Monero Talk: Howard Chu Interview

Howard Chu interview by MoneroTalk

Transcription provided by Cake Wallet for Monero


Speaker 1 (Monero Talk): 00:04
Alright, we're live. Another episode of Monero Talk we are with Howard Chu. You’re one of the core desk, is that correct?

Speaker 2 (Howard Chu): 00:15
I guess you can say that, I don't know if there's actually a title of core dev but yeah.

Speaker 1 (Monero Talk): 00:22
Okay. So I mean I think there's a ton of things that we could actually talk about. I have lots of questions but I guess I wanted to focus on the Random JS proposal that you've recently been discussing. Are you the originator of this concept?

Speaker 2 (Howard Chu): 00:41
Uh, let's see. Well, there was………….. It actually came off of another Reddit thread. So the original suggestion came from someone else, but I kind of took it and ran with it. Yeah!

Speaker 1 (Monero Talk): 00:52
Cool! I guess before we get into it. I know you don't have a lot of times, so I don't want to spend too much time going into other questions, but I can guess maybe just a quick background, I know you have, you know an interesting background in an open LDAP and I, I'm not very familiar with it but I assume there might be some connection to that world and how you got into the crypto world.

Speaker 2 (Howard Chu): 01:18
Okay well, so open LDAP is an open source directory project. It's heavily used in enterprise software. We actually just had our 20th anniversary this past fall, this current fall actually, yeah! And I've actually been working on that project almost from the very beginning, so twenty (20) years on that

The connection into crypto-currency is kind of indirect. I mean, I've worked on a lot of cryptography software before; I've worked on open SSL for several years, and have worked on curb rose and a couple other projects But the connection here came through LMDB, the database central that I worked for open LDAP. And that became the database that the Monero guys selected for the blockchain.

Speaker 1 (Monero Talk): 02:09
Okay. So were you interested in Bitcoin as well before that? Or is that what kind of brought you into crypto?

Speaker 2 (Howard Chu): 02:18
Yeah, I completely dismissed Bitcoin when I first heard about it. I said you know, this is a linear data structure that they expect to have global reach, it'll never scale, it will never work and I just said, yeah! This is garbage. So I completely ignored it and I only got into crypto-currency because of the Monero project reaching out to me to use LMDB.

Speaker 1 (Monero Talk): 02:40
Now, so has your opinion of Bitcoin changed since then? Or how, I mean obviously it's changed in some way but what has kept you kind of in the Monero world versus, you know being pulled into the Bitcoin maximalist zone?

Speaker 2 (Howard Chu): 03:00
Well, you know it's obvious, right? Bitcoin is not fungible and Monero is. For all these projects that call themselves crypto-currencies, you know none of them live up to the name. You know crypto means hidden and currency requires fungibility and you know, none of those other projects amount of that. You know Bitcoin or any others.

Speaker 1 (Monero Talk): 03:24
Yep! Couldn't have possibly agreed more. So I guess we want to jump right into then the, ASIC resistance, maybe if you could give it just like a quick little background on that………….. Why do you think ASIC resistance is something Monero should be striving for? And what is it essentially, what are we trying to achieve with that?

Speaker 2 (Howard Chu): 03:51
Okay, well I think it was actually spelled out in the original crypto-note white paper. The idea is, you know if you look at how a crypto-note started, it's obviously, was written by people who had observed Bitcoin over the course of years, you know and they saw that Bitcoin mining had become heavily centralized in a couple of mining pools, whatever, they are all based in China.

And that centralization was largely fueled by the development of ASICs for Bitcoin mining, right? And so if you follow this in the crypto-note paper, the basic point is when you allow ASICs to dominate your currency, you know, you you've got a very strong force for centralization.

It's not everybody who can access these ASICs, you know the people who are manufacturing them tend to hoard them to themselves preferentially, you know, so there's no egalitarian access to these things.  And so you know that really does violate one of the principles of permission-less blockchain where you want everybody to have the ability to use it. Centralization really harms that principle.

Speaker 1 (Monero Talk): 05:12
Do you feel, do you think Bitcoin has become more centralized over the……………. Is this something that we could actually measure, that we know is actually happening, I mean are ASICs actually having this negative effect of centralizing things?

