The Thing with Privacy is

well it is a terrific panel uh we are here to talk about privacy and we have legitimately some of the world experts who have been trying to figure out how privacy uh and uh sorry how crypto can help actually enable better privacy here to lead the conversations corey petty from status uh but i think we're gonna have an amazing conversation with uh people here from zcash and orchid etc so uh corey i'll let you take it from here yeah um privacy is clearly important um especially the decentralized space as we try to attempt to minimize the gap between those with power and those if not um and and how much influence they have how much the top have influence on the bottom uh i've tried to put together a panel to talk about some of the implications of privacy as um more often than not people talk about how much it's needed but those of us who are trying to implement it in a myriad of ways have run into difficulties both with our users and in the technology itself so uh today we have a nice panel um so i'm corey i'm the cso of status i'm also a founder and host of both the bitcoin podcast and hashing it out and um if we can bring our panel in i can give them a quick moment and introduce them uh if you want to know more about me look me up i'm everywhere uh we have dr stephen waterhouse just want to give a quick one-minute introduction as to kind of who you are so i want to spend most of time on content yeah sure can you go see america yep okay great uh yeah hi um stephen waterhouse uh aka seven um co-founder and ceo of orchid labs uh we're building privacy technology um ethereum and uh our first product is a decentralized vpn uh previously i was co-founder and uh cto and partner at pantera capital and then before that many different careers and building startups just helps how about you hello my name is justin holmes i'm at new cypher i'm uh the ranking unlicensed plumber over there uh and i occasionally play some music that's quite a few talks and i always appreciate the flare music of these in these uh generally kind of bored talks about about the crypto cryptography and privacy so uh we have a few more they may join us later on down the line but we can go ahead and get started um and out the rack's kind of the first question to you justin but um feel free to jump in uh stephen if you'd like to and i guess the first concept i'd like is is like as we keep building better and better privacy technology it's getting to the point where um it's inevitable uh but and and our users are going to have much more power in their hands um to do the things they'd like to do without uh being beholden to those who'd like to surveil them how does that suck what's what's bad about that i i i'm glad that you started there because i think that we take it for granted that the advances in privacy technology and specifically cryptographic privacy technology are going to be viewed with uh open arms and optimism and i think that that's often not the case largely because uh the appearance is um and this is quite reasonable that privacy appears as an abating agent in the flow of data from users from minds to the corpus of human knowledge in other words there is a widespread belief that we are moving toward a society where all of human knowledge will you know be wonderfully available for future generations and if only if it weren't for those privacy folks you know we'd have a more clear picture of who humans are and i think this is distinctly wrong-headed chiefly because it ignores what i regard as the main role of privacy which is to preclude the self-censorship that comes sometimes in the early days of a good idea that prevent that idea from making its way to wider and wider circles of society so i think that um the the the weakness is partly in in the way that it's sold and partly in in the way that the technologies are used but by and large i think that it's important for us to uh make inroads into communities with a reminder that uh in fact the self-censorship that comes from fear uh in an absence of privacy is a big part of what keeps some of the best ideas from reaching their full potential i want to welcome yang to the to the panel uh you can kind of jump in after you kind of start the next section or next question but stephen do you have anything to say about that yeah uh it's sort of an interesting counterpoint to this assumption as he said that we everything gets better if we're more private um [Music] i think as albert from union square ventures gave an interesting talk at block stack many years ago in berlin and he gave this kind of point where he said that we actually need radically less privacy as science and technology improves because what we should be scared of is the 15 year old in the basement building the next buyer weapon by mistake or perhaps on purpose and releasing it out to the world and so his his sort of amusing counterpoint which was said someone in jess but was something to like a thought experiment to think about as well maybe we should be surveilling people more not less um i disagree with that but it is it is kind of a thought experiment to think what are the boundaries um and how should we separate the lines between mass surveillance and target surveillance and what is the role of law enforcement in that kind of world where um you know radically improved privacy as possible um i mean i'll take a little bit of you with with your initial statement which is that i think that um when it comes to targeted surveillance of individual the kind of tools and the kind of exploits that can be gained over your devices are significant and more significant than we even realize um so certainly tools like our the ones we're building which which attempt to prevent mass surveillance um are helpful we hope we're helpful um but the sort of target surveillance in an individual by an agency is still very powerful and probably hard to resist at this point i could i could agree with that kind of perspective yang the question was um like privacy is getting better it's inevitable we're giving more power to the users to um disclose themselves the way they'd like to with the platforms that they use but what sucks about that okay um hi my name is zing tong that that is my first name um what well that's a good question um yeah you can come back to me let me think about that yeah sure you can always come back on that one move to kind of a a similar