Digital Ocean Versus Google Cloud, Polychain’s $1Bn Fund, a16z’s $300mm Fund, How to be a Shady Hedge Fund Manager, and 2nd Layer Solutions for Ethereum and Bitcoin

Here’s a transcript of QuantLayer Crypto Podcast #6.

On this episode Faizaan and I discuss why we are moving our alerting platform off of Google Cloud and into DigitalOcean. We discussed Polychain Capital’s $1 billion SEC filing, and how you need $1 million to invest in them. We look at Andreessen Horowitz’s $300 million fund launch and how they are thinking about good crypto projects. We go over the shady techniques that hedge funds engage in order to adduce their returns and what a high-watermark is. We cover some alerts that our platform picked-up and get into a really interesting discussion around second layer solutions for Bitcoin and Ethereum. This episode has it all. A crypto team moving away from the world’s greatest centralizer, crypto investor market news and where we think second layer solutions are headed. You will enjoy this one.

The episode in its entirety can be listened to here:

• iTunes
• Google Play
• Stitcher
• Spotify
• Libsyn

QuantLayer is a software consultancy based in Brooklyn, New York. All opinions expressed by podcast guests are solely their own opinions and do not reflect the opinions of QuantLayer. The information presented should not be construed as investment advice. Guests may maintain positions in assets mentioned in the podcast.

Vikram: Hey everyone, you got QuantLayer here, Vikram speaking. I’m also joined by Faizaan, known as the Wizard.

Faizaan: Hey.

Vikram: Hey Faizaan, how’s it going?

Faizaan: Pretty good.

Vikram: Cool. Yes. This has been a pretty busy week both on our product side and the crypto market side. One big thing that we’ve had to do is actually begin migrating our staging and production app over to a different service provider. We’ve been on Google Cloud for a while. Just for a number of reasons we decided to move over to DigitalOcean.

Faizaan: Yes, I can speak a little bit about that. Up till now just for the purpose of our prototype we had been using a relatively simple architecture for the alerting platform. As we get ready to launch we want to make it a little more reliable in terms of up-time. Also, easier to work on parts of it without risking taking the rest of it down. Complicating the architecture a little bit, breaking it out into a handful of services. Part of that process was just configuring a lot more resources on Google Cloud platform. Normally, I use Terraform for this. I’ve been very happy with Terraform. I’ve used for AWS, I use it for Google Cloud and it works really well. I also use DigitalOcean for some of our stuff.

The one thing that I’ve always found that’s a little bit frustrating with both AWS and Google Cloud, and probably some of these other larger providers is the identity and access management. Permissioning configuration is so very complicated to get started that either you’re doing something overly permissive or you end up spending a lot of time just configuring all that stuff correctly. Because it really is designed for larger organizations that might have different levels of permissions they want to assign people.

Whereas, for a smaller team like ours, it really eats away at our leverage to be able to deploy quickly. I was getting a little bit frustrated with just having to handle all of that every time I need to provision new service. That’s not that big of a deal, Terraform makes it a lot easier.

The really nasty surprise that we got was, so we signed up for Google Cloud platform relatively. Recently it comes with a $300 credit which is awesome. When I signed up, we put the credit card down, put the billing information and it was my assumption that we’ll burn through the credits and then we’ll just start getting billed. I sent alerts for you know if the bill exceeds that much, let me know. I assumed we would be getting charged. What instead happened was that once the credits run-out without warning, all of our services got suspended. We couldn’t access our bucket, we couldn’t access our instances, we couldn’t access our database. The way I use Terraform to have, so that our configuration file is not distributed but rather single source of truth. It’s based out of that bucket so I couldn’t actually change any configuration without having to like leave Terraform and mess up my Terraform config.

It was pretty unpleasant. I went through there console and re-enabled stuff, it’s not very obvious how that all works. It was a pretty unpleasant experience combined with the fact that it’s all anecdotal evidence. A lot of the places that I frequent people do have a bad experience with when Google shuts them down they don’t have a good explanation. It’s very hard to like have immediate recourse. With how critical these services are, that’s basically a deal breaker.

Vikram: Yes.

Faizaan: With some of the configuration, frustration we’d had and just how much nice of an experience I’ve had with DigitalOcean, we decided to move everything over to DigitalOcean for really three reasons. One, they’re like community team and documentation is by far the best I’ve seen. Not just among cloud providers but, It’s just like some of the best documentation I’ve seen, period. They’re very responsive.

I know we’ve discussed this before but, like my big criteria is always leverage. Like we want to pick tools that don’t stop us from doing anything that is possible. Also let a small group like two of us or three of us do as much as we possibly can.

Vikram: Yes.

Faizaan: The really good documentation combined with the really how nice the UI is makes like DigitalOcean a good choice. Especially now that they provide load balancers. They have their own storage. They have more services so, you’re not like just spinning up instances and having to configure everything manually. That’s my sort of two cents on that.