Speaker 2 (Howard Chu): 05:27
Absolutely! In fact if you look, if you follow Amen Gunser, he's been a crypto developer for quite a long time and one of his writings over the past summer was showing centralization both in Bitcoin and Ethereum. You know they actually measure the number of mining pools, the geographic distribution of them and a couple other factors and you can see that Bitcoin is highly centralized. Etherium is centralized too as well, you know to a large extent but I think Bitcoin more so.

Speaker 1 (Monero Talk): 06:03
Is it fair to say Monero is seemingly one of the most decentralized coins, I mean you know large cap coins?

Speaker 2 (Howard Chu): 06:13
I think at this point probably, you know certainly we've paid the most attention to, you know encouraging decentralization.

Speaker 1 (Monero Talk): 06:23
And now so what do you think long term overall,  I mean I feel like even within the Monero community there's kind of a few different ways of looking at it. Is it that we want to, the goal I guess is, we could all agree is to maintain, be as distributed as possible. Now does that necessarily mean we always need to stay ASIC resistant or are there, is there you know, a future, a possible future where ASICs eventually do come in, but in such a way where we can maintain our distributed nature?

Speaker 2 (Howard Chu): 07:01
Yeah, that's an interesting question. And most people believe the latter, what you just said that………… In somewhere down in the future, you know, ASICs will become common enough and accessible enough that we don't have to worry about resistance anymore, okay? Now I look at that and I'd say the average user or consumer, you know, isn't going to go to the store and buy an ASIC box, you know, no matter how easy they are to get. They're not going to do it. That's just, it's not anywhere on their radar.

So the only way this happens is if, whatever this acceleration-ship becomes, is already bundled into a device that they already care about, like a cell phone. You know, so until that point, decentralization with ASICs is not going to be a reality, alright? So, you know people talk about that, you know people talk about that right now saying, oh! There’s more companies entering into the market building and selling ASICs and so this is already promoting decentralization and I think that's false. You know, because the average miner just the average user, crypto coin user isn't going to be on board with those.

Speaker 1 (Monero Talk): 08:21
Right, so because you're saying because of the nature of the ASIC which is specialized hardware, which it can only really be used for proof of, you know proof of work, that, it's really never going to become a part of our daily technology?

Speaker 2 (Howard Chu): 08:36
Exactly! I don't see that.

Speaker 1 (Monero Talk): 08:38
Now, I think there's arguments that are being made that that's all that will, that's what makes ASICs, I've heard things, you know game theory arguments, that that's kind of a positive for ASICs because if you have, if ASICs can only be used for mining then they could, there's no incentive to use them to harm the network.

Because then it's like you get, you know you don't want to kill the hand that's feeding you, right? So then, if you use them to attack a network and then essentially end up killing the network by whatever it is, 51% attack, whatever. Then at that point, you know game theory wise, what do you, now what do you, what are you going to do with all these ASICs if you can't use them for something else?

Speaker 2 (Howard Chu): 09:31
Sure! Yeah! And that's probably a valid argument in the long run but it's not an airtight argument, you know. If somebody says oh! You know we now have 51% of the network but we don't want to kill the network because it would harm our profits, still once in a while you know big brother might come along and say hey, would you please, you know divert this transaction or censor that one and in one off instances that could happen without anybody seeing.  You know so, yeah! In the long run they don't want to kill it but that doesn't mean they won't poke and subvert it occasionally Mm hmm.

Speaker 1 (Monero Talk): 10:14
Right and its just theoretical………….

Speaker 2 (Howard Chu): 10:16
Yeah!

Speaker 1 (Monero Talk): 10:17
So how is that conversation going then in the community are, you know I, kind of the inner crowd the guys who really understand this stuff and that are kind of really making, you know the policy decisions, where are people at then overall, are, is there, are there people leaning then towards this, we should aim to stay ASIC resistant? Is that kind of catching fire a little bit?

Speaker 2 (Howard Chu): 10:49
I think for the near term that's definitely the case, yeah! Nobody thinks within the next six months through the next year that ASICs will be monetized enough.  So yeah, in the short term, definitely, we're still fighting that.

Speaker 1 (Monero Talk): 11:09
And then, so the next question becomes, and I guess this is where Random JS I think, starts to become part of the conversation. Do we do we hard fork every………….. I mean we, that seems to be the case, right? We're doing these upgrades every six months. Do we fight ASICs off in that manner or is there something else that needs to be done to………..

Speaker 2 (Howard Chu): 11:32
Okay, you know, yeah! So Monero already does an upgrade every six months. That's just normal, right? So what was unusual on this upgrade, on the previous one is that we changed the mining algorithm each time. Now, I believe that's not a sustainable model going forward. You know, we can't keep on doing these minor tweaks to the mining algorithm, partly because it's hard to do these tweaks.

