TG ZKP Task Force Meeting - 2026/08/25

Just waiting for Scott who's probably coming over from that other call. Feel like we just got a an hour and a half today to talk about the KPS. It's It's happy day for me. >> There you go. That's uh that's good stuff. Here comes Scott. Yeah, some of us are just coming or three of us coming from the credential editor's call. Um, yeah man, there are just so many so many things on the docket. Um, okay. I'm I'm I'm just here back more in listening mode um to uh Yep. Just a moment. Need a uh AI system on my computer where I just say, "Hey, share that screen. Fire this up." Why am I doing this by hand? Can you see my screen? >> Yes, you can. Now, >> all righty. Uh, welcome everyone. >> We are joined by Sakshan. That's awesome. as a new new member. >> Highly highly active member. >> Yeah. Yeah. Yeah. Yeah. I know you weren't expecting to be able to make these calls, but great great to hear from you. How are you doing? Uh in lie of that answer, [laughter] hope you're doing well. wanted to call out the antirust notice. Um I will be taking notes today and we just came out of I think as Drummond mentioned a cross functional task force session with the credentials task force. So we'll have a recap on that today. Um kicking off the recaps overall since last week. Um version 0.4 of our requirements doc is merged. Uh thank you Srashan for that. Uh we also just came we came out of that credentials task force call um which gives us a clear steer to get concrete on how we're going to work together but also Drummond's Drummond raised a point on developing a prioritized list of the of the actual proofs that we think we need to solve for and the why on that so we can start to triage them and actually make it real uh and storing that in a repo uh that we can all work together on. Um, uh, Glenn posted a strong delegation design note and the Berkeley team is connecting with us targeting September 8th. Uh, did you have anything to add that? Oh, you did, I think, Mitch. >> Yeah, I wanted to share something really quick. Um, oh, wait. Oh, it's only one. I do have permission. We're all good. Okay. Um this is just like a lot of uh like editing note and something that I am happy to share with people. Um, a lot of the there's like lots of content in the git discussions which I really love and I'm enjoying the direction of having these sort of like pathways of obviously curated by AI but a way for us to then like compute everyone's context at once and then provide a reply from our perspective being the ZKP task force. And I'm happy that there's like permissions and um in place for that to be possible in this community. So that's great as a note. And the followup to that is I'm just going to share the system that I've developed for this like discussion posting human in the loop framework. Um and see if it's helpful. So it's the ZK task force post viewer. Um and it's a pretty much just a copy of the different requests that are relevant to me or I've been requested on. Um but the interesting thing here is that it uses like a proverb protocol where um it summarizes the posts as like three doors were built on a yes not yet spoken say it where the builder can hear and then we go to this issue and that is a compression of my response to Glenn's um thread on um the ZKP task force's answer. So I'll just go back to that proverb again. Three doors were built on a yes not yet spoken. Say it where the builder can hear. And that's what my agent encapsulates my perspective of the response to Glance um thing. And I actually think that that could be a trust task that we could use in this community for like verifying that we've read and audited the things that our agents are eventually going to keep like posting and discussing. Um, this is more like of an editing note. It's not as related to like ZKPs in specific in the purpose of the task force. But as you can see here, there are some that like I don't completely agree with and it's optional that my agent came up with and I'm not going to say this to the group, but I will say these ones that I've mentioned where witnesses who cannot collude may still each lie. Keep the two dangers counted apart. And this was replying to one of the issues that a um asynchronous contributor Dundry made and it was actually folded into the spec by Sanka and his his agent, but I wanted to make sure that there was a reply and then we socialized that in in the call. Um I'm going to pause there. I think you guys get the point. This is the one on the Cirrus catalog. So I'll just say that as well because Drummond was the one who proposed that conversation where a thing came from and whether a stranger can remake it a different truth. Let neither stand in for the other. So So this is pretty much the response that I provided in the previous call, but that response is and Dennis agrees with this response. it's like um validating his comment as well that the [clears throat] circuits that were provided by Cybos are mostly like presentation artifact based circuits for when someone signs something. Um and if we're going to allow people to like construct their own constructions like the proof of livess 4 document, we're going to need like a bit of customization on the circuit. So we shouldn't just let them replace each other. we should um entertain both ideas is