Vikram: Yes. You know the move made a ton of sense and I think what sealed the deal for me as well, was that just the lack of communication on the part of Google Cloud. We literally didn’t know. I mean I just went to the site in the morning when I got up and I couldn’t get to it. Then we chatted about it and then realized the whole thing was down.

Faizaan: I [inaudible] [07:04] my emails to look for an alert email, and there was none.

Vikram: Yes.

Faizaan: Obviously alerts are important.

Vikram: I tweeted at them, I said, “Hey? What happened?” They were like, “Sorry for the inconvenience. Can you give us some more information?” I mean, we’ve literally given you all the information we have.

Faizaan: Yes.

Vikram: Well yes, it makes sense DigitalOcean on the documentation side, it’s really nice. Like you said we got a small team so, all this stuff is super, super important.

Moving on to a little bit of crypto investor market news. Not necessarily on the coin side but the actual investor side. Kind of stuff that we learned about this past week. The first little piece of news is that Polychain Capital which is run by Olaf Carlson-Wee, one of coin bases original founders. Looks like as of February of this year had a little more than a billion under management. That’s gigantic in crypto terms.

Faizaan: Is that mostly crypto or a lot of fiat or you know, how that?

Vikram: I’m not sure what the breakdown is. Because it doesn’t get into.. The filing that all this stuff is sourced in doesn’t say.

Faizaan: Right.

Vikram: It’s probably a mix of fiat and crypto. I would be really surprised if there were a 100% invested. It’s probably a mix. That number is gigantic. Because most of the crypto funds that you hear about are pretty small, you know? Someone made a ton of money in Bitcoin and Ethereum and they pulled together some assets and maybe they have ten, 15 million under management.

I had always just assumed most crypto funds were in that kind of range so, hearing a billion-dollar number is just obscene. The exact number as of February 2018 is 1,044,064,537. That’s their total assets under management. If you look at their SEC filings they have a venture fund, a master fund, a couple other funds, different versions of those funds. My guess is, they all have different strategies like the venture fund maybe more long-term hold, maybe there’s like a trading fund that’s more short-term type of hold.

Faizaan: Right.

Vikram: That billion-dollar number is the amount of all those funds together. You know it’s really interesting to all this information comes from the SEC filing that they have. We’ll put that in the show notes. Everyone should take a look at this because it gives you a ton of interesting information about the fund. Things like how they charge their investors, for example. Who their investors are? Specific details on what the requirements to invest in their fund are?

With respect to Polychain, so like the most hedge funds they charge a management fee and a performance-based fee. The management fee is a small percentage of total assets under management. The performance-based fee is a cut of the total profits made over the course of the year. Typically, this is in place to keep the fund incentivized in order to perform well. With traditional funds you’ll hear things like two-and-20, which means the fund has 2% management fee for assets under management, and then 20% of the profits. That’s like the average type of hedge fund.

With really successful funds like Jim Simons Renaissance capital, now this is probably one of the most technical funds out there. They basically hire all the top PhDs who never leave. It’s unheard of in finance where there’s always a revolving door and it’s always swinging. You know they hire the top PhDs, top mathematicians. You hear stories about how none of these guys want to ever leave because they’re doing amazing research and they’re actually applying the research. I would say it’s the top of the funds out there. They charge five and 44.

Faizaan: Wow!

Vikram: Yes. There $84 billion fund. If they’re charging 5% on assets under management and 44% of the returns, they can pull in a phenomenal of income based on that.

I imagine some funds like Polychain could get away with these high fees. Because they have the name of Olaf Carlson-Wee, start a coin base. The coin base people like to call themselves like the coin base mafia, like how there’s the PayPal mafia.

Faizaan: Right.

Vikram: Anyway, so fund like Polychain could probably get away with it because there was so much interest in digital assets last year. I think it’s going to be harder for new funds launching to have those kind of fees now that there’s a massive slump in the price of crypto assets since the beginning of the year. A couple other interesting things to notes from that filing 61 million, 61 1/2 million were from foreign clients. That’s just interesting so obviously most of their clients are American but, they have a pretty sizable chunk that are from abroad.

Also the filing has specific details on what it takes to be an investor in Polychains private fund. The minimum investment required is a million dollars.

Faizaan: Okay.

Vikram: Not just being accredited but being accredited and having a million dollars in change to be able to invest with them.

Faizaan: To put into just one fund.

Vikram: Yes. To just put into just one fund. The approximate number of investors they have is 267.

Faizaan: On average just under four million or so per investor.

Vikram: Right, right. I’m going to guess that Olaf’s been in crypto for a while. Probably a good chunk of the fund is his. Plus spread out over all these investors. Then plus on top of that the returns from last year so, you know, billion dollars I guess make sense even though the number is totally, totally insane. That was interesting.

Another interesting piece of investor news is Andreessen Horowitz. They’re a big VC firm based in Silicon Valley. They just launched a $300 million investment fund. That’s going to…

Faizaan: Traditionally a tech fund not a crypto fund.

