bandwidth if I'm not storing the blockchain, which I don't think end users can be expected to do.

So, yes, an asset tracking server is required. I think it's the right architectural model, just like the web has hosting services. But the issuer is *not* "required to run a web service on their servers" - we provide free asset tracking servers with 3-way geographic redundancy that track all assets, and they are used by default. The point is that the issuer *can* run their own server or servers if they like, so they're completely independent of us, and we provide an open source solution for that. They can use ours and theirs together for extra redundancy. The address of the server/s can be changed at any time on the asset web page.

The reason to still use the blockchain is that it provides all the benefits of bitcoin - objective proof of ownership of the asset, the possibility of direct peer-to-peer exchange, the issuer being unable to take your asset away, or choose who you can transfer it to, the issuer not having to run their own balance-tracking system. Basically my answer to this question is exactly the same one as you would give, or indeed the same as MasterCoin / Counterparty. The tracking server is simply a computational aid for end-user wallets and it cannot change the 'truth' about asset ownership - that is defined by the blockchain and the asset contract, and can be verified/audited by any third party, completely independently of the issuer, and let an asset holder taken them to court.

Anyway I hope this clears things up a little. BTW backtracking would only have to take place to the genesis event, since the block number of that is encoded in the asset reference, which is how assets are referred to in the scheme: http://coinspark.org/developers/asset-references/

Best regards,

Gideon

all...@gmail.com

unread,
Oct 14, 2014, 4:56:52 PM10/14/14
to bitc...@googlegroups.com
Is not installing on MAC

Gideon Greenspan

unread,
Oct 15, 2014, 10:39:38 PM10/15/14
to bitc...@googlegroups.com, all...@gmail.com
Hi. Can you please get in touch with me at gideon AT coinsciences DOT com and let me know exactly what problem you're seeing. We'll be happy to diagnose and fix the problem.

Gideon


On Tuesday, October 14, 2014 11:56:52 PM UTC+3, all...@gmail.com wrote:
Is not installing on MAC

Kawal Grover

unread,
Oct 31, 2014, 11:48:35 AM10/31/14
to bitc...@googlegroups.com
Wow. I"m glad I checked the threads on this group. This is AWESOME!

Congratulations, and I will definitely be checking out coinspark.
Reply all
Reply to author
Forward
0 new messages