is kind of what I'm trying to socialize here which is then written in the discussions as a comment. Um and then this is the other one that I haven't posted and it was basically the update that I provided last week. I'm not sure it's relevant anymore where like there was some moves um away from using ladder space PQC cryptography towards hashbased. Um so yeah I will stop there. a similar document was or HTML document was created in anticipation for the call just then and all of Glenn's issues that have occurred in the last 7 days um which I kind of use um as as tooling. So I'm happy to share this with everyone in the group if it's going to be helpful. And I like the way that um Glenn also had this like big HTML document that was produced to summarize the um work that was done. So yeah, we can keep practicing that. >> Okay. >> But yeah, um that was more of like a comment statement style thing. So we can continue with the agenda. >> Awesome. So coming out of the call we just had um yeah the uh the idea the the key message from that cross task force uh discussion we just had um we can all agree to stop assuming ZKP fixes issues kind of wholesale uh and instead start getting crisp on issuing a real ZKP for a real credential. Um and Glenn uh put us on the path of doing so with open vCT open VTC uh that something that we can integrate and try. Um the spec architecture that emerged from that conversation keep the credentials and trust task specs clean and minimal. Um there will be a VTI spec that can span all of them and it can show how to compose the pieces in a privacy preserving way. Um and we will get our own ZKP spec. We did not confirm to Drummond that we're ready for that repo, but I figure we are. Uh, keep >> I think that's something we should discuss. I, you know, it was Yeah, I'm not sure. Um, cuz I think it can actually fold into the task force work initially um, while we're still experimenting and sort of maintaining that list of proofs and recipes. So, I'm I just going to make a simp simple suggestion. Um, if we're going to since I have to request these other repos anyway, um, it's not an it's not an issue if we have one there for you to begin for this task force to begin to use for a spec at whatever point you decide you want to use it, but you don't have to start using it. Um Um, I I was just going to >> Okay. >> Yeah. Might as well just get it. >> So, the providence is there. That makes sense. Yeah. Yeah. And then then you can uh as things go along and you decide okay yeah we are going to uh um I'll clarify one thing there's no task force can exist whose deliverables is not a spec right so that's that's fine you in the end you can decide no we don't need a separate spec it's just going to like you said uh Mitch that everything go in the trust has spec or whatever. Um [clears throat] but this way we'll just have it covered and uh um again you don't have to do anything with it. Um but it will be there so that we don't have to um slow down if uh you get to the point where you go oh yeah we we're going to we do want a spec we want to start uh working with that repo because there's fair amount of work that goes into just getting it set up. So and once it's set up then it's easy to use. Cool. So, I think we would like that. Thank you. >> Yeah, but no pressure on the timing because we have a space to work currently, >> right? >> Uh and then some two two concrete things that we can rally around. Glenn will be writing up the first real ZKP with tight constraints for open VCC uh creation and a presentation and he's aiming to deliver that tomorrow. Um that'll be our first concrete integration target and we can align around it and develop it. Uh and then the other item was Drummond's point he raised about a prioritized proofs list. Um a living markdown of the proofs we need in priority order with real world utility. Uh noting what's already implemented live, what's conceptual etc. Uh and then the highest value proofs that generally need ZKP production livveness and biometrics included. Um and then Mitchell jumped on and said each entry becomes a recipe essentially. So we can take it from evaluation and try to reconstruct the circuit from the recipe ideas. Uh and proof of livveness itself is a really big proof. So several trust task proofs are are smaller predicate wins to get there. Um Berkeley we are targeting meeting them September 8th and we'll be feeding them questions ahead of time. Uh, and I have a note to myself here, but I think we'll develop those materials, Mitchell and I, and we can share them with the group to make sure everyone feels like we're representing the questions we want uh, in anticipation of that call. >> That's that's the call with the Berkeley Berkeley team. Is that >> Yeah, they would be joining this call, I believe. >> Yeah, they'd be joining this call on September 8th, right? Yep. >> Yeah. Yeah. Yeah, >> cool. >> Yeah, the the only caveat to that is there may be a change to the timing. I don't know how possible that is or we should just kind of force them into the the timing of >> if just from a purely logistical standpoint. if it's decided that a you know for that call there's a better timing or or you want to you know set up a special call for that it's not a big deal. Um there's we we have it down now that um if you want it on the calendar is