IPO: Slowing the Whales

ddink7ddink7 Member Posts: 49
edited February 2014 in Ether Sale
Vitalik or Charles:

Would it be possible for you to implement a maximum number of BTC that can be sent from any single address (perhaps 10 BTC maximum per address)? This is similar to what XCP is doing. It would not stop a person from investing 1000 BTC, but it would take them longer since they would have to shuffle between 100 different wallets. That would give us small investors enough time to ensure we get a share.
Post edited by StephanTual on

Comments

  • Manfred_KarrerManfred_Karrer Member Posts: 17
    I think a better solution could be that the cap mechanism change to be more flexible for the case that the 30k cap is reached too soon from some investors buying in as monopolist. Something like: 30k cap will be activated after 3 days, so everybody has time to buy in. Or when the cap is reached the IPO will be stopped after a certain period (1-3 days).
  • CryptKeeperCryptKeeper Member Posts: 7
    It would be fair to offer the option of wire transfer for buying in the IPO, so even non-techy people can participate. Please think about it.
  • StephanTualStephanTual London, EnglandMember, Moderator Posts: 1,282 mod
    The team is aware of this potential issue and has postponed the foundraiser accordingly.

    In a reddit thread, Vitalik said:
    "We are seriously worried that if we offer a small cap then some large investor will simply gobble it up all at once, and then we'll basically have Ripple. Judging by the response we have gotten so far, we realized that it is entirely possible, even if unlikely, that some whale with 200000 BTC sitting around will decide that Ethereum just might be the right coin to throw 10% into, and we need to work around such a possibility. If no whales come onboard on such a scale, then we will have a small market cap to start off anyway."

  • Benjamin_BBenjamin_B Member Posts: 12
    Sensible. I'm considering putting in 10.
  • redwanredwan Member Posts: 2
    +1 on CryptKeeper's request! Could we consider a direct wire transfer option?
  • ddink7ddink7 Member Posts: 49
    The problem with benjy's idea is that many in the community would refuse to identify. I posted a hybrid approach at reddit...any thoughts?

    http://en.reddit.com/r/ethereum/comments/1wlgcy/vitalik_idea_for_fair_ipo_distribution/
  • djanashdjanash Member Posts: 2
    One idea is to use a window of time to buy-in (For example:Feb 1 to Feb 10) During this time all intended donations/orders AND the amounts and wallets to be held privately in a list. Then on Feb 11 all those amounts can then be sorted and then a mean, median and avg can be created. Using this info, one can randomly select X number of donations/orders in the average range and have the amount withdrawn automatically. Some % of X for each range + or - 1 standard deviation from the norm, and then even less for 2 standard deviations from the average etc. Rinse and repeat a week later for more fund raising. This will eliminate all the dust and whales.
    Any thoughts? I know this means that we will have to "trust' etherium, but why wouldn't you if you are about to invest money in it. Publish the results and notify each "winner" and destroy all "loser" bids.
    Please contact me privately if this interest you
  • JakeThePandaJakeThePanda Member Posts: 15
    I wish everyone would just shut up so they can take my money already. :!!
  • FreddyFenderFreddyFender Member Posts: 39
    Double your sentiment Jake!
  • redwanredwan Member Posts: 2
    The IP address of anyone trying to participate in the fund raising could be checked so that we could enforce a cap on the number of BTCs a single IP address could buy. This has some limitations obviously since one person could use different proxies to go beyond the cap but that would at least block a fair amount of people trying to cheat the system by purchasing ethers from different bitcoin addresses.
  • MKHAUSMKHAUS Member Posts: 2
    The fairest way to raise money in an IPO is to do it dutch auction style. If they intend to cap the amount of money they want to raise then it makes no sense to offer it at a fixed price per unit of ether. Instead bids are made on the ether dutch auction style so that those willing to accept the least amount of ether per bitcoin do so and all pay the same rate.

    What they could do is schedule many auctions over the course of 3 months. Say they want to raise $30 million. Then they could have 6 $5 million dollar dutch style auctions every 2 weeks. Over time the market would find the right price.

    As far as determining the amount mined as a function of the amount raised, it wouldn't be necessary. If the mining schedule was stated explicitly the value of the ether would be determined by the market, so it could mean more or less ether is distributed than expected based upon market expectations.
  • MKHAUSMKHAUS Member Posts: 2
    I should add that if $30 million is just an arbitrarily high amount intended to limit the % amount any single person could acquire then perhaps the amount to be raised should be set lower. So for example, if you can only imagine $2 million worth of bounty projects perhaps you only seek to raise $5 million. There's no need to sit on extra cash if you don't have use for it.

    I suppose you could schedule a series of IPO style fund raisers, like one every 6 months for 3 years, based upon the needs of the community. If a year from now it is clear that $20 million of bounties is needed for whatever purpose, then it would be possible to raise funding as it becomes necessary.

