#hangout transcripts

Other than that the other thing that is going on is I'm going to start working on a proposal for a bond market. This will be a high level specification of how the bond market could work, what the operations are and a high level guide for what would be required to implement it on the BitShares network for the Community to review.

This effort is being funded by Community members. So I'd like to thank the proactive support of people out there who are helping fund the research and development necessary to make this happen. The only thing that's funded so far is the building of the specification of the feature. Implementation is to be determined.

This transcript is a work in progress.

Send tips to : bts-transcriptor

The STEALTH transaction history has to be handled in a slightly different way than the transaction history that we use for everything else. Primarily because it can't be indexed using the same way that we index all other transactions, because it's STEALTH.

So we suspect that we'll have a Beta of the new GUI with STEALTH transactions out within the next week for everyone to play with. We'll get some testing on the user interface and then assuming everything works as we would like we will roll it out officially for OpenLedger. I'm pretty pleased with the progress they've made there. I recognize that it's a week or so behind schedule from what we had originally intended, but it appears all of the pieces are in place to enable users to transact with STEALTH by the end of this month. So that's the update there.

The next thing I'd like to discuss is the Free Transactions as implemented by abit. He's done a great job taking the initiative to implement the feature. He implemented a version that should get us 90% there using coin days and Committee set parameters to specify a conversion between coin days and BTS for the purpose of paying fees only.

This means you can still pay fees the old way, if you haven't accumulated the coin days or if you accumulate the necessary coin days then you can pay some or all of your fee with the coin days you have accumulated. I believe his approach is ... the last 10% of his approach would be to automatically adjust that ratio. But for all practical purposes we can offer our users free transactions and let the Committee members adjust it periodically based upon price and demand changes in the system.

Because we support both fees and rate limit based approach it means no users will ever have a lack of service, it just means that if the fees aren't adjusted in a timely manner some people might have to wait longer or pay a fee for immediate service. Either way it's better than what we have today. So I'm very interested in merging his changes into the codebase as part of the next hardfork in the system. Which will probably be at least one month from now. Probably two months, we don't want to hardfork too often if it can be helped. So there is that status there.

[00:05:30] bytemaster: I'm very excited about what that can do for bringing new users to the system who will hold BTS for the purpose of transacting for free. It's really not free transactions, what we're really doing is rewarding people for buying and holding and rewarding people for taking their BitShares off the exchanges and move to local exchanges. It means anyone that wants to provide liquidity in our markets can do so for free so long as they maintain a reasonable size BitShares balance to compensate the network for their usage.

So if you're a whale you can transact in the marketplace for free. So I believe micropayments will be possible while the network's young. Eventually they will be rate limited severely if you don't have a large balance. But, for now it's a great way to get new users into the system to try it all out. So I think that's great.

The next person I would like to mention is Roadscape. He recently put a Worker Proposal up there for funding his block explorer CryptoFresh.com. I am a big fan of his work. I think his website provides the public face, it's what all the passive observers use to check out BitShares. It's what I use whenever I want to know what's going on with voting and it has all kinds of statistics that are not available in the wallet. Developers like him who have been working for us are invaluable to the longterm success, we should reward them. And so I fully support funding him and I encourage all proxies to also vote for him.

He did work first and ask for money later. abit did the same thing. That type of initiative needs to be rewarded. If we don't reward it then those developers will feel unappreciated and go somewhere else where their efforts are appreciated.

So please keep in mind that anytime someone brings resources, whether they're bringing time or money, they're boosting BitShares and all those resources, the whole is greater than the sum of parts. If we all work together the value of what we produce is greater than our individual contributions. So please support him.

So the next thing I would like to talk about and that people have requested is some feedback on tonyk's suggestion of making BitShares only tradable on the DEX and forcing exchanges and anyone else to go through either bitAssets or IOU assets. This is a way of saying, let's make BitShares the exclusive asset that can only trade on our own blockchain.

I think that it's an interesting idea in that it forces people to use the system and it gets the BitShares off the exchanges. And it adds more liquidity in the market because anyone in or out and speculate on BitShares will now be forced to do so through trading on the BitShares exchange itself.

I find the approach is simultaneously heavy-handed and would greatly disrupt a large number of users. A lot of the liquidity that BitShares has is from speculators who have money on the exchanges such as Poloniex and BTC38, but who are unwilling to provide the same liquidity outside of that environment for the same reason that people don't move to other exchanges.

Also they have bots that are programmed to use their API on Poloniex and they enjoy the instantaneous transfers of buying and selling between all of the assets on Poloniex and BTC38. So anything we do to restrict BitShares at this point in time is going to have massive short-term impacts on the liquidity of BitShares and liquidity is king. Liquidity of BitShares itself is king as far as providing value and collateral for the bitAssets we're creating.

That said, we need to encourage liquidity in our own markets and liquidity is something that is not free to provide, it's as valuable or probably more so, than any Worker Proposal that we are currently funding. Someone who risk their capital in our markets is getting a relatively small profit if they are successful in their market making, but they are taking large risks.

Anything we can do to subsidize liquidity has a leverage effect. We put $1 in, we get $100 liquidity. That's a huge, a very efficient way of amplifying the influence of our limited funds is to subsidize those who provide liquidity to BitShares.