just as long as like 24 ideally 48 hours notice at least you just uh send a uh an email to support uh LFD centraliz.org and uh you just request you just say, "Hey, we need a uh a meeting scheduled for this time." You tell them the name of the meeting and and uh who who they want to invite. It'll just go on the calendar so anyone can come. But um so I I I guess I'm saying if if there is an issue on timing and you want to set up a special call, um that's all it really takes. Okay. I think that's still on the cards depending on um Sanjim's reply. So follow up with that before September. Um like Yeah, sometime next week we'll follow up. >> Yeah, it's sam. >> Yeah. Um yeah, cool. Um it's J. There you go. Send up. There you go. Um, yeah, unfortunately Hart will not be in Geneva uh next week, so won't get any in-person time to talk with him about any of this. But anyway, we can catch up with all of them. >> Cool. And then Mitchell uh for the construction selection inputs. Mitchell and Dennis. Uh before that SC could we somehow mark that like call 8 of the September to be like to have some special not etc. It's a calendar so it will be you know like it will been like we will not forget at least me I will not forget that Bartley is supposed to be join >> I think before I'll I'll write a note to send you a DM cool thank you >> I mean like I could set the reminder but like if that call could be changed in terms of the time and >> yeah if if we make any changes socialize it with the call. >> Awesome. >> Uh do you want to take the screen? >> I think so. I Let me double check this one. I looked at Zkit. I think it's it's the same answer um as the Cybers one. Like each of these open source tooling kind of have their own place. The my agent had a problem with CKIT because they haven't had the circuit verified or something like that um since 2024 and the way that we would do a construction is slightly different to the way that it was outlined in ZKIT. And um there was a few notes around that that I can share and I I have like a discussion post if we want to explore that further. But I think that's actually a good item to like encapsulate in this um recipe document uh that has been proposed. So I will probably instead of exploring that today use that as material for when we get that sort of recipe and meth methodology going with um Glenn's first use case we can then be like okay well zkip can be used for this part of the use case but it's not useful for that one and then same with the cyos um circuits or the long fellow circuits and then same with the um the other circuit like the the growth one that that I produce. So um >> the the the long fellow circuits it's like totally another thing you know like so use circuits based on the circum and like this zit like actually aim the cir the cir aim the circum circuits you know like circum ecosystem circuits uh zip circuits and and mostly that's it like it just to avoid uh agent uh row uh scripts which you have in repo. So your agent wrote like dummy not dummy but you know like just like simplified version of the script to compile the circuit etc etc and here we are at least human written so at the end it's the same thing but you know it's human written it like it's tested not just once by agent like another people use it blah blah blah but overall it's the same thing but uh since you mentioned the long fellow and and like another ecosystem which does not which not a circum is totally another thing you know like the kid does not suit it your scripts in your repo does not suit it long fellow it's overall another story it's super special circuits uh etc etc so it's not there is no silver bullet here it was just my proposal you know like just to avoid self-written uh >> scripts yeah to compile the >> yeah yeah I think that that's what it's a good practice like if our agents are going to hand roll this stuff we should work out at least what some of the human versions were originally and then try and work out that as and I I actually think that should be a part of the recipe book is like we have this reference to Zkit and the points in the scripts and then we're like okay we we reference that and then this was the agent's hand roll of the circuit and then this was the original one that's verified um and run um prior to to that being possible Um, so yeah, I think we agree um on the appropriateness of Zkit. >> Can I I I think I missed something along the way. What What's the providence of Zkit? ZK kit. Um, is that >> Yeah, make sure you It's ZKit because Zkit is something completely different. It's another um interesting red herring. Uh so it's uh well Dennis provided the reference but it's um like an open-source repo for compiling a a CKP circuit. >> Okay. The open source repo for that. Okay. >> Yeah. >> Um got it. But I would say yeah. Yeah. Yeah. if we use it in the way that could be anticipated um like we would maybe even want to be a part of the maintenance of CKit like I'm not sure it's a like fully maintained um >> I was going to say do we know much about the providence of that repo how active it is you know how old it is what you know is it >> uh like we know it's not it's not Uh it's not the shiny coin [laughter] in terms of this stats but you know like in like if not there uh instead of that we have just self-made you know like uh in scrappy