When Bitcoin crashes

To maintain buy and sell walls someone has to hold a significant number of bitcoins in the exchanges. I recall from somewhere that NBT walls are actually maintained against USD, right? Then how is it possible that in bter there are also NBT/BTC walls that reflect the correct price? When BTC crashes significantly then whoever runs the bots to maintain NBT/BTC walls at ~1$ will lose a lot of wealth. As a result, the walls shrink until the peg is lost. This puts NuBits in a position where it depends on BTC. How big of a threat is Bitcoin’s demise for NuNet?

1 Like

You have missed the fun

Yes I have indeed missed the fun. Anyway, I don’t like the idea of using some coinapult kind of service.

Here’s what I propose:
Custodian will open a NBT/BTC online shop. The custodian will never hold BTC. Instead, they will hold USD in Bitstamp, for example. Because the custodian has to be a trusted entity in the first place, we would have no problem trusting them to run their independent NBT/BTC online shop.

Selling nubits:

  1. Customer comes to custodian’s online shop
  2. Customer wishes to sell NBT.
  3. Customer sends 1000 NBT to custodian.
  4. Custodian’s bot immediately buys 1000 USD worth of BTC on Bitstamp.
  5. Custodian’s bot withdraws the BTC.
  6. Custodian’s bot forwards the bitcoins to customer.

Buying nubits (puts risk on customer):

  1. Customer comes to custodian’s online shop.
  2. Customer wishes to buy NBT.
  3. Customer sends 10 BTC to custodian’s bot.
  4. Custodian’s bot sends the BTC to Bitstamp.
  5. Custodian’s bot sells the 10 bitcoins for X USD.
  6. Custodian sends X nubits to customer.

You can think different risk distributions for buying nubits yourself. For example, by making use of bitpay and what not. My point is that we don’t actually need NBT to be traded on conventional exchanges. It’s rudimentary way of thinking. If you think that such an idea would be cool then I could develop the needed functionality for bots and online shop. Let me know what you think.

3 Likes

So you are basically describing some kind of custodian-run shapeshift service?

Yes.

i think is very nice idea

Thanks for the support. If enough people think it’s a good idea then perhaps I can find time to develop a robust proof of concept. The software should be free for anyone to use and modify so that eventually we would have thousands of nubit online shops/converters.

Both the contracting system and the shapeshift-like proposal have their advantages. But isn’t all this the decision of the custodian?

If the custodian accepts to take the risk in order to avoid contract fees with OKCoin etc and to provide instant liquidation opportunities, then she or he can use her or his own funds. In an uncertain market however, the custodian proposal could be formalized using one of the safer models you described. In the end the custodian with the most appealing offer to the shareholders will win.

Or is the plan to create one predefined way for custodians to provide liquidity that must not be altered?

I would not want to force anything on custodians. However, as a shareholder, I would much prefer to elect a custodian that uses the system I just described rather than a custodian who risks all the funds by holding something as volatile as Bitcoin.

I totally agree and creating a set of freely available tools that will make it easy for custodians to implement such a system is a great idea.

This is similar to an idea I proposed a few days ago, but seems much easier to implement! Great idea and I definitely support it.

I think exchanges are still excellent for NBT / USD exchange, but this system would be superior for NBT / cryptocurrency exchange. Particularly for cryptocurrencies with greater perceived risk.

1 Like

Superb! Here’s how I would implement it:

  1. There exists a central meeting place for custodian bots and customers which provides an API for both parties.
  2. Front end web interface uses the API to make an order to buy/sell nubits for bitcoins.
  3. Custodian bots poll the central database for new orders.
  4. When bots see new order that they could fill they will accept it.
  5. Central server only lets one custodian bot to accept the order, others will fail accepting the order (no race condition can ever happen).
  6. The custodian bot will be provided as free to use and modify software preconfigured to be easily integrated with bitstamp.
  7. Custodian is responsible for keeping their bot running 24/7 on their local computer.
  8. Custodian is responsible for making sure that their Bitstamp account always has enough dollars and that their NuBits wallet has enough nubits so that their bot is able to accept orders.
  9. Custodian is required to run Bitcoin-core and NuBits standard wallet that support the RPCs.
  10. Custodian’s system should be Linux.

The reason why I was so specific about the above requirements is that I already have developed most of the functionality needed for such a software solution. If I was to put something like that together it would be done pretty fast. I would make use of the infrastructure I have developed for http://cryptograffiti.info/paystamper/
(you can check the above service out to get a rough idea how this whole thing would work and function)

Your points are certainly nice, and they made me think:

What about providing buy-side liquidity through a future OpenBazaar version with a backend service?
The methods to save their funds could be very similar.

If the open-bazaar platform allows automated online shops then I would think it could be reasonable. I don’t know the specifics though because to my understanding open-bazaar is still in beta and so I haven’t paid much attention to it.

Hi Hyena - Nice idea :slight_smile: Not trying to steal your thunder, but i had dreamed up something similar and we had started the process of building it. Unfortunately one of the developers i employed recently decided to move on to other opportunities so it put a hamper on it for a while as we are busy with customer projects at the moment. I had registered the domain exchange2.nu for it actually… The other thing that strays me away from the project is AML / MSB implications. So though we had the project slated i still might have never went through with it.

Between the ideas of systematically hedging with derivatives and the kind of system you have outlined, i think we have some interesting options to combat custodian losses. It’s great to have options.

All the best to you if you decide to develop this, will be watching!

teek

It’s useable beta, but development seems to be quite active.
Afaik, APIs are planned.
Let’s see what happens. It would be an easy way to have a (at least partly) decentralized NuBits exchange.

It’s NBT <–> BTC exchange and does not involve USD directly. Thus I don’t AML will get in the way. Look at btc-e and other shady exchanges for example. how can they still operate?

unfortunately, this is probably an unsustainable situation… they operate because they are set up in such a way and such a jurisdiction that makes this possible (at least temporarily).

i don’t agree with it, but being that i live in north america i have little choice but to be weary of it.

It’s not unsustainable. The worst case scenario is that a custodian needs to comply with the KNC and AML laws, it’s not the end of the world. The online shop for nubits can still be achieved. I think people are too much afraid of the potential regulations. Such attitude does nothing good, only hurts the motivation of others.

Oh i’m not saying no one can do it or shouldn’t do it. I am definitely not trying to discourage anyone either. If you or anyone else feels they want to do it, go for it!

I am just saying that from what i know and have seen following everything cryptocurrency since 2011, the MSB problem is real, and i am unwilling (at least at the moment) to go through the red tape required to satisfy the government OR re-locate and structure in such a way to mitigate that concern.

That is not to say that no one else can. As a matter of fact i certainly hope they do and fully encourage them to. Though proceeding without acknowledging the issues exist is pretty well asking for trouble.