Vikram: Right, right. Traditionally tech but now they’re going to launch a $300 million investment fund that’s going to be focused just on crypto assets and crypto asset related investments. First of all this is super interesting. Andreessen Horowitz is a giant name of itself by being really different compared to other traditional and older VC firms. They hire a lot of non-financial people for investment roles. This is really different in venture finance. Typically you had an XI banker, they did banking for two, three years, they went to get their MBA from Wharton or Harvard or Stanford. Then they go work for some VC fund. I think these guys realize that if you have the exact same kind of people all doing the same kind of investments, they’re all going to get the same kind of returns but they’re also competing for the same type of investments.

Faizaan: Yes. Just from like a depth perspective, when you just being on hacker news or some of the other places that was hanging out, they tend to have some of the best like blogs that speak to what’s going on in the tech market but like as it relates to developers. I think they speak better to founders than a lot of these other VC funds do, maybe as a result of that.

Vikram: Yes. They provide a lot of support for their portfolio companies. They try to help them hire technical talent and business talent. They’ve done a big service to themselves by getting involved in marketing techniques to help build their brand in non-traditional ways. The New York Times actually covered all this and a pretty interesting piece, we’ll leave in the show notes from a couple years ago. It was a pretty interesting move because it’s not your typical VC fund and then not typical investors so they’re launching a crypto fund.

Another interesting part of this whole thing is that they brought down Kathryn Haun, board member of coin base and former federal prosecutor who’s worked alongside the FBI and SEC. I think this move is pretty significant. I just want to read her bio here because of how significant this is. This bio’s from her faculty page at Stanford Business School where she teaches.

“Katherine spent over a decade as a federal prosecutor with the US Department of Justice where she focused on fraud, cybercrime and corporate compliance failures alongside agencies such as the SEC, FBI and Treasury.

She was the DOJ’s first-ever coordinator for digital assets and led investigations into the Mt. GOX hack and corrupt agents on the Silk Road task force. Before that she led prosecutions and jury trials involving organized crime, public corruption, RICO murders gangs and money laundering. She also held senior positions of Justice Department headquarters in both the National Security Division and his counselor to the Attorney General where her portfolio included anti-trust tax, national security and civil matters.”

She obviously has a ton of relevant experience. I think this is a pretty big hire. It’s pretty heartening to see someone with that kind of experience at a large crypto fund, especially with the name like Andreessen Horowitz. We’ll need more actors like this is in space. Then the moonboys it’s full of right now.

Faizaan: You’re going to have to clarify what a moonboy is.

Vikram: I was hoping people just understand that immediately. It’s basically all the people in the space who were just asking, “When lambo, when moon?” Like you buy a random thing you saw on Twitter and then you go on to the telegram and you ask, “When moon, when lambo? That’s a moonboy. Yes. Not a good investment method or scheme.

Faizaan: Yes. It’s actually gotten kind of ridiculous on telegram because we follow a lot of these chats for the alert platform. I know and I think one of the coins, I want to say VChain but if I’m wrong that doesn’t matter. They’ve actually created a separate room that’s like serious crypto discussion or something. Their [Inaudible] [16:42] was just full of like wine, women and gifts. Theoretically a separate Channel and literally title it like, Serious Crypto Discussion or something. There you can only talk nonsense or posters and stuff. I just thought that’s funny. You have these like financial products worth hundreds of millions or even billions of dollars and think this is what’s having to happen. Like you imagine the same thing for like I don’t know GM or something.

Vikram: Right, right. Like at the bond market.

Faizaan: Yes.

Vikram: Yes. That Chris Dixon, he’s another general partner at Andreessen Horowitz. He outlined his thoughts on why they’re opening the fund in an open letter. We’ll put all that stuff in the show notes. Just want to highlight a few parts that are really interesting. I highly recommend everyone read this because it really outlines in a very simple format, the bullish thesis for crypto and sub-sectors within crypto and where these people are seeing things go. This is their thesis upfront, this is from their letter.

“Blockchain computers were first proposed in 2008 by Satoshi Nakamoto in the Bitcoin white paper. Those original ideas have since been dramatically expanded by developers and researchers around the world. Blockchain computers are new types of computers where the unique capability is trust between users, developers and the platform itself. This trust emerges from the mathematical and game theoretic properties of the system, without depending on the trustworthiness of individual market network participants. In exchange for these new capabilities, blockchain computers trade off their other capabilities such as transaction scalability. This can lead people to dismiss them in the same way that people dismissed early smartphones because they traded off computing power and screen size for portability and new sensors.”

A couple of thoughts about this, this is me again. This I think is one of the fundamental claims where making as crypto market participants that blockchain computers allow for trustless systems. There’s some gives and takes like because of how slow and bulky these computing systems are. You can’t expect the kind of throughput you would see with the centralized system.

Faizaan: Sure.

Vikram: They go into more from the letter here, for example, “Developers are working on upgrading the core infrastructure of the Internet including storage, networking, identity and distributed computation. Stable coins can enable more mainstream user experiences for digital payments and financial services. Crypto goods can unlock new experiences and business models for games and other forms of media. Entrepreneurs are developing crypto powered financial services like the tokenization of traditional assets and payment services for the unbanked. We expect many other applications to emerge in the coming years.”