So our transaction fees are the opposite of subsidizing liquidity, so that's why I think the first step in bootstrapping is to temporarily remove transaction fees and market fees. And then go to the other extreme and have Worker Proposals that subsidize market makers in particular markets and the BitShares stakeholders can vote on these workers to figure out which markets should be subsidized and with how much. So that's my proposal for improving the liquidity on BitShares.

First we make it the cheapest place to trade. Second we make it effectively negative fees. If you put your money in our order books and you leave them on our order books and it eventually gets filled, congratulations you've earned some money.

That should dramatically change the risk reward profile and encourage traders to adopt our platform. Because now in addition to the money they are making from trading they are also making money simply because they are trading on our platform and they are sharing in the long term success of our platform.

So doing something like this in the bitUSD, bitCNY, bitGold, bitSilver and bitBTC markets would have a dramatic impact on the liquidity, on the tightness of the peg and on the volume traded there. If we had significant volume on the bitBTC / BitShares market you better believe a lot of people who are currently trading on Poloniex would switch to trading on the internal exchange.

There's lots of people in this Mumble session and other participants in the Community are BitShares fans who don't trade on our own exchange because there's not enough liquidity. So think of the magnification impact of paying a little bit to incentivize people to move from Poloniex and BTC38 and other exchanges into the internal exchange.

So that is my proposal for short term roadmap. Liquidity is everything. If we don't get liquidity then it's going to be slow to grow. And we don't have forever to grow. I'd like some feedback on that approach and see what you all think.

[00:15:28] Roadscape: How can we prevent abuse?

bytemaster: The way we prevent abuse in the liquidity provision is require all orders to be on the books for at least a minimum period of time. That might be 1 minute, it might be 5 minutes, it might be an hour. And you only reward orders that get filled within 24 hours.

This means anyone who tries to trade with themselves is not guaranteed, they're taking the risk that someone else will take them up on their order. And anyone who tries to be too conservative by placing their order too far away from the current price is unlikely to get filled within 24 hours, which means they're not actually providing liquidity today. They're providing long term base support in the market, but they're not providing liquidity on any given day.

So if there's no one else trading in the market then those people can trade against themselves, but it's a competitive game. If someone puts a buy-wall up and hopes someone will match it and no one's trading then someone else will put something right in front of it and then they'll get the points for the liquidity instead of the other person.

So the impact here is to create a very tight market where the parties are constantly fighting to be at the front of the line so they can get the reward when there is a legitimate trade. And by giving time for participants to see the market and place orders that get in front of the liquidity providers it means that you can't trade it against yourself unless there is literally no activity in the market. If there's no activity in the market then we're effectively paying people who are self dealing interest on the money they leave on the books waiting for orders to show up.

But either way you're not going to be able to abuse the system. The network has fixed cost. There's a certain amount of money that will be distributed to the people who provide liquidity proportional to the amount of liquidity they provided. And with that system in place this works very much like mining.

Someone who's self dealing, well they're not going to make any more money, they're just going to increase the percent and everyone has equal opportunity to self deal. Which means it will actually create liquidity and it will create volume. Furthermore people are rate limited on the rate at which they can transact without having a large balance, so all these things come together to prevent abuse.

[00:19:13] Community: And then the trading pair would have to involve BTS?

bytemaster: The trading pairs would be selected based on Workers voting. So if you wanted to subsidize OPEN.USD / BTS we could. And at the same time OpenLedger could subsidize their own assets using the exact same logic. Any asset issuer would have the ability to subsidize liquidity in their market. We would just have Workers that would fund the subsidization of liquidity in certain markets.

If you subsidize liquidity in BitShares versus bitUSD that makes it much more appealing for shorts because they have a liquid market they can trade in. And at the end of the day the increase in people who have their money on the blockchain who are trading and who are willing to go short because there's a liquid market that they can easily cover in. Or if they do get margin called there's a lot of market depth to minimize the cost. It makes the whole system more secure and makes the pegs tighter which makes the whole system much more valuable.

So like I said the value added to the network is leveraged by 100:1 for every $1 we pay in to enhance liquidity we're adding $100 worth of value to the network. So the [inaudible] I think are much larger. It's like the butterfly, we flap a wing and we have a hurricane of liquidity and growth. That's my mindset there.

We want to encourage people to come to BitShares and trade. If they're not trading then we don't have a market. And we want to encourage traders and those who come to BitShares and trade provide more value than those who come to BitShares and hold. At least ... because that's what BitShares is.

[00:22:23] Thom: So the hardfork coming up this Tuesday is to get the backend changes in place in preperation for STEALTH and Fee Backed Assets, is that correct?

bytemaster: Correct.

Thom: Regarding Roadscape's CryptoFresh website, it is fantastic and I really appreciate his initiative as well. When I was reading his Worker Proposal I didn't get a firm commitment that the list of features that he was anticipating adding would be what he was putting forth for the actual proposition. In other words it was a little open ended. I still think he needs endorsement anyway even if it isn't really rock solid in terms of what he's going to deliver specifically. But I would have liked to seen that tightened up a little bit more.

bytemaster: Yeah I'm basically endorsing him because he's already done enough work in the past to justify the Worker Proposal even if he doesn't do anything new in the future.

Thom: I agree. It's fantastic work and he's done this all on his own initiative. I don't disagree whatsoever, he needs our endorsement for sure for this Worker Proposal. My only comment is really that if you put a Worker Proposal out there it should be something that is, "I've got this and you get that", type of thing to quid pro quo in terms of give and take. But it's a minor point because I believe his Worker Proposal needs to be approved anyway as you say bytemaster from the merits of what he's already done if nothing else.

