Topic: [GLBSE] Bitdust - Make your bitdust work for you (Read 3508 times)

I'd recommend to follow up on my questions in https://bitcointalksearch.org/topic/reading-the-block-chain-with-a-library-77650 - It's really uncanny, as I planned to do something similar ("bitdust" was even a ticker symbol I considered) as soon as the dividend API would be released. Before that it wouldn't really make sense and I hate poor and/or rushed execution...

Anyways, you can calculate for each second since the blockchain started how many BTC 1 MH/s (or 1 KH/s) would have generated... then just sum up accordingly for hours or days and pay as dividend. All you need is either a running bitcoind + asking for the block headers via RPC or running bitcoind + armory and having it already nicely packaged. Libbitcoin might also work, I haven't tried that one yet.

If I only read "some day I'll do the work to pay out correctly" I start looking like this -->
Even if it's tiny amounts of money, you're still resposible for it and should put in some effort (it's not even hard, my scripts were finished in under 2 hours) instead of looking for a quick payment.

To avoid over/underpayments due to rounding, couldn't you pay out when the amount is a whole number of satoshis, rather than paying out per day/other time-based scale? Is this even feasible?

Turns out the GLBSE API does not support paying out dividends to shares at the moment but Nefario has assured me that feature will be available soon. I've got everything situated so I'm going to let that script start running and improve it over the weekend. I'll stick a preliminary contract up in this thread to solicit some feedback.


I appreciate your incredibly helpful comment.

I'll just stick to browser mining with my iPhone

Alright, I think I have a preliminary solution for the script that will run this thing. Due to not wanting to scrape the blockchain info(I'm azy) to see exactly when a difficulty change occurs, I will instead be calculating the amount of BTC earned every hour with the difficulty listed on blockexplorer.com/q/getdifficulty. Working now to support polling bitcoind from my p2pool server to avoid depending on blockexplorer.com. Someday I'll do the work to calculate precisely on a difficulty change.

I will keep a running "balance" per share which will store the total amount of BTC generated per share to the limits of what python provides. I will payout from this balance per share to each share on GLBSE at the granularity bitcoin supports which is 1 millionth of a BTC.

So, between payouts, each share will accumulate a few millionths of a BTC that will be rolled over to the next payout. The downside of this is that if you buy-in and sell between one of these periods you could miss out on this bonus payout.

I now need to see if GLBSE's API supports paying out shares from the issuers GLBSE balance. With that last piece in place it'll then be possible to just have this script run on a cron job every 3600 seconds and payout every 86400 seconds.

Well, if you go for 1Hh/s it's fine but then you have to be careful about payouts rounding ... and it would get even worse if the difficulty would go up.

Well there are some bonds/shares smaller than 1Mh/s as mine is ... but they are still very big compared to 10Kh/s or even 1Kh/s. My point for my offering was to have enough liquidity with relatively small offering ... but you brought it level higher. :-)

Excellent point, the 10 KH/s was pretty arbitrary. 1 KH/s should be fine but I'll have to check the math on the rounding of payments. We're going to end up with something like a million share IPO 


True, if blazr releases their 0.15 MH/s bonds it would be a superior offer for individuals looking for a long-term mining bond. My goal however is not to run a traditional mining company but rather provide a service for the gap between bitdust and the 1+ MH/s bonds available to users. I do NOT want to divert my full 16 GH/s(another machine came online  ) towards this venture. Think of it more as a temporary solution for your bit dust rather than a permanent solution.

I plan on daily dividends. I need to script up the auto-payment and see how much overpayment would occur with the rounding.

The idea is interesting ... the sub 1Mhash/s bonds are rare ...
If you do 100% PPS or more it will be attractive ...
otherwise holder of your bonds would need to check if they have enough your shares for 1 "better" larger bond to not lose on lower %PPS and that's probably tedious work
are you planning on paying daily dividends? how are you going to round? ... as that might be already important in your case
anyway if blazr puts out the blazrbond it would be superior offer ... so consider sticking at 100% PPS

I like the above idea -- even more liquidity.

Good approach to a mining bond though. Since this will be the only option for small investors in other assets until someone clones it, I'm sure you'll do well.

IMO the price should be 0.0003 (for 1 kh/s) rather than 0.003 BTC. More liquidity; put those hundredths of a bitcent to work!

Anyways nice idea!

Thanks for the discussion everyone. In the next day or two I'll sort out all the numbers and contract so I can go about organizing the IPO for this experiment.

Sending pretty much any amount to the faucet would be seen as a "loss" by many and probably work well to discourage long term holding compared to other mining stocks. Sending it specifically to the fountain would be a great way to give back. Maybe even give it away each month to a different bitcoin charity?