We have to keep them secret almost to the last minute, you know, to the last day before the release to make sure that no manufacturers get a head start and you know there's, it's also easy to make mistake and suddenly we have a mining algorithm out there that has a backdoor or a shortcut that we don't know about. Yeah!

So there's a danger in continuing the approach that we have. Now obviously these first two upgrades, you know the first one, you know, we did as quickly as we could a little tweak that would break any existing ASICs and that was it's only purpose. You know we knew this is a very small tweak and any hardware manufacturer can easily patch it into their next board revision, alright?

The second week is a little more extensive. It would take a little bit more work for an ASIC designer to accommodate it. But you know, overall, you know, neither of these is designed to prevent ASICs coming in the future. They are simply, you know, point fixes to break the existing ones and I mean I don't believe that's a sustainable way to keep developing.

So we're looking for a proof of work that that will last long term. You know like crypto-note lasted for three or four years before, you know, ASICs finally came around, so we're looking for something else that will also last for at least three or four years.

Speaker 1 (Monero Talk): 13:28
So since the let the most current hard fork and looking at that, was there any drop in hashing power there, were there suspected ASICs to even have come on since the previous fork?

Speaker 2 (Howard Chu): 13:42
Okay, there was a drop in hashing power. I don't think it shows signs of ASICs, alright it's just that the new, with the new tweaks the algorithm is like 10% slower on a lot of CPUs, it's five or 10% slower on a lot of GPUs. So there is a noticeable drop but there's nothing like, you know, the drastic decline we had after the previous outbreak where you know, half the hash power disappeared.

Speaker 1 (Monero Talk): 14:08
Right, and it was very evident that there were ASICs, considering they then forked off and created our own coin.

Speaker 2 (Howard Chu): 14:15
Exactly! Yeah! So there's no evidence that that happened this time around. Yeah!

Speaker 1 (Monero Talk): 14:18
Do you keep your eye on that at all; the other Monero coin that's sticking with ASICs? Is that something that's even relevant in the community, that people are looking at to see or is it…………….

Speaker 2 (Howard Chu): 14:29
I haven't really been paying attention. I think once in a while someone will poke up and say hey! I just looked at that and nothing's happening. You know so, as far as we can tell, you know all of those other, you know, the groups that decided to stay on the old fork have pretty much died off.

Speaker 1 (Monero Talk): 14:47
So then what is Random JS? How is this now going to potentially be a, you know more of a long term solution where we won't have to continually upgrade the proof of work to avoid the ASICs.

Speaker 2 (Howard Chu): 15:03
Okay, well, basic idea, you know the reason that an ASIC is so much faster than a CPU or more efficient than a CPU is that it's only doing a single job, alright. And a CPU is designed to do anything, alright? It’s a general purpose processing unit. So, what we want is something that's more complicated, something that's too complex for a single purpose chip to execute, alright?

That's one of the conditions, the other, you know and along with that is something that uses features that CPUs have that current ASICs didn't need to have yet. Alright, so the idea, kind of is, if you were going to build an ASIC for this, we're going to force you to make it into a CPU, alright? And once we've done that, once we've said you need a CPU to be able to perform this proof of work, then that puts everybody on more of an even surface.

And now the thing is any simple algorithm you design, you know, if you can describe it in like a paragraph, or in a page of code that can be truly turned into an ASIC, alright? Any single algorithm that you go with can do that. And, you know, there are other proof of work systems out there where they combine like a dozen different hashes or, you know, they sequentially run different algorithms and in every case, you know, an ASIC can easily be built for that, because you're talking about a small number of, a finite number of algorithms.

And each one of them can be designed and burned into a circuit and then hooked up in any sequence they want, alright? So the problem that we're trying to solve is, you know, how do we come up with something that isn't a single fixed algorithm? And so that's where Random JS comes in to say, well, we're going to randomly generate code, alright? So there is no single algorithm that anybody can burn into a hardware.

Speaker 1 (Monero Talk): 17:12
Now, is this a new concept overall in the, you know, in the computing community or has Random JS been used for, maybe in some other way and some other fashion and the computing world or it’s completely novel?

