5 ms·

I've been thinking about this for a while, particularly with bandwidth as the scarce resource. I don't quite grok the connections between the crypto aspect, the

I've been thinking about this for a while, particularly with bandwidth as the scarce resource. I don't quite grok the connections between the crypto aspect, the stock aspect and the currency aspect. The problem is that (the way Naval treats them) each of the main variables seem arbitrary, but could easily determine whether the economy you create is viable. > Pre-mine or early-mine Appcoins and keep some non-threatening amount. In Bitcoin, though the system itself is "trustless", participants trust that they are all treated equally according to the protocol & open source code. In the case of a startup raising funds, you wouldn't get the same sense of altruism. The incentives are not aligned for the major players: the startup founders, the protocol developers, the other coin owners/shareholders and the miners. Should an arbitrary pre-mined amount be set (as the founders' equity), and simultaneously be non-threatening, what is to stop the network from invalidating the pre-mined amount at will? If it were instead threatening, why would miners be interested in spending compute power? Naval also writes of supporting the open source developers with transaction fees. Unlike the commodity on which the coins are based, these are not easy values to calculate. Certainly there are a lot of possible new models, and autonomous corporations are very exciting. I just don't see how you can add a founder-controlled, for-profit startup to the other players in a crypto currency ecosystem (the miners, the non-profit developer foundation and the coin owners) and get something that works.

jamesk_au 13y ago
> In Bitcoin, though the system itself is "trustless", participants trust that they are all treated equally according to the protocol & open source code. But Bitcoin participants do not have to act on "trust" because they can inspect the protocol and open source code to verify equal treatment. In Naval's system, if the premise for participation (eg equal treatment) is equally verifiable by inspection of the protocol and open source code, is there any material difference between the two? It is true that there is nothing to "stop the network from invalidating the pre-mined amount at will", but participants will only participate in the network if it is easily verifiable that the network gives effect to the premise for participation. > I just don't see how you can add a founder-controlled, for-profit startup to the other players in a crypto currency ecosystem (the miners, the non-profit developer foundation and the coin owners) and get something that works. I think I agree with that statement, but only because of the words "founder-controlled". If a network like this is to succeed, the founders must forfeit control to the network. Given that it is the founders who write the protocol, they should be happy to do that, and then set it free. I don't think it would work if the founders made some attempt in the protocol to reserve to themselves a power to intervene in the operation of the network. Can I offer another interesting hypothetical: Is it possible for a network to incorporate some kind of mechanism for democratically amending its own protocol according to some objective characteristic (or subjective vote) of the participants? If participants are mining 'shares' in the 'company', should they have the power to act as 'shareholders' and to exercise control accordingly? (Assume that some aspect of the protocol means that the controlling majority is always held by nodes that are not cooperating to attack the network, which I think is meant to be the case with Bitcoin.)
- argumentum 13y ago
  > But Bitcoin participants do not have to act on "trust" because they can inspect the protocol and open source code to verify equal treatment. Sorry I wasn't clear enough. I agree that trust is the wrong word, I should have said "participants can verify that they are all treated equally". Perhaps I'm stuck thinking of these startups like I would think of most for-profit startups where each round of fundraising adds complexity (different "classes" of shares, etc). One of the more important advances in startup fundraising over the last few years has been the lengthening of time where the startup is under the founders' thumb(s). Fundraising in the manner you suggest would give you the money of an angel round (or less) + the loss of control of an IPO. Many startups severely regret going IPO, because (unless you carefully manage the process a la Facebook or Google) you immediately cede a great degree of control to the shareholders. In this case, the only benefit to the founders (assuming their coins aren't invalidated) is monetary, while the costs are manifest. I think the pre-mined allocation will be competed out of existence, as it adds no value to the process and because as long as the currency gets adopted, the founding developers (like Satoshi and the early bitcoin devs) will still benefit greatly. Basically I can see very few benefits to the startup's founders for going this route vs normal fundraising. Democracy may be the least worst form of governance for a nation-state, but the point of a (good) nation-state is to preserve liberty, not to realize a particular vision (unless that vision is liberty or in the case of bitcoin, is the currency itself).* What could work in the manner suggested is more akin to an autonomous, non-profit corporation. (Yes, individuals can profit by being early adopters or providing value-add services, but the corporation itself is self-perpetuating and not-for-profit). I'm not saying such autonomous corporations shouldn't be built, of course they should be if they fill a need (just like non-profits). I just don't see this as a viable alternative to fundraising for a for-profit startup. * There are of course several examples to the contrary (of a "vision" being achieved in such a manner). Wikipedia (and now bitcoin) being perhaps the most notable, but all that I can think of remain non-profit organizations.
Nursie 13y ago
It seems to me that bitcoin is the wrong model for this. There are lots of other crypto-currency models that have central control as a feature. These are anathema to the bitcoin sect, but allow all sorts of other neat stuff - offline transactions for example. As they are also centrally controlled they don't require the crazy processor thrashing that is bitcoin mining, either.
- rplnt 13y ago
  > they don't require the crazy processor thrashing that is bitcoin mining, either How do you verify someone did something worth of getting the coin?
  - shawabawa3 13y ago
    You trust the central authority
    - rplnt 13y ago
      I get that part. But how would clients mine? Can't they just tell the central server that they did something, when in fact they didn't?
      - Nursie 13y ago
        >> I get that part. But how would clients mine? Why would they need to? I'm not sure I really understand why it has to be part of the model in the first place? The company in question (as central authority) could give out coin any way it saw fit. Is it because you need their buy-in or the currency won't work? Why would that be? Miners make up a smaller and smaller proportion of BTC users all the time so. They may be needed to bootstrap a crypto currency as we understand it in the case of BTC, but it doesn't seem to be necessary that they make up a large proportion of the continuing userbase. In BTC they're also important because they secure the network, but a centrally issued crypto-currency doesn't have that problem.

