text stringlengths 0 1.82k |
|---|
**Erik St. Martin:** The UTXOs are kind of interesting because basically when a transaction occurs, you basically absorb the UTXO that they had before and then generating a new one to go to the person that the transfer is going to. |
**Tess Rinearson:** That's right, yeah. Every transaction uses up UTXOs and creates new UTXOs that can be used in a later transaction. |
**Erik St. Martin:** And the general idea of that is the double-spend problem, right? |
**Tess Rinearson:** Right. One of the things that is interesting or tricky is, like you said, this double-spend problem. The double spend problem would be like if I claim to send the same five-dollar UTXO to both Carlisia and Brian, that would be a double spend of that UTXO, and that would be invalid. So that's one thi... |
**Erik St. Martin:** And now the other implementations you're talking about, I think there's some that just kind of use a ledger and no UTXOs... |
**Tess Rinearson:** Yeah, I actually don't know much about the mechanics of non-UTXO blockchains, but certainly there are... I mean, like I said, a blockchain is one of those things where if you meet a certain number of criteria, you call yourself a blockchain, and one of those things could be a UTXO model. So there ce... |
**Erik St. Martin:** I guess from a generic perspective it's a distributed system that maintains some sort of ledger that they validate previous transactions in order to confirm a transaction taking place, and many parties have to go through their own ledgers - whether they're using UTXO or something else - and determi... |
**Tess Rinearson:** Yeah, that seems fair. I mean, I would say that also there's an inherent transfer of value in them, but not everyone thinks that, too. I think when you look at a blockchain, you can almost think of the assets (Bitcoin, or whatever other asset you have issued on a blockchain) as tokens of value, and ... |
**Carlisia Thompson:** I have a question. The Chain Core - is it meant to be used by financial institutions that are brokering initial transactions? Or is the idea that there will be a hosted version of blockchain somewhere and the financial institutions will use that? |
**Tess Rinearson:** \[08:00\] Right, so we're working with these financial institutions to spin up a variety of different blockchain networks. In the case of VISA - VISA is running a blockchain network of Chain Cores, and all of their clients have nodes in the system. You asked about a hosted blockchain somewhere - gen... |
That said, we also have this thing called the testnet. Sometimes I joke it's the TessNet, but it's the testnet. \[laughter\] |
**Brian Ketelsen:** It's all about you, Tess. |
**Tess Rinearson:** Yeah, I know... That's a term that we also borrowed from Bitcoin, where there's a Bitcoin TestNet that sort of runs parallel to the actual Bitcoin system and is reset at fairly regular intervals, and basically just has fake money on it. So we have this Chain testnet, and it's validated by us and Mic... |
**Erik St. Martin:** And now you said most of your customers are using this for financial stuff where they're tracking bonds or actual cash, or things like that? |
**Tess Rinearson:** Yeah. What people are using it for is any asset that holds value. That can be stocks, that can be currencies, that can be credit card points, frequent flier miles - all kinds of stuff. |
**Carlisia Thompson:** It sounds to me that an entity that could benefit from Chain Core wouldn't necessarily have to be a big financial institution, or even a small financial institution. It could be a business like LivingSocial, if they want to use it for loyalty points, or Groupon. Is that right? |
**Tess Rinearson:** Yeah, I think, potentially. We're definitely starting with financial institutions, and we're starting with financial institutions that are well-known enough that they can power their own networks and people will want to join them. That serves the power of this partnership. We just on Friday announce... |
**Erik St. Martin:** Yeah, and I think anything that you want validated to ensure that it hasn't been tampered with is also a good candidate. |
**Tess Rinearson:** Yeah, I think that's less of a problem that we are solving right now. Obviously, when you're looking at things like validating UTXOs, you have to make sure that transaction history hasn't been tampered with, and things like that. There are so many cool things that you can do with a blockchain... |
\[12:04\] If you look at something like the Bitcoin blockchain, which is this public distributed ledger where everyone has a grade on the history of it, you can do crazy things like prove that you have copyright -- or, I shouldn't say "prove that you have copyright", because that's probably technically-legally somethin... |
If I wanted to prove that I wrote a paper, I can hash that paper and stick that in the metadata of a Bitcoin transaction and then put that on the Bitcoin blockchain. And if anyone's like, "Oh, did you actually have this paper at this time?", I could just point them to that transaction and give them the hash function an... |
And because in Chain's case we're not talking about public networks, we're talking about private networks, you lose that ability. But there's other benefits to having a private network. |