Speaker 2 (Howard Chu): 17:29
I guess…………. Okay, the, like the code that I started with as the proof of concept was program to randomly generate C code and that was actually developed to test C compilers, alright? To try and exercise them and surface any bugs in the compiler. So this kind of random code generation as a tool has been used before. I don't think it's been used before in crypto-currency proof of work.

Speaker 1 (Monero Talk): 18:00
And so why random JS? Because I guess there's other things I've heard that kind of are similar in concept, but doing it in a different way, like programmable proof of work, Is that a ………………. What would be the comparison there?

Speaker 2 (Howard Chu): 18:18
I actually, yeah! I actually did some discussion with those programmable POW authors as well. So the basic, there's a couple key differences, you know, the prog power guys are focusing on GPUs, alright? So they, and the algorithm that they've got would run poorly on a CPU. Okay, so that's one of our key differences, is that I'm still focusing on CPU support.

The other thing, the most of their algorithm is based on F hash and still looks a lot like F hash. And so the part that does random code in their algorithm is actually fairly small, okay? And they're only talking about sixteen (16) randomly selected instructions or, you know a set of sixteen (16) instructions that can be randomly strung together.

And, you know, if you look at that from a hardware designer standpoint, you say, well, you know, integrating sixteen (16) operations in a chip is easy. So I don't think that what they're doing is really going to deter any ASIC builders. The reason I chose JavaScript is because it's a fairly large language, it's a fairly complex language and to embed the entire language on a chip would be difficult. It would use up a lot of chip real estate.

Speaker 1 (Monero Talk): 19:48
Yes, I mean maybe this is a real theoretical question. So what would happen, would have to happen technology, in terms of technology to reach that point where Random JS would no longer be able to thwart an ASIC or prevent the building of ASIC. On a technology level, what would it, what would we need it to be?

Speaker 2 (Howard Chu): 20:10
First of all, I have to say that there are shortcuts that that we know about now that might render this not quite a viable approach. But if we just take it at face value of, you know how do you accelerate JavaScript? You know, that's a problem that, you know web browser companies have been working on for five or 10 years now.

Okay, so the, I'd say it's a difficult problem on that face. There probably aren't any big breakthroughs that are going to happen, okay? If something did happen, you know, that could actually be a huge win because then maybe we'll see JavaScript accelerator chips in Smartphone’s in five years from now, you know that could be great.

But the fact is, the language itself is one of the more, you know, awkward, ungainly ones, to implement in a computing system. So I don't see JavaScript itself being something that anybody comes out with a breakthrough with anytime soon.

Speaker 1 (Monero Talk): 21:20
And now but you're saying, that you saw there's other issues that we have to worry about?

Speaker 2 (Howard Chu): 21:24
That’s right, yeah! And so we, you know we've been actively discussing this, you know for quite a while now. One of the, one of the issues that surfaces is, well, there is no reason for an ASIC developer to actually execute JavaScript, alright? They could take the random generator and modify its output to just, you know, modify, pure machine code or something that's very easy for them to execute.

Now in our algorithm, we do require a JavaScript source code to be generated and we check for that by you know, running a hash on the output, alright? So if you submit a proof of work solution and the hash doesn't match, then we say, well, you didn't (Inaudible: 22:10-11) The thing of it is, even though you generate valid source code, there's nothing requiring you to execute that source code.

If you generate source code and machine code in parallel at the same time and you only execute the machine code, then, you know you you're kind of short circuiting our algorithm there. And so this is one of the key problems we're focused on right now, is how to do, how to avoid that shortcut.

Speaker 1 (Monero Talk): 22:42
So do you think this is something that could be solved then and overcome or I mean, where do you see Random JS going?

Speaker 2 (Howard Chu): 22:52
Not, it's not really clear right now, it may be that we don't use JavaScript in the end and we actually use machine code ourselves because at that point, the shortcut is equivalent to the main code, alright? So then there's no further execution advantage there.

Speaker 1 (Monero Talk): 23:13
So what kind of support are you seeing in the community for the, you know, the general concept if it can be tweaked correctly?

Speaker 2 (Howard Chu): 23:21
Well, I think there's a lot of, there's a lot of anticipation and interest in the concept. Yeah! Obviously, you know, we need to, if you look at how Bulletproofs got deployed, you know, Bulletproof code was available on Test Net for over a year and then it was forwarded to three different independent auditors, yeah!

And we waited for all three of those auditors to come back with their whole reports before anybody would say okay, let's deploy this on the Main Net. And so that kind of scrutiny is going to happen here as well for you know, Random JS.