The feature GLBSE is missing that would really help this kind of venture out is being able to place buy orders to execute once sufficient funds are available. This would eliminate the need to watch GLBSE like a hawk to take advantage, and then you could come in less frequently to sell shares and buy more expensive ones or whatever else you'd like to.

That is a very interesting idea. Bitdust could donate a portion of the income to https://freebitcoins.appspot.com/ and other various pro-bitcoin establishments. I like it, will have to figure out the various percentage of income to divert to pay for power and such.

What about sending like 25% to the faucet?

Any thoughts on the discouragement of holding these bonds for extended periods of time? Should I not worry about this? Or find another way to discourage this?

I would also like to not run this operation at a loss, so the difference between the mining PPS and the payout PPS would help pay the powerbill. Ideally these will be running on my Enterpoint Cairnsmore quads(the first of which arrived today) so the powerbill should be manageable($4/month).

I like it. I personally feel there are too many of the ol' 1MH/s securities on GLBSE. I like the idea of 1% of 1MH/s also, it's very straight forward.

I think investment options are good.  I like this idea.

I made a mistake in the initial calculations. Easier to script/debug/run versus plugging things in to a calculator.

edit: In the end this would let people min/max their GLBSE balances versus letting their dust do nothing. I understand in the end we are talking trivial fractions of a USD. It would be roughly equivalent to purchasing fractions of a normal mining share for those people who check the GLBSE like a hawk.

I really doubt if anyone will use it the way you describe.
Personally I have invested around 24 BTC on glbse (planning on buying more bitcoin) and consider myself as a small investor.
At the moment I have around 0.11 BTC in my balance.
For 0.003 BTC each I can buy 36 of your shares, they will give me 0.000004574320992 BTC a day extra, on a year it is 0.00166962716208
0.00166962716208 / 24 * 100 = 0,006956779842% extra profit.

/edit
Your bitdust profit/day is wrong, should be 0.0000063532235831
Will give me 0.0002287160489916 a day, 0.083481357881934 a year.
0.083481357881934 / 24 * 100 = 0.347838991174725% extra profit for me.

For smaller investors it will be better, someone with a total of 2.4 BTC invested will have 3.478% extra profit.

I can't take the time but I have this gut feeling that you complain about non-100% PPS bonds so I'll take your complaining with a grain of salt. Odd numbers don't matter to scripts, only humans.

Regarding discouraging individuals from holding these bonds for long durations, my goal here is not to create another mining company that directly competes with the more efficient operations out there such as YABMC. Instead, my goal is to compliment their services by giving their users a place to store their bitdust while waiting to generate enough to purchase something useful. Liquidity should be provided by users frequently buying in and selling out for short periods of time.

90% PPS, 110% PPS, 1% PPS, who cares?!

You could just offer 1% PPS of 1 MH/s instead of 100% PPS of 10 KH/s...

Odd numbers just make it more difficult for investors to calculate how much the "standard" of 1 MH/s would cost.
Also your calculations seem weird: Why do 1% of something generate completely different amounts of BTC and not just 0.01 * [revenue of 1 MH/s]?!

Also I don't get why people should be discouraged to hold 100 bitdust shares instead of 1 YABMC share for example. It would be even better, since there might be more liquidity.

Bitdust is a growing problem on the GLBSE.* Those fractions of a Bitcoin that are paid up as dividends from mining stocks/bonds and other investments just sit, gathering dust, instead of earning you more bitdust.

I propose to create a sub-1MH/s mining bond with the purpose of permitting owners of normal sized investment stocks/bonds to generate a small profit on their dividend bit dust. Below is a sample concept that I may move forward with if there is interest.

Normal mining bond.
1 share = 1.0 MH/s
1 MH/s/day generates 0.000012706447166.  BTC/day @ difficult = 1,583,178, blockreward=50 BTC
price/share = ~0.3 BTC.

Bitdust
1 share = 0.01 MH/s = 10 KH/s.
10 KH/s/day generates 0.000000127064472 BTC/day @ difficult = 1,583,178, blockreward=50 BTC
price/share = ~0.003 BTC

I am willing to put this together for fun. These bonds would not pay out at 100% PPS in an attempt to discourage individuals from holding these bonds for extended durations. I think something in the 80-90% PPS range will encourage people to pick up bigger bonds when they acquire enough bitdust to buy 1 MH/s or whatever is the entry-level bond value at the time.

I'll go about automating payments so that I can pay out on a daily basis. I have yet to decide on the amount of hashing power I would throw at this but my farm is currently averaging around 13-14 GH/s. I would doubt more than a few hundred MH/s of these 10 KH/s shares could be absorbed by the GLBSE but that's just me pulling numbers out of my.. hat. The point is though that this is essentially a mining bond and I've got plenty of hardware to scale if desired.

Thoughts?

*- may or may not be true. Who cares?

edit1: Changed 1 MH/s/day generated from 0.00063532235831 to 0.000012706447166.