way agent scripts for doing exactly the same. I mean like means means if we will not do that we will we will build such so eventually we will end up doing ex exactly the same thing by ourself and then maintaining by yourself so like this is this is another coin which we're moving so far >> got it >> very helpful thank you >> but but but but but totally you know like valid points like it's not shiny it's not shiny coin Frank is saying like there is not ton of the stars in the GitHub etc etc so But at least something which you know like I I know I used in the past and I I I and I know that another people also you know like using that [snorts and clears throat] but that's it. No no no push or anything around that. Yeah. [clears throat] Uh cool. Anything else on that? >> That's all good. >> All right. So, the next item was the backlog and design time decisions. Um, two came out of the cross task force work that we would want to take a position on. whether the verifier performs no live lookups, no live lookups becomes our privacy profile default and whether delegation folds into the design time sweep. Now uh and we still have the semantic backlog B3 to B10 um to open as threads. There's a whole topic being presented on delegation and I think our input is going to be assumed. So that may be something that we don't lead with if that makes sense and instead we reply and um ratify as a task force open to opinions on that. Um that was my first take on that question and this actually gets the heart of what Glenn was talking about. Um the verifier performs no live lookups. Like live lookups is practically what um is helpful for the kernel for like knowing that this is a maintainer. Um but that live lookup if protected behind dead zk becomes this like house with no names which I would actually think that it's an interesting framing where um one side tries to build this house of no names and the other side tries to build this like public verifiable um graph and then the one in the middle emerges naturally that takes like a little bit of both. Um that's again another sideways thought. Uh but the verifier performs no live lookups would be a boundary that we would set for the house of no names using that new terminology. Um but we would then forgo if some of the trade-offs and the design decisions of the VTI is um in favor of transparency over privacy um being the pendulum. Uh >> VTI sorry I don't know why I said that. I'm uh I'm going to put um a link in the chat to that to to Glenn's um post in case uh Dennis hasn't hasn't seen it or or because it as I study it, there's a lot in there. The these um metaphors about the house with no names and what's going on there. Um there's a lot of info in that. Yeah, that's the uh I I Yeah, >> the works going from writing to reading. It's um reading is becoming a first class. >> Yeah. >> Activity. Okay. Uh, anything else on this item? No, I would just like reiterate the the git discussion system that I mentioned earlier in in this part of the notes >> and like I can share the HTMLs of those as well in these note documents which would then like show that I said the thing in the call. Um, Yeah, I think it's kind of front running a future issue, but it could be a good structure to to put in place early on. Okay. Uh on the open items, we're still tracking Dennis's uh probabilistic sampling prototype. Um, you walked through it last time. Do you have any more any uh I guess anything new to share about it? >> Uh, frankly saying not. Uh, I mean like I did not uh like I did not update the anything from last time. So that's why the answer is none. uh just to just to remind the numbers which I which I like highlighted the last time but let let me do that one more time. So in such approach with the sampling uh uh like and according like to the mathematic formula which like behind that uh for proving for proving around like several thousand of the credential like this connection credential to gain the trust. It's enough to pick uh around like 29 or stuff like or something like that uh credentials to proof for the verifier and uh and this my prototype have the proof uh like I mean like there is a dart like uh flatter based application where you could run that that is integration test to see the numbers on your site on your uh like iOS simulator or you could connect the real device and see the real numbers on the real device. But uh that thing is saying that at the end the holder could generate such proof in route in around six like till 7 seconds. So this is this is a approximately range to prove uh yeah at the end is to prove around the 29 uh to generate around 29 like proofs but like at the end goal like we kind of uh prove that like the user has a thousand of such credentials. So this is the thing and again that circuits have every requirements from dramant um uh first person project document like the the like so it's have it solved all the requirements which is dram posted on the first person project documents like so like it should be unique uh issues it should be like each issue should have the separate VCN should be and all all of that should be proven in the ZP manner uh that that issue have the identity credential issued by some predefined to trusted issue etc etc. So yeah, [clears throat and cough] >> cool. >> So this is the time just to just to