Speaker 1 (Monero Talk): 24:06
So will that kind of be next step then, to maybe move something to Test Net or we're not really near that stage yet.

Speaker 2 (Howard Chu): 24:13
We’re not, yeah! We're not really near that stage yet. But when, you know, when we do get something on to a Test Net, then we're probably going to start looking for auditors and you know, reviewers. Yeah!

Speaker 1 (Monero Talk): 24:27
Cool. So what do you think? Any other thoughts on Random JS before I start asking you other questions?

Speaker 2 (Howard Chu): 24:36
Well, no, I think that covers it for now. Yeah!

Speaker 1 (Monero Talk): 24:40
So I guess since I have you, I'd love to be able to kind of hear your view on Bulletproofs now that we have them. I mean, I think it's a no brainer that it's obviously, it's working and it's a great thing.

 But I think there are…………. one of the questions I'm seeing and maybe this is an obvious one but with, how much more if, or the question that's kind of being posed, with Monero’s dynamic block size and now we have Bulletproofs, basically if Monero had the same transaction volume as Bitcoin today and was at the price that Bitcoin’s at today, just trying to quantify and compare Monero to Bitcoin, in terms of, you know, its efficiency and transaction price, you know, kind of an apples to apples comparison, where do you think we would be at?

So if Monero had the same transaction volume as Bitcoin today and the same price as Bitcoin, would our transaction fees be comparable to Bitcoin? I know it's kind of a hard question to answer on the spot. But we’re trying…………..

Speaker 2 (Howard Chu): 25:59
Yeah! That’s an interesting one. I think, the way the fee mechanism works, you know, as usage of Monero goes up, the per byte fee actually decreases. So there's a strong chance actually that the Monero network would be cheaper to use, alright? As far as, you know having equal transaction volume, you know Monero transactions are still larger than Vanilla and Bitcoin transactions, okay?

So that would impose a lot more network bandwidth requirements. And it's not clear to me that all of the people, you know all of the people using it right now have a good enough internet access to sustain that. I don't know the answer to that one. But aside from that, yeah! The per-transaction fee would certainly be lower.

Speaker 1 (Monero Talk): 27:00
Yeah! I think that's certainly an exciting thing and I think that's, it's something that people………….. because with Bulletproofs it was overnight. You know, you woke up the next morning and you went to send a transaction and it went from, whatever it is, 30 cents down to half a cent whereas with the dynamic block size, I don't even think people are realizing, you know the potential benefit there.

I mean obviously people in the know are, but I mean for, you know, the casual crypto user. It's pretty interesting that we're not, that's, I feel like it's not being even talked about enough as to how Monero is built to scale and is made to, you know, the more transactions the better. Which is…………

Speaker 2 (Howard Chu): 27:45
Right, right…………..

Speaker 1 (Monero Talk): 27:46
An important feature to have.

Speaker 2 (Howard Chu): 27:49
Yeah! If you want things to actually be usable, yeah!

Speaker 1 (Monero Talk): 27:52
Do you think Monero is better architected then in terms of scaling than Bitcoin is at its protocol level?

Speaker 2 (Howard Chu): 28:00
Yeah! I do believe that, yeah! You know, simply the fact that we do have dynamic scaling versus a fixed block size, yeah! That's going to go a long way. It's not having a huge impact, you know this week because you know, transactions are so much smaller now that the blocks aren't, you know we don't have whole blocks at the moment, alright? There's not enough transaction volume to build the current blocks. But yeah it's definitely more forward looking, you know more (Incomprehensible: 28:34) compatible, yeah!

Speaker 1 (Monero Talk): 28:36
Yeah! It's interesting, I mean because if, you know in Bitcoin land the meme is kind of like, you know that Bitcoin can scale and Monero can't, which is, you know because it’s, which is, it’s kind of the opposite.

 So what do you think of, you know the Bitcoins approach with focusing on the second layer to where, you know this, for where they'll get their fungibility and also where they'll get their scale with things like the lightning network?

Speaker 2 (Howard Chu): 29:10
Yeah! See that's………. Okay. It's a simple principle of, you know, the software architecture or any architecture actually. And if you don't have a good foundation, it doesn't matter what you try to build on top of it, it's going to collapse.

With Bitcoin and the lightning network, you know, if you imagine that Bitcoin is going to be the currency of the future for everyone, and you want 7 billion humans on Earth to be able to use it, you know, you're going to………… it's going to take five hundred and fifty (550) years for 7 billion people to even create their very first transaction and open the lightning channel.