This is me again. What they’re arguing about here is that the notion of trust is fundamental to these new systems. There will be new applications on top of this new software primitive of trust that they’re referring to. What do you think Faizaan? What out of this sort of things would seem the most interesting?

Faizaan: I think a lot of it are things that have been proposed but not implemented. Just the idea of being able to have trust between parties. They talk about like financial transactions for the unbanked but, like a lot of that same crowd there’s just a tremendous amount of corruption and like predatory behavior. Even in America if you look at the sort of financial products people have access to the way they get ripped off and especially in countries with higher amounts of corruption. Just being able to have these Tressler systems execute a lot of the sorts of transactions and services that people avail themselves of, I think can cut down a lot of corruption predatory behavior. That’s what’s most interesting to me like in the long term. In terms of functionality outside of just the financial markets.

Vikram: Yes.

Faizaan: Then I think within the financial markets a lot of that has already been discussed around what it can do for as a currency, as the security, as a utility token, etc.

Vikram: Yes, for me besides the financial services side because that is the most obvious play here. What seems interesting in their list, so they list a storage networking identity and distributed computation? Identity and distributed computation seem pretty interesting to me. I think what the on the identity side people want to solve for is, you know not having to enter your username and password every time you use a different kind of application.

Faizaan: Right.

Vikram: Have some blockchain based solution for that. That would cut down on a lot of the third-party service type of hacks that we see where information is stolen. Kudos to them and I hope that whole sector is able to find a solid solution to that. Distribute computation this one’s interesting to me because I participated in the distributed SETI platform.

Faizaan: Yes, where like you leave your computer on and it uses your compute cycles to help.

Vikram: Yes.

Faizaan: Was it search for aliens, is that the steady one?

Vikram: Yes, yes exactly.

Faizaan: Okay. I did a similar one before protein folding.

Vikram: Okay.

Faizaan: Yes.

Vikram: There’s actual crypto coin called, I think the ticker is FOLD, F-O-L-D.

Faizaan: Okay.

Vikram: It’s based out of California, I don’t know much about it. I think they’re trying to solve for that as well. The thing that I wanted to understand better was, you know I’m interested in the research that said he was doing. That’s what prompted me to want to give my computing power and it didn’t cost me anything it was in my dorm room, right? I’m not using that much electricity. I don’t have to pay anything, school’s already paying for it.

I always thought the only way to get this, to have everyone participate in this, because I think their goal was like, “Let’s get everyone who has a computer to give just a little bit of computing power to this Non-time Critical Problem.” You know if people got paid for that that made a ton of sense to me. I think some of these, there’s iExec tickers RLC, they’re trying to do this with Golem Network GNT. They’re trying to do this except for rendering graphics I think. There’s a handful of other players who are trying to do this. Those to me seem like very obvious next place within the space.

Faizaan: Yes. That makes sense.

Vikram: Back to the Andreessen Horowitz side so, they finally in their letter they provide insight into how they’re thinking about good investments. Basically they said they’re patient, they’ve been investing in crypto for a while and they’re willing to hold investments for more than ten years so, they’re trying to take a more long-term outlook I guess, and all the moonboys. They also have the assets to invest when the times are tough. This is always really important when investor says this everyone says it, it’s a question of whether or not they’re going to actually do it. Who has the courage to put cash to work when the markets getting destroyed? I mean, that’s how people make real money and there’s only a handful of funds that will do it.

They’re saying that if there’s another crypto winter that comes around, they’re willing to invest in that. Like their parent fund they’re also provide operational support to their investments, that’s pretty cool. Finally, their focus is on non-speculative stuff and they’re agnostic to stage or geography or asset type. It’s pretty cool to see a letter like this.

Faizaan: Yes. That’s nice. I mean if you just look back to like say 1997, 1998; that’s around when Google started, Amazon started, Steve Jobs came back to Apple and like launched the iPod, right? Like all of those things went through the first tech collapse where I assume a lot of capital dried up but those that you know stuck around probably did pretty well in those companies.

Vikram: Yes, the other piece of financial investor market news in crypto that was pretty interesting. Spencer Bogart, he is a fund manager. He was talking about bitcoin and crypto prices being depressed due to forced selling by hedge funds. I think he was saying it on, he was talking about this on CNBC. This is a really interesting topic and I’ve heard similar arguments before about why a particular stock is selling off at the end of a month or a quarter. One thing that everyone should know is that all fund managers are always trying to game the system at the end of months, quarters and years so they can dues their returns. I’ll explain what that means but they’re always trying to do it.

Back in the 90s, funds would mark their private tech investments with, they didn’t do it as whatever they pleased but they tried to make it look reasonable and give themselves a boost in return to make their overall return look higher. If you have fund and you had an investment in a private tech company alongside your public stocks, you could mark up the value of your private investment to help mitigate some of the losses of your stocks.