have some numbers for us, you know, like to to rely on and understand like what it could be in real world. [clears throat] >> Yeah, I actually think that percentage so like 29% of credentials or >> 29 from So it's not 20. No, no, no. uh it's so it's concrete number for let's say four 4,000 credentials that like the number of the credential which need to be picked up is 29 accord uh if we use 95% of the confidence and 10% as a as a threshold for the best fraction but this is a little bit like technical but >> oh yeah yeah that's what I was trying to yeah >> just wanted to provide some you know like some concrete numbers just not tell you know like just to have some concrete numbers in our like communication >> I think the the point I was um going to make is >> bad fraction and 10% threshold of bad fraction it's got I'm saying for you but >> oh sorry can you say that again >> uh add fraction 10% threshold of Bad fraction >> of bad fraction. >> Fraction. Mhm. Yes. This is a >> it's like an error, right? >> A band name now. Bad fraction. That's cool. [laughter] >> This is this is you know like this official math statements like I believe it's it's correct. I think uh the point I was going to make, I think these percentages or these run um time and proof sizes are actually really important numbers for us to include in the recipes as well. Um and if we can ma mirror and um make the probabilistic sampling uh rhyme with the trade-off choices that are inevitably going to fit in that document. Um I think that will be help explain to people when they are making these choices of the composition of proofs. Um yeah more clear if they if you can be like well like there will be this if you maintain this error rate we should be able to still prove the graph at a 95% confidence in in under 10 seconds. Um that then becomes a decision that again the V uh TI's uh what we could even be the VTN's like the networks that um actually deliberate on. It's like a governance call um in terms of the those thresholds and those numbers that are appropriate for for that trust graph. The other thing that's related that I've worked on um is this concept of interle. Um so you you like create a proof which is half of two people and you interle them. So you give a percentage of like how much I'm willing to be public. I might be willing to be 70% public in this relationship but in another relationship I want to be 30% public. And there's a way to have two hashes and you can choose this is a little bit outside of CKP but it's like more on the hash cryptography side but uh I the word important here Scott is called like a visibility ratio where like how much of the hash you can see is the choice that the person wanted to um share as a disclosure of the nature of the relationship. ship. So for public commitments, I would happily share the entire um hash because it's a public commitment and I want people to know that the whole hash is reconstructable. I can I can prove it all. If it's a medium to private affair where it's I am happier with 50%. Then I would rely on the other party 50% to sort of reconstruct that um as a presentation. So, it's a little bit related actually, Dennis, to Dennis's probabilistic sampling. Um, so I'd like to just leave a note to join those two thoughts together. Um, >> can I ask I I can you explain this idea of being how you could be 70% public uh in a relationship? What does that mean? >> So, it's okay. I need to add a clarifying point. to the words that are being hashed. So if say for example I'm going to use the one that I've presented in the call where I spoke a proverb which is the compression of my public note in a discussion. Um if I you and I conferred on a discussion post and we both came up with our own proverbs which are the compressions of our understanding of that and then we split them in half and then 50% is mine and 50% is yours. That's us and then we hash it and then put it into the proof. That's us like taking 5050 responsibility of the content that went in to that or the compression that went into that like as a presentation of us being like, "Yeah, yeah, we were the ones who commented on that and we were the ones who read that comment and and did that thing." >> Um, >> okay. If that's what you mean by it, I understand that's okay. Got it. So it's like interle um the data or the metadata that goes into um what it is that is being proved um where then if we have separate systems for holding that information then we have to confer in order to um like we would need a VRC in order to um present that together. Um if if there came a time where we wanted to reveal the information behind that proof. Okay. So for GDC for GDC. Yes. Uh, >> I'm planning on being there Monday to attend the the um >> the the meetup. >> Good. Good. Scott, what's your schedule? >> What's the meetup? I am not aware of that. >> Oh, you're Are you not on the first person mailing list? >> Uh, I am. I I saw your email yesterday. I I haven't Is there a meetup on that? Scott, you're supposed to read those emails immediately. >> I [laughter] know. I need an agent to do it. My bad. >> Yes. Um there is going to be a meet up on Monday afternoon. Um but you know, it's it's for anyone who's available to uh come hang on. >> I arrive I arrive at 11:00 a.m. Monday and have >> Sorry, I don't know where that came from. Um