Okay? So saying that second layer scaling solutions will solve Bitcoin’s problems is clearly false.

Speaker 1 (Monero Talk): 30:00
Can you like, can you…………Yeah! Can you………………. I saw you tweak there, can you explain that a little bit more? Why is there, why would there be such a slow onboard time?

Speaker 2 (Howard Chu): 30:08
Well, it's, it's simply that, you know, every channel that opens has to be part of a Main Net Bitcoin transaction, okay? And so, again if you're looking at, you know, one megabyte blocks and one block every 10 minutes, you know, there's always so many people you can enroll in 24 hours or in 24 years. You certainly can't get 7 billion people on board all at once.

Speaker 1 (Monero Talk): 30:38
And then so what is the reaction to that? I mean that seems like a very plausible issue?

Speaker 2 (Howard Chu): 30:48
Well, there's certainly a lot of denialism. And I suppose there's some belief that not everybody needs to open their own channel. I'm not sure how that works, but you know, I guess they assume that, for example merchants will open a single channel that handles multiple customers and, okay maybe that works but to me it still comes down to every individual who wants to use it has to have their own payment channel. They have to be able to open it somehow.

Speaker 1 (Monero Talk): 31:23
Now, what do you, how about for Monero? Do you think Monero should be looking at things like lightning network as well or for the reasons you're talking about here, it's not really even something that should be considered?

Speaker 2 (Howard Chu): 31:38
I suppose there, you know there's probably no harm in it. Lightning network on top of the narrow would probably function better than it does on top of Bitcoin.

Speaker 1 (Monero Talk): 31:50
Yeah! Can you explain? I've been asking that question as well. I'm curious what your reasoning is as to why that might be the case

Speaker 2 (Howard Chu): 32:00
You know the people, people in the Bitcoin community talked about lightning network, not only solving scaling problems but also solving it’s privacy problems. And, you know, that's not really true. Again, because you can monitor the coins entering and exiting the payment channels, alright? Whereas in Monero, you know, all of those Main Net transactions are still opaque. So you can't actually see coins moving around.

Alright, so it's a better privacy solution on top of Monero than it would be on top of Bitcoin. Now, you know, if you look at Monero usage today and the network today, we don't need it yet, you know, we don't need that kind of a scaling solution yet, because the current network is doing just fine.

Speaker 1 (Monero Talk): 32:51
And then so about, so in terms of privacy, I get it. So it's, we're going from a privacy at a core protocol level to now this second layer, so it's only going to help. But then how about in terms of the architecture as well because I mean to run………. what I don't understand is if you're running lightning network on the Bitcoin network and the idea being, you know, the lion's share of transaction volume will now be taking place on the second layer, how is mining going to continue to secure the blockchain if there's no incentive to mine, especially without something like a tail emission that Monero has?

Speaker 2 (Howard Chu): 33:35
Yeah! That's a really good question. Yeah! Yeah! If the majority of transactions are happening over lightning then, right, miners aren't going to see those transaction fees and obviously when Bitcoins emission ends, you know miners are going to have no reason to participate at all and at that point the network is dead.

Speaker 1 (Monero Talk): 33:55
Yes. So I mean people in your echelon, I mean what are the……….. How are people responding to that, you know?

Speaker 2 (Howard Chu): 34:06
Again, you know, the only word that comes to mind is denialism. I think people aren't even looking at the problem.  Yeah! I mean if……….. and I suppose in some ways, you know, even if you're a Bitcoin developer and you're aware of the problem, you know, your hands are tied, right? Because you have this contracts, this, you know community contract that says, “This is what Bitcoin emission is going to be.” And you can't just go and change it now.

Speaker 1 (Monero Talk): 34:38
Right, so their hands are tied behind their back.

Speaker 2 (Howard Chu): 34:41
Yeah!

Speaker 1 (Monero Talk): 34:43
What do you think about MimbleWimble? Grin is implementing it, which, you know, doesn't yet exist, but…………………. Then, but it's also being talked about, I've heard it being talked about as potentially being used in conjunction with Monero as like, on a side chain. Are these things you've ever looked at or considered?

Speaker 2 (Howard Chu): 35:06
I haven't really looked deeply into that. You know, I've obviously heard people talking about that. It might be a good idea. But if you look at what MimbleWimble really offers, you know, they claim that you can have a very lightweight blockchain because you don't have to keep historical blocks or something to that effect.