type topic uh and that is this is i think we keep we as developers have to build this um and we have to build it for what we think are our users but in the process of introducing more and more privacy for our users we are relinquishing a lot of the power we have to understand who our users are and what they're doing so that we can maybe make decisions on behalf of on behalf of them based on the data they give us in the process of using our using our technology so as we start to remove our ability to surveil as is a correct word but it's maybe a negative connotation as we remove our ability to understand our users to their behavior uh like how do we make decisions for them and uh i'll direct that to whoever like to pick it up i guess again kind of the thinking process here is is the the kind of the the business models the internet are built around being able to track people and and you know give them the things that we think they want and in many cases they do um whether they're on the platform that they're on or whether just a more general broad internet um so yeah i do think it's a challenge it does it does disrupt the business models and could even disrupt significantly the uh the actual ability to personalize things i think that [Music] with careful design and use of things like zero knowledge proofs there's ways to disclose information that doesn't require the disclosure of everything about you in some kind of one massive swoop right um the question is whether or not they kind of come back a lot of us to this question a lot is so how do we motivate people to want that like they already have something that works like this idea that like oh let's build a decentralized social network and everyone's gonna just want it it's like well it turns out they just really don't care they just want the thing that works so there needs to be something more compelling i think um i mean that's a separate conversation but yeah i think that uh there are ways to do it but it's complicated yeah i think it's a great question about um how do we know what our users needs are and how do our users communicate them to us i've i've joined the zcash community for maybe a year or so and i see that community engagement um through you know github through public forums through the discord they're still really active channels and there's a lot of lively debate and discussion and people do bring up features and um issues that they're interested in so for example um i think in the horizon for the z cash protocol is um transparent z cash extensions which extend the functionality of bitcoin script and allow people to basically uh work with more general preconditions for their transactions um and a project called bulb labs is building zero knowledge payment channels using this extension so that's for for instance one feature request that came out of community engagement and then there are other apps like zbay for example which is private messaging using the memo field of z cash and yeah like basically keeping in touch with the community and making the decision to engage is it's just good ux and it's just good project management like regardless of whether this is a privacy focused project or not but i think corey the point that maybe you were starting to make was that if the projects are sufficiently private then we won't even know water who they are and thus what or who the needs are is that kind of where you were saying yep that's a good part of it yeah yeah i mean i think like you have to understand i'm sure this comes as no surprise to other layer two people who are engaged in layer two engineering that right now that's not a practical problem right now we're still um trying to find the sweet spot to even inspire the beginnings of user engagement we probably have i would think forecast another at least year or year and a half of things uh being such that um a lot of our user engagement is happening in hackathons and in places where people are wanting to be seen but at a deeper level your question is profound and wonderful what do we do when we if success means that we can't know very much about our users and and i think it that is one of the things that it means then what do we do and i think that the answer to that at least mostly is that we look at society and understand the if our project can be used is being used and we think about it being used to help existing marginalized groups marginalized communities then probably we're doing the right thing and probably more affluent communities will find ways to use it also um i you know the idea that people are need to be which seven this is a good point and we think about all the time of course people are sort of insufficiently motivated to seek out these features like a private social network but i think that's also because um we are not really ready as a society yet our our technological infrastructure hasn't matured yet to keep track of the really serious secrets that people have uh if people are there many people in the united states have essentially a boolean flag on their personality called undocumented and if that's true they can be arbitrarily capriciously hauled off and put in a cage that's i i a completely unacceptable situation and i think that if we're building privacy technology that allows a group with with such an incredibly marginalizing feature to use it successfully then that's a good sign even if we don't necessarily know and can't know the particulars of how it's being used it being used in contexts like that i think is a strong signal that it's a useful technology and will be useful to many strata of society i can agree with that kind of the concept of understanding that it is being used in conjunction with yingtong's uh concept of of having active engagement outside of the use to get feature requests and um actively asking people for what they want uh helps us understand better products to build yeah and i do want to underscore seven's point about the prospect of using some kind of zero knowledge uh you know the the accoutrements of zero knowledge in order to facilitate engagement support is is actually a wonderful point i i 100 agree um in fact most of the people in here have experience with zero knowledge proof at least in terms of this panel barry why don't you give yourself a quick introduction and uh add anything to that if you'd like to hey friends yeah so i'm barry i do a lot of um work with like