Roadscape: It's a valid question. I guess the big thing is I plan to stay agile because things change quick or a feature comes out we want certain data. Without knowing what's going to come up, I would just say the top priorities are, if people are excited about the social network experiments I would really like to experiment with that. The API is a pretty high priority and then the upcoming features I listed are all kind of either nice to have or important to have. So I would just use my discretion along with Community feedback to figure out the priorities for those items.

Thom: Sounds fair to me.

BrindleSwan: I guess I'd like to say, I think the work of this Community over the last couple of months has been amazing or the last couple of years. I know I'm a zero of a presence in the forums and have just been showing up here since like September. I'd like to ask other people to speak up. I'd love to hear from the proxies. I'd also like to say that as much as I think traders are an important first niche market, I'd eventually would like to see this become a platform where people who are not interested in trading, but just want to earn a living selling the POS systems and getting referral fees and maybe winning some jackpots and just having fun with it. I see that as the really big breakthrough moment. And thank all of you guys, I think everything is really awesome.

bytemaster: A lot of the systems that we want to do depend upon having the liquidity and the strong market support and market confidence. So  liquidity builds confidence and confidence has as a very powerful impact on everyone else. So if traders have confidence then everyone else will have confidence. But it's kind of hard to go the other way around.

I agree with you at the end of the day, point of sale systems and other things on the blockchain, even the social networking stuff that Roadscape talked about, those types of things have huge impact. The prediction markets on the blockchain are also huge.

It's just we have to figure out how to incentivize things we want. I think that's the biggest thing that I'm coming to appreciate is you get what you pay for. If you don't pay for anything, you get nothing. If you pay for hash power you get tons of hash power. So if you want orders on the books, you need to pay to have orders on the books. The opposite of that is, anything you charge for you get less of. So eliminating the fees is the first step of incentivizing what we want. We want people to use the network to transact and the fees were preventing that.

So a lot of this stuff is actually critical when we're growing and as we get larger then we can use our network effect and monopoly if we are so fortunate and charge small fees for things to make money. Things we can sell that we could raise money for is the premium names and asset symbols. Charging slightly more for that could have a way of generating revenue. Those are assets that are not based on transaction volume or usage, it's like real estate. You buy it to hold it to do something with it.

So that's a reasonable thing that we can continue to charge fees on. But until we actually have liquidity on the order of Poloniex in our own internal markets, charging any fees is going to hurt it's friction. And if we can get negative friction then we can grow. And once we've got all the liquidity people will stay for the liquidity even if we're charging them fees. So that's my take there on where we should go.

This transcript is a work in progress.

Send tips to : bts-transcriptor

So that would be the 18th I believe. So that should have everything required to pay the STEALTH fees to the STEALTH Fee Backed Asset. The management where the top 5 stakeholders of the Fee Backed Asset get to decide things, automatic buy-backs, the markets, all those different things he says are implemented and working. So that should come out then.

I haven't gotten the latest status report from James or Valentin. They are on the Mumble session today so maybe they could just give you an update on where STEALTH is. If one of you guys wants to speak up, let me know, otherwise we can give you an update probably next Wednesday or the Wednesday after that, we'll have the GUI update.

fuzzy: Do you guys want to talk at all?

[00:02:57] James: I can say that we're able to use the wallet importing some private keys into the new STEALTH wallet and Valentin has got some components that he's going to be integrating. He started yesterday, so that's functional for him. I've got a lot of little things to convert the old wallet in there and a lot of testing to do as well.

bytemaster: So to summarize, we've got most of the parts. A lot of the things that's taken unexpected time is the integration backwards compatibility migration. It's one thing to implement a feature, it's another thing to implement a feature and integrate it with older code and make sure you don't break anything. But thank you very much James for that report. So there's that.

Other things this week, it's been a big week in BitShares news. A lot of excitement on the forum, a lot of great ideas. I guess Tony's not here but he actually came up with an interesting productive idea, even though I concluded that it's probably not in the best interest of BitShares to implement the idea. The discussion and the ideas were incredible.

DataSecurityNode: We've had the highest market cap since November and highest 24 hr volume, over $1 million in BitShares.

bytemaster: All the BitShares Community members who speculated on Ethereum and brought the profits back to BitShares, thank you very much. So it's been an exciting week.

I released a whitepaper / blog post on the rate limiting stuff that we talked about last week. If you haven't seen that you should check out my blog. I think that there's a lot of excitement after last week's Mumble discussion. A forum poll showed something like 73 people for and 5 against. That's the most agreement I've ever seen on almost anything on our forum and lots of very positive discussions.

I'd like to address the number one concern that is what does this do to the value of Lifetime Membership. I believe the answer I came up with are two-fold. One we can give Lifetime Members a longer averaging window which would allow them to burst to higher throughputs. And two, we can create two different classes of rate limiting. Lifetime Members are allocated some percentage of the bandwidth. Non-Lifetime Members are allocated another percentage.

Considering there will be fewer Lifetime Members that means they'd have, it would be like the HOV lane. Which means you can transact and post more when you're a Lifetime Member while holding a lower balance.

