Patrick Atwater, innovation program manager at the Metropolitan Water District of Southern California, argues that protocols exist simultaneously as documented formalisms and undocumented implicit norms, using case studies from a neighborhood park, bridge/infrastructure permitting, and track-2 diplomacy to illustrate this gap. He proposes practical strategies for navigating illegible terrain—black-boxing unknowns, cultivating domain-expert practitioners as guides, and lightweight fieldwork logging (e.g., a 'protocol watching' Notion database)—and shares a real example of a water utility valve project where jurisdictional coordination revealed pseudo-official knowledge and unwritten norms.
Shared by the speakers for the symposium. Email addresses have been removed.
Transcript
YouTube’s automatic captions, unedited, so names and technical terms are often misheard. Each timestamp opens the video at that moment.
0:00 Um, all right. I think it’s a time I’ll answer the rest of the questions in chat. So, back to you, Timber.
0:08 Thank you, Benat and Humble. I’m excited for the the roster of AI researchers to um expand over the course of um the rest of the year or two or maybe next year. Uh yeah, next up we’ve got Patrick Atwater.
0:27 Alrighty. Hello everyone. Um, my name is Patrick. Let me make sure kind of a meta protocol for transitions. Does the slides work? Is can you hear me? All right. Can you um everything coming piping through correctly?
0:53 All right. Uh, first a bit of uh Quibono, Latin for consider the source. um wanted to share a little bit about kind of where I’m coming at from protocol studies. Um currently manage the innovation program for the Metropolitan Water District of Southern California. Uh beyond that, I well I have two small kids so I do a little bit less of that but I enjoy uh skiing and kind of decision-m and avalanche train is one piece where there’s a lot of protocols and a lot of things to ponder. And the other one is I am a a backyard gondner. Um so my my little prompt uh apppropo of nothing just a kind of you know kind of attention focal one is looking at this garden here uh what do you think which plant do you think is uh grown the most since you know I took this right at the start of spring so over the summer what do you think has grown the most so we’ll come we’ll circle back back on that at the end um but uh this is up in the this skiing photo with my dog Scout uh is up in the mountains uh above where I live in SoCal. And so there’s a lot of things around, you know, kind of like slipfall risk if you’re going to go do uh alpine terrain up there. Um but the water there is ultimately uh theoretically at least going into the ground um and could be pumped up and used um you know in the garden over at the right. Um and what I’d like to talk to you about today, I wanted to kind of start with some physical examples was around um the idea of illegibility around the things that around that aren’t specified in a protocol. Um and my bias is more in the kind of uh you know if you do like natural and moral philosophy be more on the latter on like touchyfey human stuff. Um but I wanted to give a couple of you know physical examples though that you know even there you kind of see that. So, you know, my toy kind of reductto had a certain argument is, hey, what if we were to try to see how water flowed from the, you know, skiing situation all the way down like theoretically, you could do some kind of, you know, some tracing. Um, you could use some remote sensing, you could do some incentu stuff. You could try to be really clever beyond cost. Um, I think the point is that the, you know, there’s going to be a little bit, uh, that is kind of always missed. Um so here today what I wanted to do is talk about a um kind of similar a couple of case studies around how do you how can you navigate that beyond you know in kind of the garden and the snow examples there’s a lot of kind of rules of thumbs there’s things that you know you don’t necessarily need to know precisely what the soil moisture content you don’t necessarily need to track the snow pack precisely you know gardening I use uh just kind of a running series of like Gemini sessions to be able to kind ask hey like is this plant getting enough water things like that there’s also sort of um a stigma sort of element as we have different folks in the household you know watering the plants like is the soil wet is it done has it been done recently um but there are parts that are kind of indelibly illeg ill uh [clears throat] uh not legible illigible as I can’t say that word um that are are difficult to to kind of measure um and and track. So, a couple examples. Um so, I wanted to talk through um one uh another place I’ve kind of spent a lot of time just kind of loosely pondering protocols, which is, you know, the the park where I take my my daughters a lot. And then two, um kind of adjacent to a lot of the work that I do around like cities and infrastructure, um looking at uh bridge construction over the time. And three was you know also relevant to water where there’s a lot of you know difficult negotiations is kind of around um folks talk about diplomacy and diplomatic protocols uh looking at uh uh strategies for for doing that um for like track 2 diplomacy and so the lens I wanted to bring to ill uh things that are difficult to make legible is like okay like well what can you do with that what can you do that if it’s like unknown if it’s difficult because it costs to to kind measure everything or if it’s difficult for any particular reason, you know, what can you how can you still navigate that situation because you still want to hypothetically, you know, grow some plants or uh ski down a slope or whatever the goal is.