uh privacy and scalability um using ckp's and fraud proofs and other things i've also been doing some like fun mechanism design stuff more recently which is also pretty interesting um yeah so just as to answer the question the last one that you asked so the last one you asked is like oh so if we have this like utopia or this future dystopia whatever where everything is very very private we won't know what users want us to to build or what they need and like i think i kind of land a bit withing tongue on this that like don't worry people will tell you like people are very upfront about what they want from their things that they're using um but like the difficult and the challenging thing is to find out what they really mean when they say what they want that like some people will say oh i want this and this but that's not necessarily what they really want like for example tornado cache is like this thing built on ethereum that's really successful and it's basically a kind of a mixer you put your coin in you can withdraw it and there's there's no way to connect the deposit to the withdrawal um so this is really successful and and like a lot of people yeah sorry i didn't really know where i'm going with that i think my point was like something along the lines of like like what's happening with tornado cache yeah yeah it's pretty i think bringing up tonight is actually very interesting um i think the interesting aspect there which i've been looking at is is the the concept of this anonymity liquidity mining which they added and the challenge of the mixer right it's like how do you get enough nodes in the mixnet and it was very interesting they sort of essentially added in define incentives to an anonymity mixer and oh this is like again we're kind of tangential here but i thought it was like a really fascinating thing i've actually been looking at really carefully recently yeah so i'm actually so i'm not really convinced about the anonymity incentives i have a link about a polls where i talk about it but in the chat so basically the poll says that like like these things have already been tried we already have these kind of incentives so they weren't really as big as they are with tornado cash uh in zcash zcash used to force miners to mine into the private pool but all of this all of this mining was was actually removed from the anonymity set because it turns out that the miners would just instantly withdraw their funds to the same address in the transparent pool as soon as they had been earned so you were able to sort of gather together all of these people and remove them and like i i i think that the tornado cash liquidity anonymity mining is an interesting experiment but i think that we really have to be careful about about the anon i think it's interesting experiment and the i'm kind of waiting for some deeper analysis about how this affects the anonymity it actually springboards beautifully there's something else that i wanted to cover uh we've both already basically brought up topics that i wanted to cover one being uh sevens mentioned that users don't care we can get back into that in a moment but to continue this conversation like in the process of of implementing privacy in in web 3 technology um you have this consequence that information tends to stick around for a really long time and it's very well structured and so you built this corpus of beautifully structured data to mine and as you continue to use these products regardless of the level of privacy you're releasing more and more information about yourself and so you have this weird concept that as you use this these products more and more and more your the ability to create um or disclose more and more information about yourself specifically on privacy previously private aspects about yourself and maybe de-anonymize yourself in some ways gets gets more and more prevalent how do you combat that or or think about that when you're adding features to something that gathers all this data into a single place or a few places that interact with each other i'll let anyone who'd like to pick that up pick that up i'll take a crack at it uh one of the ways that at new cipher that um that we address that issue is by separating the cryptological actors uh such that the actor who decides who gets access to the resource which is alice is not the same actor as the actor who encrypts data for the resource we call that one enrico and it's one of the big advantages of threshold proxy re-encryption and when we sort of got our head wrapped around that it was kind of exciting because we realized that uh alice the authority of the policy can sort of go to sleep or delete her private keys um but i i don't know i tend to think questions like that i tend to think um solid engineering you know and you know we uh we need to just a big thing is we need to continue to attract our usership through solid engineering that's easy to understand and i mean it's an ongoing challenge but i think it's the one you know that we all signed up for and we're having fun uh rising too yeah i think new cypher has a really cool it potentially has a really cool ux like anyone who wants access to that data can just request that it be re-encrypted to their key right uh that is one possible that's one possible scheme uh alice the original authority of the policy needs to be online you know needs to use her private material in order to grant that access so in the scenario you're describing you you have some kind of an alice oracle okay any other thoughts on the matter it's something i've been worried about um because we add more and more features um that allow you to just like to give yourself options to disclose yourself more about yourself and in a model like ethereum that's account based uh it's all kind of around a a single entity so you may enter into the ecosystem with a lot of anonymity but as you use more and more surfaces you build up a much much larger corpus of information about yourself which means which means that you remove what you previously thought was anonymous about yourself in the process of using features like what do we do about that because like can privacy tech mitigate that and if so how and and when does it come about so so like there's sometimes when it just can't do anything for example if you live in like a very low population time zone uh for example i think greenland has like very few people in that