[00:07:12] fuzzy: Is there any way this can hurt the mechanics of the market in terms of the fairness of it? I haven't been able to bump up against it and I'm not an expert in this area and I think a lot of you guys might have some insight on stuff like this, but are there any downsides that have been talked about that I haven't seen?

bytemaster: The primary downside is having to tell people, "No you can't transact right now".

fuzzy: But that's only going to be a very small window, right?

bytemaster: Yes. And it should only affect us once we're already saturated.

fuzzy: So it's essentially kind of like how there are different fees in Bitcoin as far as the priority list.

bytemaster: So as far as the cost to develop this based on our experience with STEALTH, I'd say it's the integration making sure we don't break anything in the process that's the biggest challenge, but it'd be a couple of weeks to implement it. I think abit has offered to implement this feature and he's got a pretty good understanding of the code base. He's been doing the percentage based transfer fees.

So I am all about diversifying and decentralizing the development of BitShares, so if he'd like to make his own estimate of what he thinks it would take to do this, I think having a few more minds looking at the varied details of the proposal and figuring it out would also be useful.

fuzzy: Out of curiosity, you said there would be potential things, that the concern that you make any big changes like this there can be problems that can arise. Is this planned or going to be tested on the testnet that xeroc's running?

bytemaster: I think it most definitely should be tested on xeroc's test network, particularly because this test depends upon flooding. So the goal is to make sure that someone can't flood the network regardless of how big their stake is. So a test networks the perfect place to test this.

fuzzy: With respect to that, I was talking to people in the Committee the other day when they were talking behind the scenes, what they were kind of talking about with respect to this stuff, the amount of volume on the test net and the problem is getting a lot of liquidity and volume on there. Have [inaudible] potential of using like a Worker Proposal or something like that or some pool of fees to occasionally have like competitions, because I don't know if you guys are familiar with the Motley Fool, but they have a test net and they have a winners list, like who's the best on this test trading portion of their site. And then they have an exact same replica of the real where you can actually go out and trade and do other stuff through using like all the stuff they offer. So I'm trying to think, is there a way of making like little competitions, maybe like rewards to get people onto the network. Have you guys thought of how we can get people onto the test network itself and actually incentivize the use of it?

[00:12:11] bytemaster: Yes we can do all kinds of things like that. Running those competitions takes time and money. You can probably make it a pay to play kind of thing where you put money in and then you do the competition on the test network and then some of the proceeds go to the people who are running the test.

But with respect to rate limiting, one of the challenges we're faced with BitShares is new users should be given enough stake so that they don't encounter rate limiting in normal use. Which means that right now when we create a new account the faucet will need to fund that account with maybe $1 worth of BitShares. And if it funded the accounts with $1 worth of BitShares it would need to do so with like Facebook validation or something to make sure that people don't come and drain it dry.

It also means that it's probably something that the network would have to subsidize. So the network will have to trust someone to hand out those type of signup rewards.

xeroc: Are you saying that every user or customer of our network has to have some BitShares in his account to use this?

bytemaster: Yes. It's very much a membership organization. You have to have a club membership to use the club facilities. But then when you have a membership the usage is free. So it's sort of like saying pay $1 and you never have to worry about fees again.

xeroc: I get that point, OK.

bytemaster: So a new account wouldn't have to wait to transfer at all. Well they might have to wait a couple of minutes. If you put a dollar into the account it should accumulate enough to transfer in a couple of minutes, probably less, but that's not really that big of an issue.

betax: Will this be a faucet for new users?

bytemaster: Yes. It would just be the same faucet we have today, it's just that it would fund the account when it's initially created with enough BTS to do transfers. The other effect of this would be a minimum balance requirement. Users will not be allowed to withdraw BTS from their account if it would render the account unable to transact.

hadrian: How long will a new account have to wait until a transfer can be made?

bytemaster: I answered that a couple of minutes ago.

fuzzy: [having mic issues] bytemaster can you read out the questions?

bytemaster: Alright, so some people have asked, "Why not have a fee as a backup?". In my whitepaper I address the issue very clearly on the economics of having a fee alternative. The fees can now ... with the whitepaper I introduced an object method for determining fees based upon network capacity. And for a given transaction we know how much stake you need to hold and for how long you have to hold it in order to transact. And if you're going to pay a fee rather than hold stake for a minimum period of time, then you effectively need to determine the interest rate you're going to charge on borrowing that much money from the network for short period of time.

So read the whitepaper for more details, but it should be possible to calculate what the fee is and to make sure that it's high enough that it's beneficial for the network to allow someone to pay a fee to consume bandwidth versus having a membership. Just keep in mind that allowing people to pay a fee to consume bandwidth can make it cheaper to flood the network.

In my example I showed how someone can completely cripple the Bitcoin network for a whole day for just $10,000. Someone with the same $10,000 attempting to spam under the rules I presented wouldn't be able to affect the network for more than 30 seconds.

That is a huge difference. And so even if you have relatively high fees, if the fees would scale up automatically as the flooding commences. So there's an equation there, you can have a fee fall back. You can also have a situation where rather than charging a fee you take the users money and you hold it off the market. You lock it up for a period of time and you return it to them. It's still zero fee, it's just we allow them to pay it with future commitment to hold rather than past holdings.

Which means if you start to spam the network you're going to end up locking up more and more of your stake and then get it all back later. So those are a lot of the ideas.

So the issue here is why it matter that the network gets flooded? Two things, flooding the network creates bloat in the sense that it takes longer to download the blockchain, it causes the blockchain to consume more RAM and makes re-indexing slower. All those things impact people running the infrastructure for the network. It means that we have to pay Witnesses more to do the job.