5:28 Uh so this is the park. Uh it’s pretty and so there’s a variety of different spots. Um, one of the things that’s kind of fascinating is, you know, just kind of it’s like a, you know, Jane Jacobs style, uh, approach of, you know, people watching. There’s no shortage of protocols, right? So, you have both like a variety of formal and informal sports. You have folks coming together for like a picnic. You have degrees of formalization there. You have folks coming in for uh things like a birthday park, uh, like a birthday party. So like you actually can like reserve through the official like government apparatus like a um you can uh reserve a picnic table. Um there’s things like pickup sports. There’s things like formal sports for like kids where they have leagues. Um they have things where it’s like kind of implicit there. Um there’s a lot. Right. So here’s a a little effort to kind of list some of these out. Um I pulled these from I have a little like running list in notion to do a protocol watching. Um, some of these are more or less legible.
6:32 Um, and I think one of the things I want to talk about here is that a lot is actually quite legible like you know there’s things that kind of you take for granted like in a park there’s that is uh you know like the county like the folks just take care of uh making sure like things are watered, the trees are fine, that the you know you have certain expectations around um you can go to the play equipment, it’s not going to fall over. There’s the permit reservation system. Um the various signage like little things around like um you know like they’ll put up a sign like oh don’t go here because the bathroom’s closed or because they’re doing maintenance or stuff like that. But then there’s a whole lot of stuff. So like like in pickup sports um if anyone’s ever tried that like how you know there’s always kind of a little bit of tribal knowledge there. It’s like what qualifies as next in a game. Um there are things around like uh you know having dogs like letting dogs over like folks will let them over in the the baseball diamond area. Um you know there’s a lot of implicit protocols there um around how you negotiate that how you do that there. Um there’s things around like you know who and how you can let up signage.
7:43 So um you know is that allowed or not. Um, and the piece that I really want to talk about, and we’re going to revisit this in kind of each of these case studies, and hopefully find some generalizable tools, or at least tools that I find helpful, so maybe others will find it helpful as well, is a couple of different things. Um, one is, you know, I was once upon a time a math major. So, it’s like kind of the idea of, okay, how can you deal with this as a kind of a black box? So like even if something like uh you know like pickup sports you don’t like know the people. You don’t know like their specific like rules or like what’s the culture it’s like you could kind of see it’s like okay like who who is the one that’s kind of like the center of gravity socially like uh and you could just ask them you know be like and then ask them it’s like okay like I want to play or like when do you play? Um there’s kind of different questions there. Um that kind of relates to the second one of like cultivating practitioners as guides. Um, so hey, you know, who’s someone who could, you know, if there’s someone, if they have a friend, um, that maybe plays a pickup sport that like plays basketball or soccer or whatever, um, how can they go there? Or, you know, maybe, you know, there’s kind of different communities that will go there. So, there’s kind of like different etiquette around um, you know, like kids birthday parties and things like that. It’s like, oh, okay, if you want to try to, you know, bridge and, you know, communicate to like another group like, hey, who would be someone who could be uh uh kind of a bridge building kind of diplomat? Uh, actually, I think kids and dogs are actually great for that cuz they’re an easy way to break up the conversation and like, you know, meet kind of meet new people type of stuff. Um and then the third one is I have a little bit of a stack around using um just like a cloud skill to do like the protocol by order and then log that within notion which was some of the results um from the one before.