yeah, from um uh 1 to 1 to 5 we're going to the first half we're going to be actually um um testing sorry um the um uh the uh open VTC um code that um um Glenn and and Jeff are preparing for the uh plumber summit uh which is only at that point another month away. Um and then and then wow was I don't know why this something's making me sneeze. Um this in the second half we will uh uh be you know just having an an update about uh progress across various aspects of the uh firstp person project talking about a go to market strategy that we've been uh developing that's just want to get people's feedback about um so anyway uh everyone is welcome and uh I I reserved a space last week and they said they'd be right back to me with the confirmation and it's been like it's getting to be like five days now. I just checked it again and they still so I don't know how to get in touch with these folks and say okay can we nail this down. Um yeah anyway um we will uh I guess the the direct question here is in the sessions that our workshop organized that um is there is there going to be a place where we need that ZKP interjection um should we be prepared with something for that uh or should it more be like audience input of someone who is part of closer to the coldface. What What do you think? >> Uh are you talking about in the in the uh you're talking about the workshop? Excuse me. >> There's you you're talking about in that first half of the meetup where we're >> uh No, I'm talking about in the GDC >> the GDC presentation. Yeah. in in the meetup, I assume it's it's fine to just like hang out and interject, but in the presentation, does there need to be something >> in the presentation? That's up to uh Scott, you were the one who submitted the the presentation proposal, right? >> Uh for ZKPs, I I didn't do one about ZKP. >> What What did you do one on? >> Holder binding. >> Ah, that's right. That's right. Holder binding. Okay. Well, I sort of assumed that you were gonna maybe mention TKPS and that, but anyway, that's um um so so that's uh I mean, I'm going to touch on it. So, I have one on the decentralized trust graph where um but it's a highle presentation, you know, to introduce the DTG to that audience. So there will be a mention of ZKP in there because it is a key piece of our story, but it's very, you know, it's just like a mention on one slide. We won't be getting into any any [laughter] details about it, although there might be people in the audience that are interested um and want to follow up. Um and then uh so Scott on holder binding, are you going to be uh what are you going to planning to cover there? Uh well it's a panel discussion uh moderated by David Keltz and including Paul Grassy, Brent Zundle, someone else to be determined myself and kind of covering all the lanes of what are the challenges with older binding and uh I will be providing the privacy preserving biometrics lens of it and David will be will just kind of be pushing for what are the challenges and opportunities there through the different lenses. is that those speakers represent uh and I can definitely represent ZKPS but it was not explicitly like this is a ZKP session that I'm getting. >> Yeah. >> Yeah. I just wanted clarity on the what the field is. I think I have two workshops that I'll be hosting. Um one is on PQC crypto agility. So, it's more on the quantum side. And if that is interesting to you, I would recommend that one because we have one of the guys from the Google quantum um >> Oh, >> yeah. He's going to present. It's low key, but he um one of the guys, one of the authors of that paper that published the circuit as a ZKP that over the last two months we've been able to recreate in the auto research community. Um, Adam so that one I wanted to give this group heads up on because it's quite it's yeah it's relevant in two years but we should learn about it now. Um and then the other one is more about wallet um security target profile things. Uh so a little bit less but obviously it's all kind of folded and related. Um and then there is a I think that's the two >> what uh what where do they um so I could mark them on the on the calendar. >> Yeah. One of them one of them has an overlap. Um, so if you go to the GDC site, it's on one is on September 2 and I think if you Yeah. So, >> and what's the name of that one? Give me a search term. >> Uh, BG. >> Yeah. And on day two, there's the postquantum cryptography migration on DTS deployment, agility, and governance. That's the one. >> That's the one that I mentioned that we'll have the um Google guy. And then on day three, um there's vulnerability handling in the agentic AI era. Uh that's the other one. That's the one I've done most of the work on. Um so this is kind of getting into what my agent harness does. Um in terms of being able to like disclose vulnerabilities inside security without telling the frontier model or whatever cloud it is that's hosting your agent um that this is happening. And then the final one that is um these two are hosted by Julian Bringer but I'm a part of the beginning group that he is and is wallet security assurance STP >> um which is yeah more more about like um standards and frameworks for what's a good wallet. >> Got it. Got it. Yeah, I see these these two um on uh on Tuesday on on excuse me on Wednesday the 2nd of