I think this got fixed because the SEC realizes this was going on. Then they basically caused a stop to this. It didn’t stop funds from trying to buy a momentum towards the end of a reporting period. I’ll just give us some examples with some math. Say your fund 10% of your portfolio is in a liquid public stock so, something that really doesn’t trade that much. Because it doesn’t trade that much, any buying or selling has it moved pretty quickly and like something that trades say a hundred thousand dollars a day compared to the billions of Facebook, Netflix, etc.

To give a sense of magnitude, some stock trades a hundred thousand dollars a day. One day you buy $200,000 of it. Now you’ve doubled the volume and the only way to only where the stock is going to react is going up because the bid-ask spreads are just going to keep going up because you’re buying all the asks. You can buy more of that stock on the last day of the trading year. What that’ll do is send that stock higher like we just described. Say it’s up 20%, right? Now you’ve just added it’s 10% of your portfolio and it’s up 20%. You just add a 2% to your returns. You just might bite the bullet next year but who cares about next year? You got your two and 20 in that year. You just made another 2% that you’ll take 20% of.

This is why you see crazy bull runs towards the end of a year. People are just marking their books up. If they see a stock that’s starting to move, they want to get on board too and you see this all the time. The flip-side of all this which what Bogart is talking about here is that, if a fund is performing poorly, its investors will want to take money out of the fund. If they decide to do that the fund is going to be forced to sell some portion of its holdings in order to pay back investors.

Say you have a hundred million dollar fund and you’re fully invested. Then one of your investors says, “Hey, I had invested this much in you and I want it back.” Say it’s five million, just for discussion. They want their five million back and you’re 100% invested that means you have to sell at least five million in order to pay them back. Often you’ll see fund say they’ll have a lock-up, meaning your investors can’t get their money back for like six months, a year.

I think I’ve heard like three years before too which is a really long time. Maybe a venture funds it’s a lot longer. Some kind of trading oriented funds they could give lock-ups of 30 days, which means you have to give 30 days notice before you can get your funds back. The fund manager has 30 days to come up with $5,000,000. The only way to do it is to sell. This can really exacerbate the prices of stocks that were under-performing before. If you’re one of these managers and you have to sell something, you probably want to sell your worst performing assets rather than your best. Say there’s some asset that’s down 10% markets up 5% so it’s under-performed. You want to get rid of that. You want to sell the one that’s up a lot, because typically people like to let their winners run and they want to cut their losers. A lot of fund managers start selling the one that’s not doing so well. If it’s down it’s going to be down more.

In the case of crypto so the markets gotten destroyed this year, so anyone who invested in one of these hedge funds in these crypto hedge funds late last year, early this year is probably upset and they want to pull their capital. That makes a ton of sense. The argument can be a little backward looking, in the same way you hear stuff like, “Tax people are selling,” because of tax season. They need to pay their taxes and stuff like that. It happens. It’s just hard to say like that’s why this is happening now. If a fund manager saying it’s very likely he talks to other funds and it’s happened to some of them so you know I’m…

Faizaan: It was at least a partial cause of some of the price moving.

Vikram: Yes, yes. Another reason why all this is relevant is on the topic of high-water marks. How high-water mark broadly is the idea that if you lost money in a prior year, so say you lost money in 2017 for your investors, you have to make them back before you can get paid on your profits. You can’t just like be down for 10% and then be up 10% the next year and take 20% of the 10% in the next year. Here are some examples, right? Say you have a hundred million dollar fund and you’re down 20% for the year. That means your fund is now worth 80 million.

Faizaan: Right.

Vikram: You need to make back 20 million. To be back to where you were before in order to take any percentage of the profits after that. You need to actually make 25% in the following year before you can take your performance fee. Now, imagine you if you’re even down more. Just consider a crypto fund. Like at the beginning of the year you were 100 million crypto, portfolios down 50%, that means your fund is now worth 50 million. Now you have to make back 50 million as a hundred percent return before you can make money on future profits.

Faizaan: Yes, that’s pretty…

Vikram: It’s brutal.

Faizaan: I remember some funds that are in horrible shape. That if they spun-up in Q3, Q4 last year, when stuff was there like 12, 13, 14 thousand.

Vikram: Yes. What happens in the hedge fund world a lot of funds will just, you made there down 50% they’re like, “Okay, we’ll just wrap this up.” Then they’ll just start a new fund. It works better for people who have historically been really good and they just had one bad year. I think people will give them a break. With descriptive funds who knows, like if you had a handful investors who were excited to give you money and you blew all of it.

Faizan: Yes.

Vikram: You kind of tarnished your name in an already weird strange space. Yes, that was the pretty interesting market news from past week.

Faizaan: Nice. Then I’ve been you know every now and then just skimming through all of the alerts that come through looking for stuff that I find particularly interesting. One of the items that I’ve already mentioned was some of these channels just having you know official serious discussion and sub channels. I think it’s something we need to address. I’m seeing a ton of banning activity too. I think there’s just a lot of either people trolling in there or what you had experienced where if you ask a legitimate questions that people don’t like, you also get banned. We’re seeing a ton of banning going on so that’s probably something worth investigating especially if you’re invested in a particular coin and the admins are just silencing any sort of reasonable line of questioning or dissent.