same time zone and you do a lot of stuff during the business hours of this place and then you're is this the kind of thing you're getting at that there's like when i'm making my transactions i always make them at 5 pm greenland time and there's a very small set of people doing this and it's able to you're able to kind of locate me because this is the thing i'm doing repetitively is this kind of what you're getting at kind of but it's more like uh yeah or like timing attacks or like even even just say for instance i have used an account to do activity i put funds through a mixer and that's recorded on the blockchain and at some point in time i decided to use the new ens feature and get a name attached to that address uh the the concept of me having privacy attached to the funds that i moved into a mixer is now gone because that's recorded on the blockchain and tied with my ens name if i forgot to not use the same account yeah yeah so you're like yeah so this is okay so that's an interesting point right because you you are these yeah it's really difficult to okay so we can make the best privacy technology in the world but the users have to use it well like one of the big problems that people make when they're using z cash or they're using they're using something else some other privacy coin is a deposit like 1.156 ether into the pool and then they do some stuff and they would draw a later 1.56 to either from the privacy pool and like that's that's just it's obvious that they are the person who deposited because they they were they were the person who withdrew because they were the only person who deposited so there's like this whole different difficult like education problems that you need to solve in order to in order to to kind of complement these private systems which is not trivial it's enough well to go off of that um you mentioned that uh it's an educational problem which it very much is we have to as we relinquish power as as people building products to the to the users uh we have to educate them so that they can make the best decision in the context that they that they would like but in the context they're presented with the information they have and so as developers we kind of have to um not only make our user interfaces intuitive so they can make those decisions but educate them about the things they should care about because more often than not they're not aware of it and with privacy related features at that that that includes making sure that they understand the disclosures they're making whenever they're making them what are we doing in in that in that realm how do we think about that especially as as seven mentioned users don't care until it's usually too late well that education is a two-way street we need to be listening about what people don't want to disclose and why and um you know with a with a careful ear and uh you know careful commits and code reviews uh ensure that the things that those things are um evident you know when that that exposure is evident the scenario that you painted with a mixer and uh ens domain is a is a really solid one and um you know i mean i i think like as you sort of talk to yourself through it it's pretty obvious that those technologies uh if if indeed we want privacy related to ens domains that we there's uh there's a re-engineering component that needs to be done um you know i'm uh i i i am interested in hearing uh the sensitivity is there and uh about exactly how and why and when um what information would we even want to be available and when you know it during the acquisition of an ens domain is sort of itself an interesting uh privacy question privacy of course has forever you know been contrasted with uh public public and private have been essentially contrasted with one another of course they both are latin words right this is a thousand-year-old notion of uh of things working together and um obviously an ens domain is is by its nature a public entity it's something that i'm going to type into my browser to my terminal and so what is the uh what is the private role there uh these are good questions and uh you know the the oil and water of of public and private are are sort of endemic to the conversation at this point which is a good sign um but uh you know i don't know i know it's an extremely roundabout way of answering your question i think because there's no terrific answer for it today except to say that uh you know these are all issues that belong in a bug tracker somewhere anyone else i think seven is frozen or is that just me or is he's frozen in thought i'm going to assume frozen in thought because he hasn't moved yet uh so kind of like i mentioned like in in the beginning zero knowledge proofs and the development of applied cryptography is growing and growing and growing um to the point where like we're getting to the we're getting to an area where we can make systems like systems of rules that allow you to work within the system without disclosing anything about yourself uh what does that mean for like i i guess the best way to answer this is like any tool can be used for good or evil and it's up to the developer to make a decision based on the potential consequences of whether or not to implement that thing and they have to think about that stuff in the event that everything is private is it is is the is it overwhelmingly good or over overwhelmingly bad because it's the potential for these things isn't necessarily equal how do we weigh uh what can be done with the technology we're building uh with like what we want people to do with it i think um i think if you just look at like kind of the some of the exploits and defy that are happening and the speed at which these things happen the kind of uh so there's some of these articles about the dark forest and memphis and so on and so i think it's pretty naive of us to hope that um by some kind of good intention of a developer things are going to work out okay i uh i think that there's some really amazing technology being developed [Music] and a comment i've made for many years now is just because we have blockchains and just because they're going to be private doesn't necessarily mean it's going to be good it's a lot of like dark architectures you can build of like very centralized control centralized digital bank currencies and so on that will have all sorts of great privacy tech in there and it'll seem like that decentralized but