Last Thursday (5/24) I finally managed to sync up with Jihan Wu for a videochat. We had spoken to each other briefly at LABitConf in Buenos Aires about a year and a half ago, right after Zcash launched, but we hadn’t communicated since then as far as I remember.
Here are my notes from our talk. Anything that’s a quotation, I got from my notes that I was typing as we talked. Everything else is my own paraphrase or memory of what was said.
I started by telling him that we as the Zcash community have to decide what to do about potentially changing the Proof-of-Work, and that there was a serious “trust gap”—a lot of people are highly suspicious of, and angry at, Bitmain. I said that I myself was annoyed at him (because they hadn’t given me a heads-up about their new ASIC Zcash miner) but that I wasn’t as deeply suspicious and hostile as most people, but that a lot of people—including people that I need to work with in order to maintain and improve Zcash—are very suspicious of Bitmain and that this “trust gap” prevents cooperation to produce value for everyone’s benefit.
Jihan said that they are in China and there is a large communication gap between them and the West, so that they don’t know how to communicate to Westerners. I agreed that this communications gap is a major reality, since that I have the reciprocal problem: I don’t know how to communicate to people in China.
Jihan said that a specific group had taken advantage of this communication gap to spread a lot of lies and conspiracy theories about Bitmain (note: by telling you what he said in our conversation, I’m not endorsing it. I’m just telling you what he said), and Bitmain didn’t know how to counteract those lies over the communications gap, so they had adopted a policy of just not engaging, being quiet and just releasing their products without saying anything. For example, he said, we don’t even say that the Z9 Equihash miner is for mining Zcash!
I said that a large fraction — maybe 60% or so — of the XMR hashrate had disappeared from XMR and switched to XMO, and asked if that was Bitmain’s mining operation. He immediately replied “No, it was not us. We released the Monero miner right after the samples.”
For what it is worth, I didn’t get the feeling that he was lying about this. I felt like he was telling the truth. I asked him again if they had been doing any Monero mining with ASICs, because it really seemed like someone must have been, and he said something like that he saw the theory that there was a large-scale stealth ASIC miner, and that he hadn’t had time to look into it to tell if it was real or just a hypothesis, but that in any case it wasn’t them. Again, he sounded sincere to me.
I said that the hashpower that stopped mining on XMR appears to have moved to XMO, and that XMO is sponsored by Bitmain/Antpool. He said Monero Original contacted them and they provided the sponsorship, but it is not something that they are actively seeking.
He said “We have never been doing a stealth mining strategy.” He said that the reason they decided not to use stealth mining is that they wouldn’t be able to keep it secret. He said they have 2000 employees, and a large-scale mining operation would require hundreds of employees to be involved in the whole process.
“Is it written down anywhere — your policy about this?”. He said no, it’s not written down.
I asked whether they use a pool for their own mining operations. I said, assuming that their mining operations are sufficiently large, they might not even need to use a pool at all. He said that was a good point and technically correct, that a large enough mining operation wouldn’t necessarily need to pool, but that they point their mining operations at their own pool.
I asked “Are you mining Zcash right now?”. He replied “Uh, are we mining Zcash right now? Yes, I think so because there is already some sample miner produced and we have also opened the service for mining Zcash in Antpool right now, and we have a few miners right now in testing. And right now in I think one week or two our first batch of ASIC miners will be shipped.”
Again, I had the feeling that he was telling the truth here.
I asked if they operated GPU mining. He said they don’t have a very large GPU mining farm, they may have some but it is just for a development use case for their own mining pool.
Then I said, I’m thinking how to say this respectfully, because you are a much more successful businessman than I am, but may I give you some advice about how to start rebuilding trust? He said yes, please.
I said, okay I think of three things. First of all, re-engage with communication. I’m well aware of how badly that went last time, with Twitter flamewars and all that. I’m not saying it is easy or I know how to do it, but I believe it is possible to do it and get better results, and I will give helpful advice if I can.
Second of all, commit to policies in writing. In a large and well-organised Western company, if there were some issue that had important consequences, they would write down a policy saying what the company will and won’t do, precisely, and post it for both internal and external people to know what the company has committed to, and they would explicitly update it if the policy changes.