Vikram: That’s so interesting because we track all the admin conversation, right? I mean, we have all the data. We could actually look up by coin how many times the word ban comes up per day. They’ll be just interesting to see if there’s some co-relation there but between that and the quality of the team.

Faizaan: Yes. I think you could pretty quickly if you just look up where the bands are and pull in some of the context. If the admins are saying something ridiculous then they called-out and then they ban someone, it becomes very clear how things are being run.

Vikram: Yes.

Faizaan: Versus there’s a lot of also just legitimate trolls and fund.

Vikram: I’m not only have they been banned, I’ve also been muted. One time I was asking, I mean it’s not like I’m trolling these people. I’m just asking real questions. That I’ve asked about something and I got a message from the admin that said, “You’re going to be muted for three days.”

Faizaan: What? Like they are punishing you?

Vikram: We’re not going to ban you.

Faizaan: It goes that without warrant?

Vikram: Basically we’re not going to ban you but you’re going to be muted for three days.

Faizaan: Yes. There’re needs to be some standards brought to this stuff because that’s pretty crazy.

Vikram: Yes. It’s cool how much we can see though in these alerts. Especially around the internet that stuff.

Faizaan: Yes. Another one that I caught, again I was just skimming through a lot of these VChain ones because there’s just a lot going on with the banning and then the separate channel. Again, like telegram for whatever reason has become like the platform where a lot of information appears first. VChain is having a token swap and they published, “Hey these exchanges are going to be supporting it.” We caught that in our chatter so that’s pretty cool.

Then crypto has been banned in India like outright. Then there was a Supreme Court hearing on it. It was upheld by the Supreme Court so I don’t know how that’s going to play out but you have a pretty large market now where crypto is just ban. A lot of the start-ups that were there and you know companies that were based out of there we’ll see what happens to them.

Vikram: Yes. We really should get someone on the podcast from India who has experience in crypto. I met a couple people when I was at that Boston event a couple months back. I should get in touch with them because I think this is a pretty interesting topic that I think would be pretty good to talk about.

Faizaan: Yes. Another one I think that’s relevant to everyone that uses exchanges so I think there’s been over $700 million of exchange hacks in the last, I want to say year but I can’t confirm that. That’s a pretty crazy number for, yes first half of 2018 so just year to date we’ve had 761 million worth of cryptocurrency stolen from exchanges. That’s just something to be worried about.

Vikram: That’s such an absurd number.

Faizaan: Yes. I recently I had been one of the people that was lazy and would leave stuff on exchanges that had no intention of trading. I moved it over to a ledger because this just keeps happening.

Vikram: Yes, how do you like the ledger?

Faizaan: It’s pretty good. It’s bit of a set it in. The instructions are really good. They have like these Google Chrome web apps that you can use to install different wallets and manage your stuff. Then you know once you’re done with it you just put it away and forget about it but now it’s not online or connected to anything so that’s nice.

Vikram: Yes. I liked it as well. I got one awhile back and when I was setting it up I think I checked their Twitter and they said there was like a new firmware upgrade. The only issue I had was that I wouldn’t have found out about that unless I went to their Twitter. Again, like we should all be doing that stuff anyway.

Faizaan: Yes.

Vikram: I think on the last podcast we’re talking about how like crypto teams can communicate with their customers better.

Faizaan: Yes.

Vikram: This is the kind of thing that you don’t want your ledger to be out of date so is worth always like checking their site and their Twitter and stuff to see if they have any new upgrades especially on the security side.

Faizaan: Yes. One other item that I want to discuss, so another one of our alerts is we just look for any addresses that are using a large amount of the ETH network. I think you notice this but over the last couple of days 40% of ETH network has been taken up by three contracts. We know like scaling is one of the big issues that’s constantly discussed not just with Ethereum but with a lot of these larger like the bigger players in the crypto space.

The problem that you have with Ethereum specifically is you know with Bitcoin is if you have transactions. With Ethereum you have these smart contracts so there’s the additional complexity of A, how many can transactions can you put on a block but B, for each of those you know contracts that have to execute how much computational resources does it need to have? The base issue is for a blockchain is that every transaction needs to be verified by every node. As the number of nodes increases but particularly as the number of transaction increases the amount of work that’s having to be done goes way up. Then if you add to the fact that certain transactions can use a lot more gas or you know need a lot more computational power that doesn’t help the situation either.

There’s been a lot of chatter in the Ethereum space about like how this can be handled? Vitalik especially, has proposed a couple of different things. There’s layer one and layer two solutions that are being discussed. Layer one is something that would be within the Ethereum protocol itself that would require hard report. Then layer two is essentially on chain implementation so something that can be rolled into a smart contract. The main layer one solution that’s been proposed is sharding so similar to in a relational database where a single table might get so big that it’s not practical to use that table anymore, too many rows. What you do is you split that table into two separate physical database shards and say, let’s just say it’s a user table. Everyone with last name A through what’s the middle of the alphabet, I’m going to say M. A through M would be on one chart and then N through Z would be on another chart.