9:40 Um, so you know, kind of beyond just like pickup sports and some of the examples I was talking about before, um, you know, like how dogs interface. There’s a lot of different, um, ways that you can there’s a lot of different etiquette around [laughter] that. And there it’s a whole culture around dog people particularly and maybe this is just like a a US thing where people we’re a little crazy about our dogs but uh you know there’s a whole etiquette around like people around like whether they like leash greens like don’t like leash screens. There’s like different things that you can kind of learn in uh dog training classes about like you know reading whether their fur is up or things like that. Also reading the body language of the person you know kind of asking around how to do it. uh there is a you know so a variety there.
10:32 Um second example I wanted to talk about was say you know you’re not going up to your neighborhood park but say you know you live in a town or a city or or something like that and some folks are you know looking to to build a bridge. Um so this is something I was kind of pondering around you know how infrastructure gets constructed. Um, I live, you know, work in a public utility where things are very formalistic. You know, it’s like takes formalized to, you know, there’s formalizations on top of the formalizations on top of the formalizations. Um, and how do you how do you navigate that? I mean, a lot of my, you know, it’s kind of kind of interesting spending time around civil engineers and trying to understand both the kind of culture and also the uh approaches there. So things around uh like take storm water recharge or capture. So like how you can let water cool at a place like a you know a park like that above or you know there’s a a case study that uh I wrote for protocolized which actually is just like a block north uh or up the mountain uh where we dug a bioil. Um but how those are often done is even today there’s a lot of formalizations in terms of how you know design bid and build like the procurement specification for something around the public works project. Um but [clears throat] then there is a um a lot of that leans on a lot of the kind of math um leans on things like there’s the uh rational method like Hamming equation things like that that are um kind of you know late 19th early 20th century um you and one of the piece that I was thinking is like there if you go back way back into antiquity you know how things So, you know, like how the Romans would build something like an aqueduct or a bridge, there’s a lot of more of uh there’s certain formalisms, but it’s a kind of a different category and approach. Um [clears throat] uh you know you have a different relationship to uh how you kind of use like the the kind of guild knowledge how you can um formalize things in terms of methods you know so how you can uh look at different recharge rates things like that um how things and one of the pieces that you know I’ve been kind of pondering around is uh how that gets impersonalized by different specifications ations and um uh today you know part of the piece that I’m really interested in in protocol studies particularly for things adjacent to you know infrastructure construction and cities is that there’s a you know there’s so many formalisms that it’s it kind of becomes difficult to kind of you know have synthesis and judgment. though, you know, how um being able to use tools like, you know, an AI or codification or, you know, integration of protocols to be able to kind of navigate and um uh understand and be able to operate in that more complex thing. Um, third case study that I wanted to kind of flag and plant a seed for, um, was and actually I just recently had a kind of fun little dialogue with, uh, chorefan who I know many folks know from, um, uh, was around track 2 diplomacy. So we kind of off diplomacy protocols things like Kyoto and or you know whatever uh whether it’s some of like the nuclear uh launch codes or nuclear proliferation and diplomacy that kind of gets put up as you know frequent example here. Um, track two is an opportunity [gasps] where you kind of it’s like messy and ill in not legible, illegible uh by design. Like a lot of the value is that it’s folks that have kind of like plausible deniability. Um, so it’s these are folks that, you know, maybe they’re a retired official. Um, they’re an academic, a clergy, uh, scientist. Um, and uh, [clears throat] the piece I was talking with Chore about was the the uh, the opportunity. I don’t know the AI safety discourse is kind of silly. I’m not super smart on that stuff, but um you know, one thing that strikes me is that kind of seems lowhanging fruit like the opportunity to have, you know, create space for where you can have folks who have technical knowledge to actually like hash things out and like figure out like um creating neutral grounds particularly, you know, creating space across like the China America divide because the geopolitics is one big elephant in the room there. Um also been kind of cogitating on you know uh water politics are very you know one adage like uh is the you know if you go back to like at the state or federal level in the US like western water is some of the the trippy trickier diplomatic angles but you know and there’s a variety of things but they’re not you know it’s like how to kind of nurture those with like the potentially with things like the Colorado River or uh local water like storm water issues um things where there’s uh uh very tricky political economy um and that’s something I kind of think a lot about with my uh with my day job um um and I think a part of a lot of that is you know you find folks a lot of the value there is you have folks who have enough kind of like trust and context and understanding with each within each of the worlds that you can start to build a bridge And then you could start to figure out like answer questions like well what do they really mean? Like why are they really have this posture?