And the reality is, that's not entirely true. You know, somebody needs to be archiving this stuff. And the other part that goes with this is, there's nothing preventing somebody from doing a full archive node and if they do that, then they can de-anonymize what's going on the network.

So MimbleWimble kind of, its privacy depends on the fact that everybody is printing information all the time. If somebody decides not to print, then they can actually break the privacy.

Speaker 1 (Monero Talk): 36:09
Yeah! I would like to do a show just on MimbleWimble. Try to kind to kind of get to the bottom of that. What are you, how about Grin? What is your feeling of Grin? I mean that’s obviously just an implementation of MimbleWimble. But do you think this is kind of a project worth looking into? And…………..

Speaker 2 (Howard Chu): 36:34
Well…………….

Speaker 1 (Monero Talk): 36:35
It's getting a lot of attention.

Speaker 2 (Howard Chu): 36:36
I would say it's interesting to look at the technology. You know, I have nothing to say about them until they've launched them on the internet.

Speaker 1 (Monero Talk): 36:48
Gotcha! What do you think of, this kind of, this other recent claim that's been going around that, you know being pushed by the Bitcoin maximalist, basically that Mone…………. something like Monero requires more trust than something like Bitcoin where everything is transparent and you know if there was some kind of emission bug or something, it would be completely obvious in Bitcoin as opposed to in Monero where it may go undetected. So therefore they're being more trust in math and trust in the developers. Do you think that's, there's any credit to that?

Speaker 2 (Howard Chu): 37:34
Yeah! That doesn't seem like a valid argument. You know, if you look at the………… you know there have been inflation bugs in Monero code and there have been bugs in the Bitcoin code, and the Monero bug was detected before it was ever exploited on them, yeah! Which was a good thing.

I think the point here is, you have to know what you're looking for, if you claim we could spot a bug like this faster on the Bitcoin network, alright? Clearly, you know, they didn't spot this one. And it's all about finding the bug in the code, you know they didn't see it on the network. Alright, so I don't believe that the network transparency or opacity is really a factor here.

And yeah! It's just, it's all about, you know, is there a bug in the implementation because we, we have to trust the math, alright? And the math is sound in both of these cases, aright? So it really comes down to, was there a bug somewhere? And that's the same in any code base.

Speaker 1 (Monero Talk): 38:55
Right, so whether it's Bitcoin or Monero, you're trusting the math and you’re trusting………… There's some trust in the developers as well who are implementing it. There's really no difference. The math just might be a little bit more complicated in Monero.

Speaker 2 (Howard Chu): 39:10
Sure! Yeah!

Speaker 1 (Monero Talk): 39:13
So do you think there's any place for a coin like Bitcoin that is, you know, transparent at its core? Do you think there's kind of a necessary, there's utility there, it's serving, it will eventually serve a purpose that Monero can't fulfill?

Or is it that if you built this core protocol like Monero and it works and it becomes large enough that it would be able to kind of, not just do the privacy and fungible thing but if you want, you could back step and also, you know, do the revealed, you know make it transparent as needed?

Speaker 2 (Howard Chu): 39:55
Yeah, yeah! I mean Monero has optional transparency, and I think that's the right approach. Bitcoin and its derivatives and all the forks and clones based off of it, you know, they are, none of them behave like real money.

You know, really what they behave like, is a big surveillance machine. And, No, I really don't see a valid use case for them anywhere. But that's just not the way you want your money to behave.

Speaker 1 (Monero Talk): 40:31
Now, how about from the standpoint of being a protocol? because you obviously are quite familiar with protocols. Do you think this is a case where it makes sense for there to be one protocol that wins? That, you know, one protocol on the internet that will be used for transferring value?

Speaker 2 (Howard Chu): 40:57
That's an interesting question. I think ultimately, yes. You know, it's like, if you look at the way the Internet has evolved, you know, you can look at E-mail protocols, you know, there were two or three competing ones back in the dawn of time and now there's just one SMTP and that's used everywhere.

And I think, yeah! for interoperability, for usability, you really only want to have one single protocol at the end of the day. It may not be Monero, you know, it may be some new one that we haven't even seen yet. But I don't see a valid use case for like five (5) or ten (10) running in parallel all the time.