The idea generally speaking is to split transactions up across different shards. One shard would handle let’s say all the even numbered transactions and then the other one would handle the odds. The problem with that is of course if A requires a hard fork of the Ethereum of network. Number two, it sounds like it’s a pretty linear solution to the scaling. Like if you have two shards then you’ve doubled your capacity. If you have too many shards then you’re losing the guarantee of having the whole network. If you have a hundred nodes in ten shards then each shard really only has ten nodes verifying its transactions.

The layer two stuff is where things get interesting. That’s essentially smart contracts that will settle off the chain. The trade-off you’re making here is you’re getting a lot more speed, you are essentially able to process a lot more transactions that don’t have to be verified by the Mainnet. The trade-off is a little bit of the guarantees or the trust that someone won’t rip you off.

There’s three that are particularly interesting. One is state channels. A state channel is really very similar to Lightning Network. A fixed number of parties ideally open up a channel, there’s a contract in place with a judge, its multisig and you can essentially have as many transactions as you want within this channel. When the channel closes it’ll ultimately settle up and be one transaction.

Let’s say we were playing a game, some sort of like crypto kiddies board game where we are constantly making a move like every time I make a move you make a move. Let’s just say a game of chess. We wouldn’t want every single move to be a transaction on the Ethereum Mainnet, you know especially as transaction keys rise. What we could do is open a single channel for the game. Part of the contract would be having that judge that enforces that each move if it moves the transaction or the ultimate event of one of us winning causes the right settlement to happen and the way to do that is by having that multisig wallet.

I’ll post some stuff in the show notes that goes into it in more detail with diagrams. It’s essentially the same idea as the Lightning Network where you open a channel, within that channel you can have all of these transactions or state changes and then you settle up and it’s one transaction on the Mainnet.

The second is a paper that Vitalik wrote so this is called a plasma. The idea is that I can create a sidechain off of a given block. That smaller sidechain then does need to be verified by all of the nodes in the system. There can be some amount of people verifying that sidechain. Every now and then the validity of that sidechain is written back into the root node of that sidechain which is actually on the Mainnet. You have this like one ledger that’s on the Mainnet that’s verifying that sidechain. With this you can actually have chains of chains it’s nested.

The trade-off here is again you give-up a little bit of security, there are attacks that can happen because these child chains don’t have the same level of security. Depending on what you’re trying to do it’s a pretty effective way of again being able to have a lot of transactions, have each of those transactions ratified versus a channel. Then have every now and then the root of that chain has a transaction.

Then the third one solves the problem that we discussed of computationally expensive transactions. The two previous ones were just really how can we do a much higher volume of transactions without having to get them all verified. This one is called true bit and I’ll post some stuff in the show notes about this as well. Essentially, you pay solvers to do your verification. Then there’s challengers that verify the solvers verification and you basically you pay gas to a much smaller subset of computers like not every single node so you can run much, much more computationally expensive contracts by essentially having it verified by a smaller subset of computers.

Vikram: Yes.

Faizaan: Those are the big ones for Ethereum that are in place. It’ll be interesting to see how that plays out because the ones that are layer two are just a matter of people implementing them. It does require a hard fork, there’s a handful companies that are doing different you know their implementations of these. We’ll see how well they work because again you do get lower security guarantees. We’ll see how it works-out.

Vikram: Yes. There’s probably pretty significant architectural, I mean you were going through it I guess but there’s pretty significant architectural gives and takes between each possibility. Are different groups trying to implement them?

Faizaan: Yes. The plasma one Vitalic wrote a paper but yes there are you know a handful of companies that are working on different solutions to this because it’s essentially gives you an infrastructure layer on top of a Ethereum that you could then build application on. Like if I wanted to build a game of chess, I don’t want to have to go implement state channels from scratch. If there’s a company that handled the state channels for me, they’re like a protocol layer for state channels that I can build an application on top of that. That’s how I see some of these plans.

Vikram: Yes. In some ways it kind of reminds me, it’s very different but I am just you know humor me for a second. It reminds me of the whole notion of convention over configuration in web development. Especially around like not having to rewrite stuff that already is in good convention like Rails, and Elixir, and Phoenix, and Ember you know handle the convention really well. There’s other platforms that people use regularly that don’t and they have to re-implement things that they really don’t need to, or I guess they have to because they’ve chosen that framework to write these things. It shouldn’t be the case.

Faizaan: Yes, exactly. Or another analogy would be networking. You have all the different layers from just the hardware, TCP/IP to then stuff has been built on top of that. Like you know HTTP protocols and stuff like that. Just the different layers give you more leverage to build higher level applications on top of it.

Vikram: Yes. Along another second layer side, so Bitcoin there’s a big discussion about the Lightning Network which serves as a second layer on top of the Bitcoin network. Very similar to what we were talking about before. Bitcoin has a limitation in terms of number of transactions per second. What the second layer solution, so you can take two views, you can have one view you want to do all transactions on chain then you’re limited by what the base level protocol supports.