16:39 What’s really gone? Is there like is there a fear? Is there uh you know is there other kind of like internal politics? You know what are the things within the coalition? because you realize it’s folks like uh like to use a water example, it’s like you know the upper and lower basin on the Colorado River like they’re not monoliths or you know like the US is not a monolith obviously and like um and for similar things for you know other things like with China and like the AI geopolitics um so that’s kind of a quick kind of speedrun of a couple of different case studies um and a couple of strategies that I found useful um what I like to do there is just kind of plant a seed um particularly to invite others. You know, I think if you take away anything, this is kind of just me kind of real wild reing and you know, you know, having some messy ideas around like, you know, potential areas for future research. Um I’ll share a link um in the chat and kind of you know, really would value kind of, you know, folks to bounce ideas off of SP there. um once, you know, this is a quote from a professor of mine from undergrad who was in um pp philosophy politics.com. So he did a lot of cool stuff around um if you ever driven in uh LA and Southern California some of the diamond lanes you kind of helped get the uh um you know congestion pricing there which is not like the world’s best example globally but it is you know LA SoCal we love our cars. Uh but he has this quote better crimson examples and gray words than crim words and gray examples. So that would be my like little branch that I would like to offer for uh the protocol community. Uh [clears throat] it’s just you know folks that are really interested in kind of crimson applied examples. You know I do think there’s some focus on things in kind of like the engineering world or like the techie world. Um but though particularly there’s I think there’s a posity of good crimson examples and kind of things that are kind of social science and humanities. I think well opportunities for improvement that’s a let’s go a glass half full. Um so towards that this is one and I’m going be hosting a workshop uh the metropolitan water district we’re going to work around it’s um it’s pitchfest and this will kind of an idea around this will be there um the research need here was around you know kind of similar to the infrastructure construction the bridge building but you know we talk about formalization this was a valve which by the way this was like an eyeopening experience this valve is like you know it’s bigger one of the world’s largest you know water valves which I don’t know for me it’s like I spend time at like a tiny we’re talking about like a you know 12 in 18inch valve but um the per uh permitting you know folks talk about permitting it’s in the zeitgeist with like abundance I think it’s really there’s a lot of folks who don’t know what they’re talking about with peritting in terms of like how you actually build stuff um and the thing that’s interesting here is it’s not one of like the politically charismatic things like SQA or NEPA the folks these are US examples that are like environmental peritting like get a lot of oxygen in the reform discourse. But the thing that’s really interesting about this one is that you know it’s just like just kind of a plain vanilla like pretty standard like you know you have to like dig you have to dig in a street but the coordination became super complex because you have it was adjacent between two different cities and then the street was one of these streets that’s big enough that it was managed by CALR. So you have this like kind of little game of telephone. So the research I was thinking I want to do is to uh to use kind of a digital metaphor.
20:08 It’s like kind of like trace the logs. Like if you actually like were to trace the log in terms of all the different communication points um is uh you know like what would you what would you find like what would what were the actual like hard and fast things that rules that were required and what were not. Um, and what you cuz one of the things that you see if you spend time in these, you know, hyper formalized domains like infrastructure construction is there’s a lot of like pseudo official knowledge where it’s kind of like, you know, someone who’s kind of in official, but it was like they were half remember something because there’s just so much to remember. It’s very complex. Um so it’s not quite clear you know what are like the hard rules and what are the kind of implicit norms that kind of just you know pass along uh and what’s kind of just like institutional like the obverse of institutional wisdom. Um, so I’ll show the link when I stop presenting, go to Q&A, but there’s a a workshop and I think I put in the Discord. Um, if anyone is like the academic folks, there’ll be kind of remote options to listen to that, but that’s a research one and we have a little pot of funding for that there. Um, flash this QR code.