September. Unfortunately, I the fiduciary session is the same time as >> the quantum >> that one. Yeah. So, well, you know, there's >> well I'll have to report and we can we can chat about it. >> Nothing we can you know it's just that's the way these things are. What can we do? Yeah. >> All right. Well, good. I'm glad I know about them. Anyway, >> awesome. And Drummond, you're doing you're in four sessions, I think I heard. What are the other ones? >> Uh, well, I'm actually in six sessions. >> Um, >> underelling you. My bad. [laughter] >> Well, I'm only I only proposed two, but excuse me. Um, let's see. Uh you can if you uh uh type Drummond into um the uh the session review here. I'll put it in the in chat. Yeah, there you go. Yeah, you're looking at it right there. Um >> uh you're if you're on uh Yeah, it's not day one. You have to go to uh day two. >> Number two, you're on >> Yeah, you got to scroll up a little bit more. I don't know why they put that up so high. Yeah. >> So, yeah. >> There you go. >> Choices. Wow. Five on day two. >> Yeah, they just they're just a bunch of different um luckily they they're we've managed to eliminate any conflicts. Um between those there was one. Yeah. But the the last one um uh is the one primarily you know presenting the work of this you know working group and then uh the second one I think the fiduciary I one is the one I'm doing with uh um Richard Wit. He'll be the primary presenter in that one. I'm just gonna Daniel and I are gonna sort of set him up and uh and then let him talk about the whole thing of fiduciary um net fiduciaries and and his book uh reweaving the web which I do recommend to anyone that's interested in those uh the legal aspect of that topic. Yeah, >> I fortunately met him last year and he gave me a copy of the book. >> Oh, cool. >> That was my Yeah. in part inspire inspiration to some of the work. >> Cool. >> Web of trust stuff. >> Yeah. A macro for that, but I can't do it right now because what is going on here? Um, >> yeah, I don't have any other items, so that's cool. >> Cool. Well, I think the items we're looking at next, Glenn sharing, and he already shared it right up first requirements. It's not the right way to say it, but uh, Drummond setting up repo for us. Mitchell sharing What would I call what you just said? Um the >> I'll share the like the yeah the HTML of the ledger um of proverbs that I committed alongside the discussions comments. Um yeah, I'll just put that in these notes and think I'll just edit into this document after this. >> Cool. Um when do you Well, It's not that far for you, Mitchell. So, you're probably not going that far in advance, but Drummond, when do you leave for Europe? >> Uh, Friday night. Uh, so I get there Saturday night and uh I'll be time adjusting on Sunday and then uh you know, attending to the uh first person meetup on Monday afternoon and then then we're deep into it. So, um when do you arrive? Uh Scott, >> uh Monday at 11 a.m. >> Oh, okay. >> Good. Coming going to London first to I don't get to see my team often, so I'll see some of them on Sunday. >> Oh, cool. You're going to see them on Sunday. Okay. >> Well, definitely. We're going to do the u [clears throat] uh first person meet up from uh 1 to 5 on uh on Monday afternoon. So, uh and it's going to be down it pretty down near the train station. um pretty much right in the center of town. So, uh it's should be pretty convenient to get to. Um >> that's again, if I can confirm this place, there's no number for me to call otherwise I'm going to get on the phone here and these guys. Um anyway, [clears throat] that's that's the plan. So, yeah, I look forward to see >> How long did it take you to adjust your clock from your body clock from New Zealand? Um yeah, it did take uh it was like five or six days um before I was getting a full night's sleep. Um that's you know that I I just that's I actually find it it's the same thing as going from here to Europe, but for some reason I adjust faster when I go to Europe and then I don't know there's just something about going east that um it's harder. Um, I had no problem in that direction. Uh, >> yeah. Going back. >> Yeah, exactly. It was just coming back. >> Yep. >> Yeah, that's uh coming from Australia, it's the same thing. It's always like the week after you come back to to London, it's the >> Yeah. >> takes a little bit longer. >> Yeah. Exactly. Anyway, it's it's all good. And in this particular case, it's like exactly one week over there and then back because my sister and brother-in-law are coming for Labor Day weekend. So, I want to be back here for that. Um, Yep. >> Amazing. >> Yeah. >> Well, see you all there. Looking forward to it. >> Geneva. [laughter] >> Yeah. Yeah. So, yeah. So, that means I don't think we'll have a call next week. Uh, Dennis. >> Correct. >> Yeah. Um, in fact, unless I was I'm checking with each task force. Unless you're planning on having one, we're going to just wipe the calls for next week and tell um LF to take them off the calendar. >> Excellent. >> So, Dennis, get a week off. >> Yeah, just what he's looking for. [clears throat] >> All right. Very good, Jen. Some take a break before the human trust experience gets together. our three-hour task force blocks and on Tuesday mornings. Good stuff. >> All right. >> All right. Thanks, guys. Take care. Bye.