The other aspect is at any given time there's a technological limit to what we can support. The stress testing of Graphene showed that we could get up to 1,000 transactions per second but we're actually running at effectively 100 transactions per second right now. We need to scale, you know manage supply and demand for that 100 transactions per second until we can add more capacity. Obviously the answer is once you hit your capacity ... the correct choice is to find a way to increase capacity, not increase fees.

So that's something that we can do and that's why I mention Bitcoin that you might have to hold $25 worth of Bitcoin to transact, very conservatively, once a week. But if Bitcoin increased their capacity to what BitShares is capable of then it would be less than $1. It would be easy for Bitcoin to increase it's capacity by a factor of 10 or 20. So there's that.

[00:22:37] fuzzy: So it comes down to the amount that the Witnesses are going to have to pay for the actual hardware to run the network that's being flooded. So it's a balancing act at any given stage of the network's growth.

bytemaster: Correct.

bytemaster: Someone asked the question, "Has the idea of a secondary market for timeshare bandwidth been discussed?".

bytemaster: That's sort of like creating another token, like BitShares. The BitShares is the token that represents timeshare bandwidth under this system. You don't need a secondary market. The thing about bandwidth is it's not just accumulating coin days. Unused bandwidth cannot be moved into the future. If its' unused in the past it's lost to the world forever.

Now there is another aspect to this. We talked about bandwidth being one resource constraint. There's another resource constraint which I think we might want to implement and that is a memory usage constraint. The idea here is that there is a maximum amount of memory that the Witnesses are currently capable of utilizing and every single account should have a percent ownership of that maximum capacity. And if you don't have enough stake to represent that ... let's use the club analogy. Every member of the club needs a locker. If you run out of lockers it doesn't matter that we've got time for people and bandwidth, we ran out of another resource.

So this system for fees can be applied in multiple directions. One is the rate at which you can transact. The other is how many total operations can you have in the system, like open orders, how many pending operations, how many people you're voting for, all these things. We can calculate the memory usage that each user is using and reject any transaction that would cause bad user to use a larger percentage of the total memory allowed than their percentage of total BitShares held.

So then the answer is, if that becomes a problem the Witnesses need to increase the amount of memory that they support. And that gives us a nice way to balance supply and demand. Because even though we might have the network capacity to support flooding, if the flooding results in the creation of millions of orders on the books, that could cause us to run out of another resource.

Community: I have a question about the rate limiting. It sounded like there was some sort of exponential backoff on when people pay a fee to try to throttle them from doing that. How does that scale as far as somebody instead of doing it from one account, they just make a bunch of little ones and divide their money up trying to do a bunch of transactions.

bytemaster: It's proportional to stake. So an account with 2 BitShares can transaction twice as often as an account with 1, so if you divide it into two accounts you're overall transaction rate will be the same.

Community: I see so there isn't any sort of exponential backoff is not true then?

bytemaster: There's not exponential backoff. There just, you've got so much stake based on the current network utilization you're allowed so much bandwidth and any time you exceed your bandwidth you're throttled.

Community: Is that done based on a per account basis?

bytemaster: Yes, each account has it's own bandwidth measurement and when an individual account's exceeded the bandwidth it's allowed based on the stake that it's got, it can't transact. But as time passes it's bandwidth usage falls, because it didn't transact, which means it will be able to transact again.

Community: So I guess my question would be if somebody was preparing to do this type of attack, rather than putting all of their stake in one account and get it all bulk rate limited they would make 100 accounts or 10,000 and then if one got limited they still have 999,000[sic] more.

bytemaster: Yeah it wouldn't work. I guess I don't know how to explain it more concisely, but imagine transactions are like water. Transactions are like water, each account is a pipe of various diameter. You can only push so much water through a pipe with a certain diameter and where they've got one pipe with twice the area of a smaller pipe or two small pipes, you can only get so much water through. So there is no way to gain or throttle it or defeating the system.

fuzzy: Essentially you're saying it doesn't matter because you as a person are going to be holding all these accounts, the amount of accounts is not important at all. All that matters is the stake that you hold.

bytemaster: Correct.

Thom: It doesn't matter that you distribute that stake over a bunch of accounts it still represents the same amount of cut of the bandwidth.

fuzzy: Exactly.

bytemaster: So the goal of the system is not to have anyone's funds locked up in any perceivable way.  Right now Bitcoin locks up everyone's coins for 10 minutes at a time. You transact and your funds are locked and you can't touch them for 10 minutes. The rate at which an account with $1 can transact will be much faster than every 10 minutes.

Only if someone was trying to abuse the system would it cause that time delay to grow. But the numbers that I showed shows that you'd have to be a whale to meaningfully impact the overall reserve rate.

Thom: Didn't you mention that abit was going to be willing to do the coding on this feature?

bytemaster: Yes.

Thom: Is this going to be a feature backed asset and if not why not?

bytemaster: Well considering there are no fees involved, it's kind of hard to make it a Fee Backed Asset.

Thom: Well I guess Fee Backed Asset is different in my concept than a feature backed asset, but your point is well taken and that does bring up the question, how does this impact feature backed assets?

bytemaster: As a network we have to make a choice between charging people and network effect. And having the freedom to decide when to charge people is incredibly valuable, right? You want to charge people for premium things and give them the first hit for free, that kind of thing.