Speaker 1 (Monero Talk): 41:47
Do you have any thoughts on whether or not, you know something like, you know Bitcoin because it's, you know obviously has a first mover advantage and maybe because it's more acceptable in, you know, the communities of governments because, you know the fact that it's transparent, so therefore maybe it won't be as susceptible to regulation; that maybe something like a Bitcoin has more of a chance of surviving for those reasons.

And it's a good enough protocol, so even though it's not the best form of, you know, money transfer we can create on the internet, that maybe we'll just have to live with it and that's the one that ultimately wins for those reasons

Speaker 2 (Howard Chu): 42:40
I don't think a first mover advantage is any kind of guarantee, you know. I mean, just today I was reminded of the existence of MySpace, you know when was the last time anybody thought about MySpace, right? Yeah! First mover is not a guarantee. And as far as government regulation, that could go any number of ways.

You know the, if you look at what's been happening in Europe with the GDPR; the General Data Protection Regulations, you know, there's been a much stronger emphasis on consumer privacy in Europe, alright? And if you put the current crop of crypto coins into that environment, you might find that, you know, Bitcoin and others won't pass muster for consumer privacy protection and a coin like Monero would.

So, you know, it's not obvious that when government regulation enters the scene that Bitcoin will be the obvious choice, you know, it may not be.

Speaker 1 (Monero Talk): 43:48
That’s an interesting take on it. Yeah! I think everybody was kind of, has been looking at it from the flip side perspective that it's more friendly to governments, but I guess it depends what government you're talking about. And more progressive governments that are actually, you know, taking a stance and starting to more highly consider, you know the need for privacy and for……………. In this digital age may actually be more open to something that's protecting privacy on a protocol level.

Speaker 2 (Howard Chu): 44:22
Right……….

Speaker 1 (Monero Talk): 44:22
That's an interesting take. So are there anything else that you're excited about in terms of development in Moneroland that's, you kind of see on the horizon, things that you're working on? What is it that you kind of Google and follow and what's your, other than the Random JS, what else is it that you're very interested in now in Monero?

Speaker 2 (Howard Chu): 44:52
Well, okay. The, I suppose the other things that occupy my time, you know obviously I spend a lot of time thinking about database work, and I looked at what's coming out in new generations of armed CPUs, new generations of AMD CPUs and just looking at power efficiency, you know, performance per watt and those sorts of considerations. Yeah!

Speaker 1 (Monero Talk): 45:18
What do you…………… What's your take on the, you know, these proof of work coins use too much electricity and if we're going to destroy the environment?

Speaker 2 (Howard Chu): 45:28
You know it's, that is something that kind of haunts me. You know, I sit here and think gosh! proof of work kind of sucks. It would be great if we had some other mechanism to protect and secure the network. I just don't know what that is yet.

Speaker 1 (Monero Talk): 45:50
And so, yeah! Even with the Random JS because I know, I, when you mentioned that, you know, if we eventually get to the point where ASICs kind of crack it, does that means there'd be a development in the JavaScript script world, which would be a, you know, an additional benefit?

So do you think……….. Are you thinking in those terms then with proof of work as well, like could it serve an additional use or is that the nature proof of work that it's just meant to be a big waste?

Speaker 2 (Howard Chu): 46:25
I think, yeah! It's the nature of proof of work that it's just meant to, you know, consume resources. You know and that's the point that, it makes it expensive for an attacker to try and subvert the network, alright? because you just got to deploy so many resources to be able to launch an attack. If you have a proof of work that has another useful function that starts to distort how strong it can be at protecting the network.

Speaker 1 (Monero Talk): 47:02
Okay. Alright, that's really all I got. I mean, I crammed in a lot of questions there. Thank you.

Speaker 2 (Howard Chu): 47:06
Okay, okay……………

Speaker 1 (Monero Talk): 47:07
And I know it's late where you are.

Speaker 2 (Howard Chu): 47:09
Yeah!

Speaker 1 (Monero Talk): 47:10
Thank you for, you know, coming on the show. Thank you for taking your time. Thank you for doing it at this time, I know it’s very late. Sorry, I was a little late, I was running from work. Really appreciate your time and your thoughtful answers.

Speaker 2 (Howard Chu): 47:26
Alright, thanks for having me on.

Speaker 1 (Monero Talk): 47:28
Thanks. I'll talk to you later.

Speaker 2 (Howard Chu): 47:29
Yeah!  Bye.

Speaker 1 (Monero Talk):47:30
Bye.

***********************************THE END*********************************