they really will be more about controlling you than they will about giving you freedom so i think that um [Music] you know a lot of the thinking we have in the space is that the technology will create the society and i think we need to make sure that we don't take that too far and remember that societies create society and then ideally they use technology to create better societies but if we just assume that the technology is going to take over and create some wonderful society for us and we don't need to get involved in that we may miss something really important yeah brilliant man i am i feel like i'm kind of dominating the conversation here i'd like to invite the others to step in if they want um but failing any unmute's happening i uh i i don't know okay so let's i'll play the i'll play the naive the naive role that you described for a moment while and that's not to say that i disagree with you i don't but i also think that it's helpful to look at the longitudinal positioning of technology and society and uh find the cases for optimism and i think there are many of them uh this this gathering that we've had here uh denver generally is um filled with just there's obviously an attractor of some kind some kind of curvature of the space-time of wisdom and knowledge that is bringing a lot of good ideas here um and and i i think that the what what tends to happen is at least for me and and i i think this is probably a widespread feeling is that um when we think those dark thoughts like you were you know elucidating there about the prospects for uh information deprival being used to control us um what's really happening there i i think is a muddying of the waters about what we're really talking about so if the the title of the panel is the thing with privacy is then i think we need to identify that the thing with privacy is that it is a micro institution of the self and of the the body and of the user and when we start to see entities that are beholden to the public when we start to see the commons filled with notions that uh you can't record that or see that or hear that or think that that's not actually privacy that's that's a different thing that's secrecy for secrecy's sake or secrecy for the sake of secret police and uh the thing about privacy is that it sort of has this walled off notion as i was saying last time in contrast to the the publicus um and so it's not to say that i would say you know oh we could leave it to the developers and as long as we develop good technology that it will develop society i agree with you that societies develop society uh but i also think that um we are best positioned to try the best we can to lift each other up with with a degree of optimism i mean the the pro we haven't even really talked about the progress of the last year in development of of our code bases um uh you know yours and mine obviously have gone through leaps and bounds and um the the enthusiasm over the deployment of novel cryptographic primitives in a decentralized way i i think i don't know you you don't necessarily call it new and novel but it's um a cause for optimism i think and it's okay to operate with that optimism as long as we also um are thinking critically about what we're developing and why yeah i agree it's good to be optimistic too and i am i agree with that i think that's a that's a great kind of parting thought but i'd like to allow others um to kind of give a quick parting thought to to the audience as we wrap this up yington do you have anything to kind of um reflect on or have our audience reflect on yeah i mean to end on a high note um there's a lot to be optimistic about and um what i'm excited for is more universal snarks are like for instance plonk is being widely used by lots of projects um including zcash and basically that gets rid of the trusted setup which was a big part of many um in production protocols and perhaps a source of uncertainty for some um and i think we're seeing advancements in uh usability scalability of privacy tech and off um privacy protocols so i'm excited for this technology to become more and more useful and scalable in the coming years very much agree barry do you have any thoughts for us yes i do i think that this is a really exciting time to build new things um one of the things i'm really excited about seeing is uh private social media um so there's some approaches to this using kind of snark-based constructions where you where you get like reputation based upon the posts you make and you can give other people reputation you know in a private way so i have some other posts about this which i'm going to share i hope that gets transferred somewhere so that everyone can see it but yeah it'd be exciting to see people uh working on this this weekend um it would be cool to i'm excited to see the projects that come out of east denver uh i was there last year and there was a bunch of cool stuff that came out and i'm just very excited to see more so thank you thanks corey of course is happy to have anything to wrap up with um i'm just i guess i'm just really excited to to you know see all the energy that goes into the space whether it's it's in privacy focused things or non-privacy focused i think that i i've been in this space since 2013 and i'm just it's it just gets more and more interesting every year whether it's a bull market or a bear market or we're not quite sure what kind of market it is it's uh it's wild and i i feel um i feel this this you know like listening to justin and barry and new car just this it's kind of like a sense of kinship amongst people when we're all focused on this stuff and i think that's really powerful and so i i agree with justin we have a lot to be optimistic for because i think there's some really good people on this stuff so that's that's what i'm into thank you all for joining me i appreciate it and i really enjoyed this conversation wonderful to be in y'all's company yingtong i don't think we've ever met in person but uh maybe next year denver the rest of you all of course we've met in person in person we've been doing personal things like that i've heard of that thank you so much it has been a really long time uh yeah since we've all seen each other uh thank you so much corey for hosting that that was a brilliant discussion i think we just learned that privacy seems pretty simple but it's really really really complex glad we have some brilliant minds on it