[{"content":"🤝 If you would like to add or comment, I invite you to do so. It’d be great to have more voices.\n⚠️ *Disclaimers:\nI have not been through Antler but it’s similar enough for info to be relevant. If you went through Antler and want to contribute, let me know! I did not team up with anyone through EF or ever pitched to them. I found my cofounder through YC co-founder matching and we’re building Peanut Protocol (hmu if you’re curious!). I still consider the programme to be extremely valuable and it definitely changed my life in many positive ways. I am often wrong and these are just my opinions.* Why write this? I get asked the same questions about EF, so it’s easier to answer them once, rather than repeat the same conversations. I want to help. Maybe if help someone with this, that someone will help me later on? Startup world as a whole is messy and opaque. The barriers to entry are much lower than people think, but ecosystem and culture knowledge is paramount. For instance, both EF and YC use obscure terminology for well-known concepts. YC asks “what do you know what others don’t” instead of asking for “competitive advantage”. They probably think that’s avoiding jargon, but without clarifying the question it could refer to anything. This actually increases the barrier, rather than lowering it, because you can’t google what they mean by that. EF uses “followership” instead of “leadership” 🤷. And I’m sure some midwit will be able to explain all the subtle differences, but idc. It’s a poor attempt at differentiation at best, and subcultural barriers to create in and outgroups at worst. I like to talk to founders. So hit me up. What is EF \u0026amp; Antler They call it “talent investment” but that’s just a term to make us feel better about ourselves. It should be called something like pre-team \u0026amp; pre-idea accelerator, or founder matching accelerator. Confused? Me, too. The above graph illustrates it. In case you’re not familiar with venture capital, read my noob intro below.\nThe key hypothesis of EF \u0026amp; Antler is that they can get in super early. They believe that:\nsuccessful founders share many similarities (you can tell in advance) many people do not found companies, because they can’t find a co-founder (co-founder search is really really hard and solo founders usually fail) many great founders need an extra cushion or an extra push, this means: personal runway (how many months you can live without a salary) ecosystem: knowing other founders really helps knowing investors helps having access to frameworks and learning materials helps So their idea is to take would-be founders and give them this stuff, in exchange for a % of their future companies.\nNoob Intro to VC In case you know how the basics of VC and equity work, you can skip this.\nCompanies have equity (also known as stock or shares). This was invented, so that people could buy a part of a company. The more valuable a company is, the more expensive 1% of it is. Companies that solve a lot of problems that people are willing to pay for, e.g. Airbnb, are worth a lot ($50B or so). Now, if you had just 1% of Airbnb, you would have equity worth $500M, and you would be very rich.\nBut Airbnb was not always worth $5B. It started off as a small company, worth a few dozen thousands, perhaps. Whoever invested early, is very well off. Early investors (like Sequoia or Y Combinator) invested a few million for 10-20% of Airbnb equity, leading to a 1000X return on that investment.\nHowever, most companies are not successful as Airbnb. 90-99% of startups fail, depending on how you define failure. Early stage venture capital investment deal with this by investing into 100 companies, knowing that 90 will fail, 5 will be barely OK, but 1-5 will make back their money more than X100.\nEarly companies are worth less, because it’s not yet clear whether and how profitable they will be. This makes them more risky investments, but with much higher potential returns. Risk and high returns go hand in hand. After all, if there were risk-free high returns, then everyone would bid for those, thus increasing the price and lowering the return.\nAnyway, the earlier, the riskier but the more % you get for your $. That’s how finance works. And VCs go for the high-risk, high-reward end of this.\nThe Deal EF’s deal is 10% equity for roughly $100k + the programme (2022 info).\nTo give extra incentive to join, they also pay you €6k for the 4 months of the programme. Yes, it’s actually 4 (not 3, like they say), because the programme starts when they release cohort names and people start networking. That helps with your personal runway (how long you can build a business without a salary), and people don’t feel that they’ve given up that much other than time, in case EF doesn’t work out for them. I think this is a gamechanger for many people.