Faizaan: Yes. There’s the proponents of we’ll just increase the block size then if you double the block size you’ve doubled the capacity. The problem of that is, that leads to centralization because the bigger you make the blocks the more concrete like memory and resources the each node needs to actually be able to run a full blockchain. You’re essentially causing like people can’t run it on their computers anymore. You essentially create centralization towards specialized hardware.

Vikram: Yes. What Lightning Network is trying to do is be able to support way more many transactions by doing all those transactions off-chain. There’s a really interesting application out there right now for the Lightning Network, it’s a Satoshis.place, it’s just S-A-T-O-S-H-I-S dot place. I’ll put that in the show notes. It’s basically a little draw/painting web app. In order to put your actual drawing or painting on it you have to pay to do so.

I was messing around with it and I could draw like two letters and it costs about a penny. Forget the actual cost itself but the idea is that in order to participate in the application that you put up some Satoshis for that. Obviously, like what do you expect when you have an open source drawing platform that anyone use, that the Bitcoin community is involved in? What do you expect to be drawn? Penises of course!

I think that was what happened initially and then finally it’s been cleaned up a little bit. Now people are just kind of, I’m just opening it up right now. Some guy named Greg has drawn his name. There’s the word horse. I don’t know there’s a bunch of random stuff someone drew an entire magic card, like from Magic: The Gathering, there’s a MasterCard logo, a nice big Bitcoin logo. It’s kind of cool.

Did you see this happening? It’s a collaborative drawing I guess. Another use cases micro-payments of course. Like two people can open up a channel with each other. Say in the case of like games and they can, as long as they know they’re interacting with each other over the long term they can just pay Satoshis to one another.

Faizaan: One of the biggest drivers for pushing like any technological standards has been like gaming. I think like being able to put games and stuff on the Lightning Network, probably one of the earlier adopters.

Vikram: Where does like a network stand right now? We log all of this Lightning Network. We log the Lightning Network progress on a regular basis and then deliver them as alerts to our users. For example we take snapshots of the Lightning Network at given time intervals. We look at things like how much PTC is locked-up in the network, the number of nodes and channels in the network. Right now there’s 27 BTC worth locked-up in the Lightning Network and roughly it’s probably around 150,000. A couple months ago is around 20 BTC. That’s a pretty big 35% in total BTC locked-up, it’s not bad over a couple months and as the stuff gets out of beta and more into Mainnet. As companies can figure out how to profit off of it these, that number will shoot up dramatically.

Show Notes:

Topics:

• Google Cloud migration to Digital Ocean
• Their bad experience with Google Cloud
• The reason they decided to move to Digital Ocean
• Investor market news in crypto
• Polychain Capital’s one billion dollar SEC filing
• Andreessen Horowitz launched their $300MM investment fund
• An open letter from Chris Dixon
• Spencer Bogart talking about bitcoin/crypto prices being depressed due to forced selling by hedge funds
• What a high watermark is
• Some alerts that QuanLayer’s platform picked up
• Crypto banned in India
• Increase in exchanged hacks (over $700 million)
• Bitcoin removes alert key
• Catching telegram announcements for VeChain token swap support
• The ‘official serious discussion’ that VeChain has
• How ETH network has been taken up by 3 contracts
• First layer scaling solutions: Sharding
• Second layer scaling solutions: State Channel, Plasma, and Truebit
• What Lightning Network is
• Lightning network application

Links

• Form ADV — https://adviserinfo.sec.gov/IAPD/content/ViewForm/crd_iapd_stream_pdf.aspx?ORG_PK=285022
• Fortune article where they interviewed Olaf Carlson-Wee, founder of Polychain — http://fortune.com/2018/06/26/polychain-capital-bitcoin/
• NY Times article on the firm — https://www.nytimes.com/2015/05/03/upshot/andreessen-horowitz-dealmaker-to-the-stars-of-silicon-valley.html
• Kathryn Haun — https://twitter.com/katie_haun
• Kathryn Haun’s bio — https://www.gsb.stanford.edu/faculty-research/faculty/kathryn-haun
• Chris Dixon — https://twitter.com/cdixon
• Chris Dixon’s open letter — https://a16zcrypto.com/
• Spencer Bogart — https://twitter.com/CremeDeLaCrypto
• Plasma — https://plasma.io/
• Sharding — https://github.com/ethereum/wiki/wiki/Sharding-FAQs#what-is-the-basic-idea-behind-sharding
• Truebit — https://truebit.io/
• Making Sense of Ethereum’s Layer 2 Scaling Solutions — https://medium.com/l4-media/making-sense-of-ethereums-layer-2-scaling-solutions-state-channels-plasma-and-truebit-22cb40dcc2f4
• Lightning Network — https://www.coindesk.com/information/what-is-the-lightning-network/
• Satoshi’s Place — https://satoshis.place/

Hey everyone, this is Vikram again. Thanks for listening to us. If you’re
an exchange, a trader or working on a crypto project get in touch with us.
You can reach us on twitter at https://twitter.com/quantlayer or email us
at podcast@quantlayer.com.