Cybnate, you are going to use a Pybot? not a Nubot?
Why? and what is the difference?
i just read the reason in Nagalim’s draft
I like the plurality of the drafts
Cybnate, you are going to use a Pybot? not a Nubot?
PyBot is used by the ALP software, NuBot is developed by Nu. NuBot is currently being developed to work on future ALP software and is the way to go eventually. NuBot has also more configuration options. But for now PyBot can take this function with minimal effort. I’m used to it and know how it works and it’s tricks/bugs.
As LiquidBits operator it took me 10 minutes to make the gateway work today. Technically I can start the gateway today with minimal effort. Just need to reveal the addresses.
Two things. Firstly, pybot is a supee innefficient way to do this because you’re creating a client-server communication stream that is simply unneccessary if you arent using ALP rewards. Still, you’re right that itll get the job done, it’s just definately not a long term solution.
As far as T3 custodians, a T3 custodian is not allowed to send money until they recieve payment for that money. How does the gateway pay in advance for the funds being sent? It breaches at least the contract im operating under for me to send money without proper advance payment recorded on the blockchain.
I was thinking a protocol could be worked out so that gateway operators could use proceeds to buy liquidity from T2 custodians. recycling funds among gateways and T3 could be much quicker than via FLOT.
Yes, if gateways are looking to rebalance they can use T3 custodians to get a fair price. There would need to be some statement about the markup % used by the T3 custodian or either the custodian or Nu will get taken advantage of. Also, the gateway operator and the T3 custodian should be non-colluding actors, but this requirement is actually fairly weak. (i.e. even if they are both malicious and colluding they can’t really do a lot of harm to Nu without breaking contract)
Are we still back to discussing gateways?
Or is depositing to one side of a dual side NuBot meant, but “gateway” said?
The former gateways were one-way streets, diodes, semipermeable membranes.
Dual side bots are better, because through a feedback loop gateways create problems:
The dual side NuBots eased the situation at Poloniex, don’t you agree?
will you please prepare it for voting?
I’m still breeding on the T3 issue.
@nagalim, what would be a reasonable markup from your point of view. I suppose I’ll need to add that into my proposal as a cost factor to the Shareholders which I would need to be compensated for acting in the T1/T2 custodian role.
I’d say 0.3%. Of course, 0.1% goes to Nu by virtue of the T3 contract, then the custodian gets 0.2% for doing the manual trade. On a $5,000 rebalance, that would give the t3 custodian 10 nbt.
In case someone wonders, I’m still experimenting with stable settings for PyBot. I think I’m close now but will have them running for another couple of days to be sure. They can place new orders completely unattended, and I would just return ‘proceeds’ e.g. every week or what FLOT thinks would work best.
Thinking of offering a dual-side gateway on both Poloniex and Bittrex. Could also do CCEDK, if anyone would support that.
Looks reasonable to me. Any reason for using Pybot instead of Nubot here?
Main reason is the ease of use. It wouldn’t add any work to run the bots as they are running off LiquidBits which I need to maintain anyway. Setting up a NuBot would create additional work for me. So using PyBot enables me to do a relatively good offer which only needs to cover the transfer of funds once in a while to FLOT.
But you can not take advantage of the parametric order book fearure with Pybot…
Gives shareholders some flexibility, in case for example an exploit is found in the nubot software.
No, that is a drawback
and my earlier reply regarding ease of maintenance for me and therefore low cost makes it an attractive proposition for Shareholders I believe.
After a period of testing time has come to publish the final proposal and submit it for voting:
Full proposal update in OP and is published on Daology: https://daology.org/proposals/9703bd96f4cc6754cf6b98431315a9133bb564cf
Daology Hash: 21cc7b3ee02aba45e464a81e75e32784c2827302 (Assistant hash is still not working)
B7jeGVDdsd4eE5ZGGmv8xXn2QgkS9nAPUC , 100
Please add the above to your client if you like to support this.
Looks reasonable to me and we need to diversify.
100 NBT is a considerable cost reduction compared to an ALP’s cost.
It is important to try and see if such Dual-side bots can achieve the same role as ALPs for a much cheaper job.
polo and bittrex are not the worst exchange out there.
The plan sounds reasonable.
Adding this to my data feed
This proposal has passed. Thanks for your confidence.
There are currently two dual-side PyBots running on both Poloniex and Bittrex. I’ve already provided the same addresses in the forum for use by FLOT. They have minor amounts in them, and you could have followed their behaviour in the last week or so. I will withdraw the placeholder amounts once FLOT transfers funds into the bots.
Here is a summary of the proposal:
Deposit addresses for dual side bots:
Both bots have an offset of 0.007 (0.35% on either side). Poloniex have been set at a maximum of 2000NBT on sell. Bittrex is set on 1000NBT. I’m fine to tweak these parameters when this is beneficial to Nu and on FLOT’s request.
I still have a maximum withdrawal/deposit limit of 2000NBT / day for Poloniex. I’m good for transfers back to FLOT on a weekly bases preferably during the weekend (UTC+12) or for emergencies. I will report on a weekly bases once the bots are funded. At this stage I appreciate to have the bots funded with no more than 15k together. I believe this reflects best the current security measures I have in place and the way the operation being managed and encourages decentralisation of the bots.