Third, transparency. Bitmain will have to be much more transparent than a smaller and less important company would have to in order to get a similar level of trust. So, I advise you to disclose to the public a whole bunch of information that might be uncomfortable or problematic and that a different company wouldn’t need to reveal, such as:
- Corporate structure — who owns what, who controls what, financial statements, who profits
- What’s your policy about stealth mining, exactly.
- Start publishing a whole bunch of statistics, like once a month or once a quarter, about how much hashrate your mining operation is running, which blockchains it points at, which pools, any policies about what sort of transactions/blocks, etc.
- Disclosure about non-production (dev and test) mining
He said he would talk to his team about doing these steps.
And, I said, you should really show up at Zcon0! I think that being face to face with someone and looking in their eyes really makes a difference. Probably a big part of the reason that I’m not quite as suspicious of you as most people I’m working with nowadays is just the fact that we met in Argentina and looked into each other’s eyes for five minutes. So I really think it would be worth showing up at Zcon0. He said he would definitely try to make it.
I said that I wanted Bitmain to contribute back more to the community. I said that it wouldn’t work at this stage to donate money. Any donation, such as to the Zcash Foundation or to sponsor developers, would be perceived as a bribe or buying influence. But, I said, maybe they could actually do things for Zcash, such as integrate Zcash support into the BTC.com wallet.
I asked, is it true that the BTC.com wallet is now pre-installed by default on all Huawei phones! That’s amazing if true. He said he had heard something about that but wasn’t sure of the details and that he would ask about it.
I asked is BTC.com wholly owned by Bitmain. He said yes.
I asked who are the owners of Bitmain. He said he owns around 25–28% of Bitmain, and that the other co-CEO, Micree, owns more than he does. And that there are three other natural persons who are shareholders, an option pool for employees, and that there are a few venture investors who collectively own less than 5%.
I asked if there is anyone who can fire him. He seemed confused by the question. I explained how in a typical Western company, like the Zcash Company, there’s a board of directors who can by majority vote fire the CEO and replace them.
He said oh, no there’s nothing like that. I said well could Micree fire you or vice versa. He said his shares and Micree’s shares each have 10X the voting power of anyone else’s shares.
The conversation was in that “we’re winding down” kind of state at this point, where people throw in things that they almost forgot to mention.
I said I hadn’t decide whether to publicly mention that we had this conversation at all, or what to publicly reveal about what we had said in the conversation. He said it was totally up to me what I wanted to say — I could keep it confidential or make it open. (I appreciated that, I thought it showed honesty/transparency.)
I said Oh yeah, before I go: Bitmain is valued at $10bn, correct? He said yes. (Bitmain is one of the four cryptocurrency decacorns — $10bn companies — Coinbase, Bitmain, Upbit, and Binance.)
I said, Bitmain’s revenue in 2017 was $7bn, correct? He said no, it was actually $2.5bn. I said is it okay to publish that number, too, and he said “yes, it is already a public number, even if it is not widely noticed. It is a public number.”
As we were saying good-bye, he said “Zooko, thank you very much for the help.”. I felt like he was emotional about it, like he was sincerely grateful for the conversation. I said something like I haven’t helped much yet, but I hoped to.
POSTSCRIPT: shortly after this conversation, but before I wrote up these notes, Bitmain started posting these reports on twitter that they term “radical transparency”:
This seems like a big step in the right direction! Transparency about orders and shipments of the Z9 aren’t one of the things that I asked for, but I definitely would have if I had thought about it — it is very central to the burning questions about whether Bitmain — or anyone — is running stealth ASIC mining, whether they mine at scale with their own equipment prior to offering it to customers, how much of the total Zcash hashpower is ASIC vs. GPU, how concentrated the ownership of the ASIC miners is, etc.
I still want to see the other stuff that I asked for, especially the total hashrate owned and operated by Bitmain for different PoW algorithms, which chains/blocks/txns they target with what part of that hashrate, etc.
I look forward to seeing what The Zcash Foundation (which is separate and independent of me and of the Zcash Company that I lead) comes up with from its investigations into ASIC resistance. For now the Zcash Company’s position on ASIC-resistance is unchanged from our previous statement.