21:19 Um, and this is a kind of a halfbaked blog post looking particularly on like uh research into protocols. This is kind of looking at government operations though I’d be interested and very eager to kind of collaborate with other folks that are looking at uh you know Tim mentioned in the kickoff around like kind of like the implementation of like the protocol studies like what are the things around you know like government is one angle to look at kind of like cities and civilization there’s a tangle of a lot of other things like kind of like infrastructure traffic like all sorts of things cities are really interesting there’s a lot of other areas we could do applied pro protocol research and kind of human problems um very interested in a lot of those and kind of curious to bounce ideas off of other folks there. Um and yeah, uh this and this is the answer to our, you know, come full circle with the whirlwind as we look at kind of illegible uh protocols. Uh well, the kale did pretty well, but and it goes further over to the left. Uh, but the sweet potato, no one really expects the sweet potato to kind of be the the eight potato. Well, at least I didn’t. I’ll have to like look in the chat and see what folks have to say. Um, but that was the the apex predator um with with my little the little garden. It kind of went gang busters. It was almost dead prior to the photo there, but now it’s going going blooming quite rapidly.
22:48 Um, so with that, if there are any kind of like burning questions happen to do dialogue there or uh I haven’t been following the chat, but if Timber if there’s anything hot off topic. Um,
23:00 I didn’t see anything come up um explicitly couple comments, but um yeah, I was definitely going to guess um kale or raspberries myself for the garden. I was not expecting sweet potatoes.
23:22 I have a a bit of a question about um yeah, how you’ve been using this um kind of like language for a while now in government and there are probably a few people here who are like either in public service or like adjacent to it.
23:45 Yeah, so I think it’s given me a set of tooling for like how to navigate it. you know, we’ve taken very ba baby steps in terms of unlining the some of the, you know, tangle and procedural pretzels that we twisted and I know we collaborated on we did this internal our CEO tasked myself and our innovation program with streamline like kind of it was a little of organizational therapy among friends where we like listen to everyone’s gripes about internal processes and procedures and Tim uh [clears throat] Timber did some amazing little workshops um with folks and that you know we took some baby steps out of that there. But, you know, we’ve kind of been able to kind of grow that. We have a new co, we had a transitional figure. There was a lot of uh we had a little bit of a civil war internally, SoCal Waters, which my institution was sort of the front and center around. Um, so this was kind of the organizational therapy and listening to folks gripe around procedures and red tape was a little bit of the healing out of that. Uh but there’s going to be like we have some new there’s been some staffing changes and some more resources aligned behind that. I mean we’ve definitely looked around like our internal payments. Um the permitting one was something that was you know a different one executives that came out there. So I’m very curious to like the research there to how we can untangle that. Um and I think you know to circle back on your question around how it’s been useful. So like that valve that I showed that was a giant valve.