So if you're building a Fee Backed Asset of feature backed asset, having the ability to control when and how you charge customers for things is very important. So I'd say a business that wants to build something on the blockchain has to decide whether or not they want to create their own token, build their own network effect or charge fees. And there's the two ways to do things.

Thom: I see. Well it is kind of counter intuitive. I guess I didn't see it right off the bat, but now that I think about it I can certainly see why this can't be a feature backed asset because the whole idea is to eliminate fees so you can't charge for that addition or that feature because it's just counter intuitive right on the face of it.

bytemaster: So this is a feature that grows the network effect it benefits everyone. It's a perfect case for a Worker Proposal. So that brings me to the next topic. I had another blog post in the past week that I think is something that is worthy of discussion and that is the problem of currency distribution.

The concept that I wanted to introduce here is a paradox that exist, I call it the paradox of scale. We all know that inflation, the printing of the dollar results in a transfer of wealth from everyone to the people who receive the new money. And that the impact is the loss of purchasing power and we all experience the loss of purchasing power as a result of government inflation.

On the other hand ... so that's when something is the reserve currency of the world, that's the dollar ... then you take something like Bitcoin. The creator of Bitcoin, Satoshi, was the owner of all the coins and he decided he was going to distribute the coins according to mining Proof of Work algorithm and so he starts inflating his coins in order to distribute them and get more people involved. And as a result, despite the inflation in supply, the value went up.

So somewhere between the birth of a new currency and the time it hits world reserve currency scale, inflation turns from a good thing and a necessary thing and into a bad thing. And the question is, why is that, how does that work? And the way I'm starting to think about it is, it's like putting two logs next to each other, alone they are worth less than they are together. When you inflate to bring new people in the result is 1+1=3. If I'm the only person who holds it, it's worth less. If I get two or three people that are willing to trade it, A) it just went up dramatically in value. If I get a million people using the same currency it has dramatically more value even though I might own a smaller percent of the overall pie.

So this put an interesting perspective on fees. Think of fees as the opposite of inflation. It would be better for a blockchain to charge no fees and inflate to cover expenses, because you're growing network effect and the result is that the blockchain grows in value by more than the lost revenue. Particularly if the only people that you're paying are people who are providing demonstrable value to the network. They are providing the infrastructure and they're providing the support and the maintenance and whatnot.

Anytime you pay people who are providing value you are growing network effect. And even if the short term price declines, at the end of the day the short term price is insignificant next to the network effect that's being built, next to the sense of belonging that people get.

And so if you view the problem of currency distribution, fees are the opposite of distributing, they're collecting the currency back and by not charging people then at the very least you're even. They can use the system and they don't feel like they're paying any fees. You're distributing the fees to the users of the system which encourages growth and adoption.

So I would say that Bitcoin decided some arbitrary thing, we're just going to distribute 50 Bitcoin's, then 25 then 12 every ten minutes to whoever digs a hole and fills it back up again. Initially it was helping, it had some initial benefit of securing the network, that's long since past it's usefulness. Initially it was distributing it widely, now it's concentrating and it's being sold onto the market.

Thom: If you can make sure that your M1 supply matches or is correlated with your Gross National Product then you have a sufficient currency to transact all the business of that GDP. So it's when they get out of balance, when they stop being correlated to one another that you have inflation or deflation. So how would we measure the GDP within our system?

bytemaster: We have the instantaneous price which is what the market's telling us, then we've got the prices that each of us has in our head of what we think that this thing that we're building together is going to be worth one day and that's very different than what the market is telling us.

So people are willing to work to build something speculatively on its growth. When people get together to form a startup, they work for low wages, no wages and get equity that has no real immediate value. but they work because they know that it will have value some day. And everyone knows that if we all work together that it will grow faster than if we just go off on our own. So at some point, it doesn't matter how much money or what the actual dollar amount that you're distributing is. It matters that you distribute it to someone who is doing work and that you distribute it to the person who is willing to do the most work for the least reward.

So the call to action here is that we should not look at every person that we pay as someone who's going to push down our price. That only matters if we're looking to sell. If we're looking to build something together the work that they are doing is increasing the probability of the long term success of everything that we have.

So there's two factors, there's current price times probability of success and the more people that are working we're increasing the probability of success which increases the perceived value of today, but it doesn't necessarily show up in the instantaneous value. Because some people want liquidity today.

So today's price is a measure of who wants to exit on any given day versus who wants to buy in on any given day. But the people who stay in doesn't change that much. My point is that Bitcoin has been inflating in its first two years at like 15%-20% per year and look at how fast it grew.

BitShares has a maximum inflation rate of like 6% or less. So even if we were inflating at our maximum rate we'd be inflating we'd be inflating less than Bitcoin and all the other altcoins out there that have ... very few have low or no inflation. And I think that if we view inflation and model it mentally as if we were the reserve currency of the world, that's doing a disservice because that's not the situation we're in. That's not the right mindset to have.

As some people might say, well if you're inflating that's going to cause the price to go down short term. If I'm a speculator and I know that the price is going to be going down short term then I'm going to sell to get ahead of the inflation. But if they inflation is known and everyone who bought in, bought in because they knew what the inflation was then the speculative price that things trade at will have that baked in. Like Bitcoin has the mining rewards baked in and then it will grow based on that.

So inflation only has a negative short term affect on the price when it changes in value. If it's a constant rate of inflation then the market price will reach to factor in the constant rate of inflation and it doesn't matter any more. Now all that matters is that we allocate what we inflate to the people who are doing the most and the highest value work that we can.