    The problem with this is obviously that this adds uncertainty with arbitrary funding amounts set in the future. It's hard to know how much you'll need in bounty money now for projects in the future, but if funding and ether supply isn't explicitly specified from the outset, confidence is undermined by the uncertainty.
  • StephanTualStephanTual London, EnglandMember, Moderator Posts: 1,282 mod
    moved to -> investement
  • illodinillodin Member Posts: 2 ✭✭
    Also very worried that big whales will buy it all. What might be good as already suggested is to have multiple rounds of fund raisings.

    For example the fist round the max amount to invest is 0.1 BTC and lasts only 1 day. Ok sounds silly but hold on. This would mean that no matter how fast you create new wallets and transfer funds to them, you can't grab it all. After this round is over, lets say you have 1% of the total amount sold.

    Next round could be for example max 0.5 BTC and last for few days.

    After that evaluate how much you have raised and what the max and duration for the next round should be.

    Doing it this way, you'd get max distribution which can't be anything but positive for the whole project.
  • Benjamin_BBenjamin_B Member Posts: 12
    It's easier to "eliminate the dust and whales" simply by setting a minimum investment at 1 btc and a maximum at 20 -- and hold with an escrow mechanism. Given the enthusiastic response Vitalik received in Miami -- and I will see Gavin Wood speak later today at the London club meeting -- I have no doubt that with these parameters, 500 btc will still be raised in one or two days.
  • illodinillodin Member Posts: 2 ✭✭
    When talking about millions, it's easy to forget that 0.5 BTC or even 0.1 BTC is not "dust" for most people. Fully agree on it being easier to handle fewer bigger chunks though. Maybe the poor bastards can arrange their own group buys for 1 BTC shares.
  • HarryHarry Member Posts: 1
    In my humble opinion Bitshares model for fundraising prevents big investor of buying up most of the shares and allows small investors equal chance. They could put limit of 500 BTC every day for 60 days and fix daily limit of Ethers to be proportionally spilt every day between donors.
  • shitscaredshitscared Member Posts: 1
    can i reserve some, as all my btc is stuck on a btc withdrawal on gox. And they dont know when tf its gonna be fixed! :(
  • oliverkxoliverkx Member Posts: 85
    How about capping investments by BTC wallet and real-world mailing address? Potential investors can sign-up on a web site with their name and address and deposit 0.1 BTC earnest money. They will then receive a BTC address in the mail, to which they can send additional BTC, with a daily cap calculated to work toward the fundraising target while giving everyone equal access. Any funds in excess of the allowed amount are sent back or (in case of repeat offenders) confiscated.
  • BeeFer310BeeFer310 Member Posts: 6
    If the minimum funding amount is 0.01 then you could add all investments up which have at least 0.01 and have an amount x(if 10000 people are investing then it is 100 BTC)
    Then you look how many invested 0.02 and add this to the sum......
    If the sum is at any time higher than 30000 BTC then you stop and all people who were willing to invest more gets their "change" back.
    With this way you could protect small backer for whale who are investing 30000 BTC within the first 2 days of the IPO.
  • davidpbrowndavidpbrown London, UKMember Posts: 15
    Fear of whales, seems to stem from the simple thought of distributing Ether based only on money. It is important that people do invest; and obviously you can solve wealth distribution before launch. What you can do is look spread some distribution more fairly.

    So, look for a way to create excitement and reasonably reward those who are interested and engaged.

    It's in all our interests that Ethereum has as many people engaged as possible. That is, that we do enable people who are interested but cannot afford to buy in. Gifting to those who cannot and those who can afford to buy at launch, will help balance and provide a wider audience initially.

    So, I'd suggest consider a mixed distribution, that is perhaps 80% Ether available to buy, balanced with wider distribution of 20% against other measures which support Ethereum's adoption and early usage.

    Gifting Ether initially will only help create the excitement too.

    So, off the top of my head then:
    - Proof of interest - NEM's signup to a thread on BitcoinTalk, with increasing trivial payments to engage users, worked in a fashion.
    - Proof of engagement - accounts here now before any announcement; or at sometime based on the number of posts, if that works before it's abused.
    - Proof of CPU/GPU - use each wallet or other program to prove interest and lock that CPU|GPU active on the network at a particular time stamp.
    - Proof of IP address - perhaps easier to implement use each wallet or other program to snapshot IP address at a particular time stamp.
    - Proof of use - run the network for free for a short time and any users of the Ethereum contracts get a small fraction of profit each day - immediately some but only once a day.

    Having free use of Ethereum for a short period might be a useful way to engage people anyway - or perhaps better have a test net available for new users to play and make mistakes with longer term.

    You could run the network for free for a period of time and then once it's stable engage it to solve the puzzle of how to distribute 20%; so, at timestamp each CPU gifts the one node that it's acknowledges some fair amount of Ether. From that you have a large network of happy users ready to support the larger investors and more demanding contract users.

    I'd be tempted to do 20% later, based on the actual network use; [20% of the remainder 80%]=16% based on gifting as above; and the remainder 64% based on ye olde traditional buying from the exchanges. So any whales in the 64% would be hard placed to dominate the entire network in the longer term.

    Brain over.
Sign In or Register to comment.