25:15 There’s also another valve that the um uh like an 18inch valve that was like really really expensive. It was like around over a million bucks for that a single valve which is like wild and it had to do with kind of like when you make things so formal and clunky it’s hard to get enough bids and stuff like that. Um but uh things that with uh like the efficiency thoroughess trade-off have been helpful in terms of communicating internally like there’s often a lot of defensiveness when you say like oh that costs like way too much money right um but it’s like hey cuz the the engineers will say it’s like well we had to be there’s a thoroughess there um in terms of wanting [snorts] to have something extremely well engineered um because a lot of where I’m at it’s like you know we’re trying to be 24/7 365 kind of be a guaranteer of safe and clean drinking water for 20 million people which is like you know it’s a very high bar but uh you know reframing it as not just a you know that we’re further away from the frontier of the efficiency thorough triumph and like there could be different ways to get there um has been productive um you know still TBD it’ll probably take it’ll take a minute to kind of translate that into kind of shifts Yeah, speaking of um EdTeto, I like informally designated September 26th as International Edeto Day. So that’s coming up on on Saturday. Um fun fact
26:57 I see Andre has a chat as well and then Ellen as well. uh Andre’s how to grow uh into heavy environments. Um my you know the best advice I ever got was if you want to do something radical particularly in government is do a pilot. So like and then you kind of like hit people with uh so that’s one part. So framing as a pilot and then the other one and a piece of advice I got from the um it was Bloomberg’s data guy uh he’s chief data analytics person was around like sometimes you got to hit people with the imagination and say like well what if like what if and why not like you know the kind of things like how things might be differently I think right now there’s people people like know right like it’s in like it’s in the air like that thing the times are changing like you know the AI like people realize that like I have board members who like you know that are you know like in the industry like there’s folks that are like from very different like people know that things are changing so it’s like there’s opportunities to do things um do the thing there yeah and Ellen yes feasibility studies could be a place where things die um and then Ellen I don’t want to forget your question but what h what works best when dealing with things that happen at boundaries of jurisdictions and competes and any insights on how to make that go more smoothly Uh the thing that’s difficult is that is kind of I would say like you know the practical answer is is that is kind of where you know if there’s like the metnotish of metatis you know like adysius the majes in the news it’s like where he’s like I think in the the Greek it was like like polymetus was what the folks talked about him but like a lot of that is like these are necessarily things that are difficult to kind of document and codify. Um, so one of the things I’d say is that I have found helpful is just like paying attention to little things that are quirky. So I think getting involved in the protocol watching habit is very valuable as a personal practice that I have found um because [clears throat] then you can kind of notice things that are odd. Um, but I’ll share and what I’ll share this final thing I’ll do is tying it to the workshop and the nice thing that Rafa did yesterday that I found very helpful on the um with Sachin. I used the some sashin there’s a little bit of a rag factory that we were collaborating on and I used that rag factory just kind of prepping for this talk for um on my protocol watching DB. So I have a little [clears throat] like micro site and I’ll show that in the chat and now we’re at 12 um or 12 Pacific and so I will um say goodbye because uh so we can stay on protocol.
29:38 Thank you Patrick. Um yeah, we’re doing a great job um staying on time. So, thank you everybody again for doing that. Next up, we’ve got Tim. And this is our last session before the Psycho History SIG group has a two-hour block for their special session. So, all yours, Tim.
Simone Cicero of Boundaryless presents O2A (Open Organisation Alliance), a proposed open standard for making organizations 'composable' by replacing implicit coordination mechanisms like hierarchy and meetings with an explicit, machine-readable contract graph of units, offerings, and relationships. He argues the protocol's compact grammar—three economic primitives (purchase, revenue split, investment) plus composition mechanisms—can represent bespoke institutional arrangements like Haier's Rendanheyi model or joint ventures as 'sentences' in a common organizational language, becoming essential as AI agents enter firms and transaction costs collapse.
Tim Beiko, core coordinator of Ethereum protocol development, traces how his own need for frameworks to steward Ethereum led to the founding of Summer of Protocols and now the Protocol Institute. He argues Ethereum functions as a 'model organism' for protocol studies—citing the 2017 Parity multisig freeze, Uniswap's constant-product market design and fee-governance debates, prediction markets' oracle problems, and atomic flash loans—to illustrate his concept of protocols as 'engineered arguments' that draw a boundary between formal rules and human judgment.
Eric Alston presents a framework of five recurring 'governance axioms' for protocol design—incomplete rules, governance-as-conflict, mechanical concentration of authority, asymmetric incentives from shared resources, and the transparency-privacy tradeoff—arguing these tensions can never be designed away, only managed through mechanisms like killswitches, veto-gates, and tiered amendment rules. Drawing on his experience advising Cardano's constitution, DAOs, and peace processes, Alston contends that 'incomplete rules' mean norms and discretion inevitably fill gaps left by any formally encoded protocol, and that durable systems require 'dynamic fidelity'—credible commitments combined with constrained capacity for adaptation rather than either permanent rigidity or unchecked flux.