Community: In my mind the one sentence way to describe what you're saying I think is, the value of what you're holding is appreciating faster through network effect than it's decreasing due to inflation. It's a net gain.

bytemaster: That's correct. And I would look at it this way. If you want to start a fire you have to gather lots of twigs and then light a fire, then build things on top of it. When you're in the building phase you need more wood. You don't say, well I've got my little pile of twigs, let's stop bringing more wood to the flame. The pie grows every time someone brings wood in the form of labor or money to the equation. And if we say, well we only want people to bring money to the equation and we don't care about labor, well money doesn't grow in value without labor. So you create a ponzi scheme if all you want is people to invest money and not labor when you're trying to start something up. You need to put money in and you've got to spend money to grow and to bootstrap.

Thom: If we can't make that correlation between the supply value and what's being transacted on our network then we're going to have too much inflation or too much deflation.

bytemaster: I'm going to disagree with you in the sense that, when you're small you have no mass. And without mass you don't have efficient markets. Which means that the short term price that you see is not driven based upon growth of the economy, it's driven based upon short term fluctuations. It's like Bitcoin, it's pretty big, it's got a lot more mass than BitShares, the price moves up and down as people enter and exit even though the Bitcoin economy is constantly growing and is probably growing at a faster rate than it's inflation.

But you can still have massive swings down and it can lose half it's value, you can have speculative bubbles that push it up and down. So when you're small it's not possible to up a 6% a year inflation rate, it's insignificant next to the volatility. Which means that even if the price goes up by 10 or down to a 10th, that's within your range of volatility the markets assessment. It has nothing to do with the size of the economy, it has everything to do with speculators perceived value and their demand for liquidity at any point in time.

If they see something like Ethereum going to the moon they might sell their BitShares to get onto Ethereum to make some money and then turn around and buy BitShares back. Who knows what's going to happen. The other markets are ... I guess the GDP analogy for inflation and correlating supply growth and price stability ... price stability is irrelevant. Look at electronic prices, I think that's not the goal here.

And all the economics lessons that we have from Austrian school and those philosophies on how global money supply, universal money supply should work ... they break down on a smaller scale because it's kind of like saying the Sun's got gravity and a marble's got gravity, but which matters more. The marbles not going to move the Sun. The marbles going to get blown all around even though it still has gravity. Other things like the wind and other objects around it have a greater impact on where that marble is located and how fast it's moving than anything else.

So when we're small you can't use the rule of thumb or the things that apply at large scale because there's too much noise at the lower scale. So you almost have to have faith buy continuing to work at it together it will grow in value and it will over a long enough time frame you can actually see the growth. But that time frame, the smaller you are the longer it stretches, the more volatile it is, the longer it takes to separate the signal from the noise.

The only thing that's going to destroy a Community is if someone says, "it's all mine, mine, mine. You have to pay to play with me". That mindset is what turns a lot of people off of BitShares. A lot of the theories that underline BitShares, it needs to be profitable, it needs to generate more from revenue to cover expenses and all those theories, they are technically true. But the one thing that we have underestimated is the value of network effect being greater than the value of revenue.

And to some extent we are penny wise and pound foolish. We charge a 1 cent fee to get some revenue and we lose a dollar worth of network effect. It's very very hard to visualize the network effect in order to understand that 1+1 does not equal 2. It's also hard to use valuations. If you use a valuation based on revenue projections ... a lot of the centralized exchanges, even the big ones, aren't worth as much as Bitcoin is worth.

Thom: The problem with the network effect analogy, I mean it's a good analogy I'm not disputing your position, but it's more speculative as to what it takes to get that than it is to deal with fees and more tangible things that within in your direct control.

bytemaster: Correct, it's very hard. Since Bitcoin entered the scene ... and actually maybe even before Bitcoin, but in the past 15 years web 2.0, the value of websites is users and their attention. And that has more value than anything you can actually charge the users for. Getting people's attention is the scarcest resource in our economy. And whenever someone's using your token you're getting their attention.

Whenever they do something for your token you have their attention, that is a scarce and valuable resource that companies pay a lot of money for, hundreds of dollars per user is not unusual for companies to attract people. Because people in large groups are incredibly valuable. Anyone can create a service like Twitter on a technical level. But you can't copy it's network effect. And that is a very ... it shows that it has very real value and we're seeing it not just in cryptocurrencies but in all kinds of services. Things have value because other people use it.

Thom: One thing that comes to mind when you bring it up bytemaster, it's really good you're getting into a philosophical level at this point because what it tells me is that, you know things like how people view anarchy versus minarchism and statism, there is mindshift. And so we're in a battle of gaining mindshare and what you just described was another example of gaining mindshare. Are people attentive to what you're trying to push. So it's just another facet of how we're trying to gain mindshare and get people over. Now whether it's individual and granular versus a broader thing for anarchy or non-aggression principle type of things that's just a matter of scale.

bytemaster: Correct. So imagine you're trying to grow the libertarian movement, but you charge everyone to be a member and you charge them every time they vote. Anytime they participate in any event you charge them. You're not going to grow you're going to die. And people are going to say, "you just want our money". But if you say, "hey join our movement, we'll pay you to come to our talks, come to our events, we'll pay you to vote for us". Those types of things dramatically impact the rate at which things can grow.