**Erik St. Martin:** Right. |
**Carlisia Thompson:** And when you're generating keys, who is actually validating those keys? Is Chain an authority to validate those keys, or is some distributed system...? How does that work? |
**Tess Rinearson:** Yeah, every node in the system can validate every transaction, and then there are certain -- I don't wanna get too much into the mechanics of all of it, but every node in the system can validate transactions, and then when it comes to actually creating new blocks and adding those new blocks to the b... |
**Carlisia Thompson:** Who has control over this federation? |
**Tess Rinearson:** In the case of the testnet it's us, Chain, and Microsoft and IC3. We are the three signing nodes in that network. But that's configurable on a network-by-network basis, and actually in the protocol we have a field for the consensus protocol, which can be changed on a network-by-network basis. For te... |
**Carlisia Thompson:** Cool. |
**Brian Ketelsen:** So what's the application for GopherCon? That's the real question here. Can we start making chains of things that we can give out to gophers and let them trade value? We can have like gopher points... Trade them in for plushies. |
**Tess Rinearson:** Yeah, that's definitely a thing you could do. You could have every attendee at GopherCon have their own asset that's issued by their... One thing I didn't really talk about, but if you're familiar with Bitcoin or you're probably familiar with mining - which is used for consensus, but it's also the p... |
\[15:58\] So what you could do at GopherCon is every attendee could issue their own asset, tied to their own key pair, and then maybe for each person you meet, you can issue them one token, and then whoever has the most tokens wins, or something. |
**Erik St. Martin:** I say we go bigger. We call GitHub, and contributing in Go gets you -- that's how you mine, by committing in Go. \[laughter\] |
**Tess Rinearson:** Oh, yeah... |
**Brian Ketelsen:** Interesting. |
**Tess Rinearson:** Yeah, I feel like every year at GopherCon... I make all these big plans to do some open source work, and it hadn't happened until Monday, when we open sourced new stuff. But I definitely think GopherCon gets me in the open source spirit, and then generally I fall through on that. |
**Brian Ketelsen:** Well, you delivered big for this year, so you've done your part. We need to take a quick sponsor break, but when we come back, remind me to ask you about how this differs from the blockchain that IBM announced earlier this year. |
**Tess Rinearson:** Okay. |
**Break:** \[17:03\] |
**Brian Ketelsen:** Alright, so how does your blockchain differ from the one that IBM released? I saw a press release - it was really early this year - that IBM had released some sort of business-oriented blockchain. Is that different? I think there was Go code and Java code... I don't remember all of the details. |
**Tess Rinearson:** Yeah, I don't actually know exactly what they're working on. There are a lot of different blockchain systems out there, and I tend to be pretty heads down on what we're doing here. One thing definitely - and I don't know if this is the case with their blockchain, but definitely one reason why we ope... |
One thing about blockchains, like I said, is there's a lot of them. With some of them it's just hard to know what's really going on beneath the surface. There's a lot of press releases or white papers, but not too many in-depth protocol documents and even pure implementations. So one reason why we went open source is j... |
**Brian Ketelsen:** So you're telling me blockchain technology is a little bit like vendor management solutions for Go - there's a lot of hype and not a whole lot of action? |
**Tess Rinearson:** It's a hot subject, to be sure... \[laughter\] |
**Erik St. Martin:** I think anything academic like that is. Before Raft, trying a distributed consensus - it was mostly white papers and things, and you're like "This sounds cool, but how do I build it?" |
**Tess Rinearson:** \[19:55\] Yeah, totally. It's funny that you bring up distributed consensus, because obviously that's a huge part of blockchains as well. There's so many papers on different consensus algorithms, and a lot of blockchains are using PBFT (practical byzantine fault tolerance) as a backbone, if not actu... |
**Erik St. Martin:** What did you end up using for your consensus protocol? |
**Tess Rinearson:** We use something related to PBFT, but we basically have - again, without getting too into the nuts and bolts of it - a single node that we have been calling the generator node, although we've just started discussing changing that to the block proposer, that is responsible for gathering the transacti... |
In the case of these financial networks, that's actually not a bad thing, because if some institution in the network is going rogue, you have a bigger problem than the state of your databases; you have sort of an organizational problem that has to be dealt with offline. So the fact that we use that as our consensus mod... |
**Erik St. Martin:** So by putting it out there open source, were you hoping for contribution, or is this mainly to get people using it and coming up with use cases...? |
**Tess Rinearson:** Yeah, we're not looking... It's this funny thing - we are not looking for contribution in the same way that a lot of other open source projects are. It really is about communicating ideas as much as we can. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.