jamesk_au 13y ago

> In Bitcoin, though the system itself is "trustless", participants trust that they are all treated equally according to the protocol & open source code. But Bitcoin participants do not have to act on "trust" because they can inspect the protocol and open source code to verify equal treatment. In Naval's system, if the premise for participation (eg equal treatment) is equally verifiable by inspection of the protocol and open source code, is there any material difference between the two? It is true that there is nothing to "stop the network from invalidating the pre-mined amount at will", but participants will only participate in the network if it is easily verifiable that the network gives effect to the premise for participation. > I just don't see how you can add a founder-controlled, for-profit startup to the other players in a crypto currency ecosystem (the miners, the non-profit developer foundation and the coin owners) and get something that works. I think I agree with that statement, but only because of the words "founder-controlled". If a network like this is to succeed, the founders must forfeit control to the network. Given that it is the founders who write the protocol, they should be happy to do that, and then set it free. I don't think it would work if the founders made some attempt in the protocol to reserve to themselves a power to intervene in the operation of the network. Can I offer another interesting hypothetical: Is it possible for a network to incorporate some kind of mechanism for democratically amending its own protocol according to some objective characteristic (or subjective vote) of the participants? If participants are mining 'shares' in the 'company', should they have the power to act as 'shareholders' and to exercise control accordingly? (Assume that some aspect of the protocol means that the controlling majority is always held by nodes that are not cooperating to attack the network, which I think is meant to be the case with Bitcoin.)

argumentum 13y ago
> But Bitcoin participants do not have to act on "trust" because they can inspect the protocol and open source code to verify equal treatment. Sorry I wasn't clear enough. I agree that trust is the wrong word, I should have said "participants can verify that they are all treated equally". Perhaps I'm stuck thinking of these startups like I would think of most for-profit startups where each round of fundraising adds complexity (different "classes" of shares, etc). One of the more important advances in startup fundraising over the last few years has been the lengthening of time where the startup is under the founders' thumb(s). Fundraising in the manner you suggest would give you the money of an angel round (or less) + the loss of control of an IPO. Many startups severely regret going IPO, because (unless you carefully manage the process a la Facebook or Google) you immediately cede a great degree of control to the shareholders. In this case, the only benefit to the founders (assuming their coins aren't invalidated) is monetary, while the costs are manifest. I think the pre-mined allocation will be competed out of existence, as it adds no value to the process and because as long as the currency gets adopted, the founding developers (like Satoshi and the early bitcoin devs) will still benefit greatly. Basically I can see very few benefits to the startup's founders for going this route vs normal fundraising. Democracy may be the least worst form of governance for a nation-state, but the point of a (good) nation-state is to preserve liberty, not to realize a particular vision (unless that vision is liberty or in the case of bitcoin, is the currency itself).* What could work in the manner suggested is more akin to an autonomous, non-profit corporation. (Yes, individuals can profit by being early adopters or providing value-add services, but the corporation itself is self-perpetuating and not-for-profit). I'm not saying such autonomous corporations shouldn't be built, of course they should be if they fill a need (just like non-profits). I just don't see this as a viable alternative to fundraising for a for-profit startup. * There are of course several examples to the contrary (of a "vision" being achieved in such a manner). Wikipedia (and now bitcoin) being perhaps the most notable, but all that I can think of remain non-profit organizations.

Nursie 13y ago

It seems to me that bitcoin is the wrong model for this. There are lots of other crypto-currency models that have central control as a feature. These are anathema to the bitcoin sect, but allow all sorts of other neat stuff - offline transactions for example. As they are also centrally controlled they don't require the crazy processor thrashing that is bitcoin mining, either.

rplnt 13y ago
> they don't require the crazy processor thrashing that is bitcoin mining, either How do you verify someone did something worth of getting the coin?
- shawabawa3 13y ago
  You trust the central authority
  - rplnt 13y ago
    I get that part. But how would clients mine? Can't they just tell the central server that they did something, when in fact they didn't?
    - Nursie 13y ago
      >> I get that part. But how would clients mine? Why would they need to? I'm not sure I really understand why it has to be part of the model in the first place? The company in question (as central authority) could give out coin any way it saw fit. Is it because you need their buy-in or the currency won't work? Why would that be? Miners make up a smaller and smaller proportion of BTC users all the time so. They may be needed to bootstrap a crypto currency as we understand it in the case of BTC, but it doesn't seem to be necessary that they make up a large proportion of the continuing userbase. In BTC they're also important because they secure the network, but a centrally issued crypto-currency doesn't have that problem.