And this has been a real challenge for me as a libertarian to balance the free market, capitalism, private property, anti-socialism mindset with the real benefits that certain socialized things have, like insurance. Insurance is socialism, but it has huge value to the individuals.

It's like penguins, they circle up for body heat and they rotate who's on the outside. Any individual penguin who didn't participate in the socialized heat sharing would die. And so it's really really tough for me because I don't like taxes, I don't like socialism, I know that central planning fails and I don't like other people having arbitrary power and particularly enough power to use violence. But there's this balance here that I'm coming to appreciate and it's very nuanced and the gut reaction to inflation or dilution or taxes or socialism turns a lot of Libertarians away.

And it's because we're not actually understanding what we're actually against. We're against the use of violence and coercion. That's what we're against. We're against not having the right not to participate. But voluntary socialism, if you will, where you can ...

fuzzy: It's family.

bytemaster: At small scale it's family, if it gets slightly larger you get Community, you get mutual aid societies and all this stuff. That's how people share the heat so that we don't die in the cold. It's how we collectively can share burdens that are too large for any one person to do on their own.

And in fact every company, they call it capitalistic, but everyone company is a socialist system. The shares in the company, in the venture, you're sharing the profits, it's socialism. So recognizing the nuances here is extremely critical to know how to build and grow a business and a blockchain in this space.

If we interpret our philosophy from a big scale down to a small scale and we make the wrong analogy or the wrong correlations we're going to die. And I guess that's my point about the currency distribution and the paradox of scale. Its' some food for thought for the Community at large.

Which would you rather have, inflation that grows and brings in new people and allows more and more people to get involved and have the price fall in the short term? Or no inflation and have the system stagnant because no one's bringing more wood to the fire and have the price fall in the short term?

fuzzy: I've always thought delegated proof of stake kind of enables true proof of work.

bytemaster: True. Yes it's subjective proof of work.

fuzzy: But subjective in a way that eventually we'll be able to hopefully find a way of judging this stuff and procedures.

bytemaster: Well, like abit doing stuff. That's something we can judge. We can see the benefit, we can see his actions on the forum. There are ways where we can judge this stuff and here's the thing. Perfection is the enemy. There is no way to perfectly assess someone's value. What matters most is the network effect of them contributing is so much bigger than any rounding error we make on how much they get paid.

Obviously we don't want to be foolish and pay people for doing nothing, but with our voting system the more competition there is, it's going to cause people like abit to come out and actually work to prove that he deserves to be voted for. Particularly when, it's not a question whether or not we're going to pay someone, it's who we're going to pay. And I think that's what the discussion needs to be changed too. Figure out who we're going to pay and what are we going to pay them for and not should we pay someone. Because saying we shouldn't pay anyone is guaranteed to kill the network effect. Guaranteed to result in stagnation and failure.

Network effect depends upon predictability and dependency, so to a certain extent anytime we can remove uncertainty it increases value. So the uncertainty about the inflation rate is a source of fear and causes the value to fall. It's possible that by committing to a guaranteed inflation rate, a guaranteed amount of work that's going to be done on something that the value can actually go up because people don't have to worry about it changing or about people reacting to a change in the inflation rate.

People don't have to worry about, "well is this thing going to stagnate because no work is getting done or is it going to not stagnate?". We remove so many worries that a 6% a year inflation is insignificant next to the volatility of BitShares on any given day. Insignificant next to where we want it to go. If we want it to go from where we are now to even the size of Ethereum, that's a 40x growth.

If we want to do that over the next year or two, what difference does 6% make or even 12%? It's insignificant if we achieve what we want to achieve. If we don't achieve that then it's also insignificant. So we're investing based on the hope to gain. If we just stay flat over the next year or we even fall 10% in the next year or two years, well OK, what did we get done in that time? We didn't disappear.

So the inflation, the reallocation, that's something that is necessary for growth. Bitcoin does it. If we don't do it, it's going to be a problem. So there are people out there, valuable members of the Community. A lot of people in China, alt and the like, who are actively voting against every single spending proposal. In fact the second largest proxy voter is alt and he's voting against any inflation whatsoever.

I encourage people to open up your eyes to the bigger picture here. It's better to have a few people sell out now if they don't like the inflation and then grow with a solid base of people who recognize the value of network effect than to stay in there and say "no, we're not going to inflate".

And this gets me to my last concept here. The problem with inflation is not that the inflation's going on. It's the power it gives to the people who get to decide where it goes. Bitcoin doesn't give anyone any power. It's going to the miners and that's it. And most of it's consumed and wasted, which means the profit from mining is relatively small.

At large scale it's the power that inflation gives those who control where it goes. Now in BitShares that power rest in the individual stakeholders. Which gives us a lot better control over where things go and hopefully it's more fair. But it's the power that's the problem, not the inflation per se. Because inflation in the grand scheme of things is almost ... supply inflation is almost irrelevant when it's only a couple percent. Considering all things in our economy are inflating, all goods have supply inflation as efficiencies improve. And no one complains about computers going down in value, chips going down in value, raw resources going down in value because we get more efficient mining techniques.

So what matters is the power and decentralizing the power. Anyway just something to think about. Obviously I'm not a dictator, I'm hopefully here to provide some perspective in lessons learned. I'm just one voice in the Community. I hope each of you get something valuable out of what I had to say today.

Send tips to : bts-transcriptor

Be sure to include in the Memo field : 2-12-16 Transcript Tip