\nEF do not invest in every company that’s at EF. They only invest in 30-40% or of the companies who have teamed up. In my cohort, there were 30 people who teamed up, and 17 who did not. That’s 15 teams. This means that EF will invest in only 6-8 teams, i.e. 12-16 people, which is only 30% of the total cohort. Don’t think that because you are in the programme, you will find a co-founder and get the pre-seed investment.\nIs it a bad deal? People tend to complain about this deal. I disagree. At EF you start with nothing. Even when the investment happens, the most you would have done, is probably a waitlist or maybe a few FFF (friends, family and fools) customers. The 100k let’s you build things for 9-12 and pay yourselves a $2.5k-3k monthly salary. Generally, you should be able to raise more capital a few months in.\nOf course, when you look at the company’s pitches, it sounds like insane traction for a few months. But it’s not. For instance, one of EF’s flagship stories musiioo, claimed that they have a database with millions of songs already. Sounds impressive, right? Well, it’s not if you think more than 5 seconds about it. It’s a vanity metric. All the songs came from an open access database. Anyone tech savvy could download them and run some data analysis on them. It just sounds impressive, because it’s a large number. Anyway, just wanted to emphasise that it’s reaaaallly early for most of these companies. It’s not common to build something big in a few weeks.\nWhen you pitch, you have “to be onto something”, not have a seed-ready company with a working business model. A few letters of intent, some decent weekly progress, etc. are good enough.\nOn top of this, you have to think where you are in the stack to see whether it’s a good or bad deal. If you’ve founded a successful startup before, you know investors already and they trust your business much more. That’s a bad deal then, you don’t need the EF money or the many talks and lectures. The only thing EF will be good for is your co-founder search, and that alone might be worth 10%. And it will be great for that, because you’re likely to already know what you want in a co-founder.\nBut if you’ve never raised much money, ever founded or worked as an early employee at a startup, then EF could be your gateway to the startup world. I think that’s the most attractive thing about EF. It’s a very easy, low-risk entry to give it a try in that world.\nIs EF a good investor? Short answer - they’re probably 2nd or 3rd tier. I’ve heard that they expect frequent updates but give little in return. However, that’s a good worry to have.\nIs the alumni network worth it? Short answer - no. The alumni community is relatively dead and there almost no events. The cohorts are very active, but once you ‘graduate’, there is nothing that sticks with you as an alumnus. Sifting through current cohort members might be useful for hiring however.\nEF Personas This is highly subjective, but also the most common question I got was “what’s the crowd like?”. So here is my very poor and biased attempt at answering it. Keep in mind that the “common weaknesses/strengths” are extremely subjective and biased.\nType in EF or the cohort name (e.g. BE9, LD5) into LinkedIn and you will find the kind of profiles who go through EF.\nWho goes to EF? EF attracts ambitious ~25-35 year olds (in my cohort it was min 23, max 42). My gut feeling, after looking at 4 cohorts (Berlin, London, Paris, Toronto) is that EF cohorts have a similar composition. I would categorise these as follows:\nEx-Founders 20% Both, business and tech. People who already have built some kind of company, usually a small startup, and sometimes more traditional companies with low potential to scale. Mostly either small failed startups, side hustles, or traditional non-scalable businesses. In that mix, there are some impressive ones too. For instance, one guy scaled his company to 30 employees and reached 100% market saturation with a small gadgets business. There is someone who bootstrapped a proptech app and had hundreds of paying customers. I’ve not met anyone whose business was Series A funded.\nMcKinsey types 30% They’ve spent a few years in a top consulting firm. Maybe the went to a fancy business school or studied econ or finance. Many of them worked in business roles at startups as well.\nDevs 30% Just good devs, with a decent mix of specialised and full-stack. You could probably find someone for most ideas.\nAcademics 20% The hyper-specialists with deep insider knowledge. Examples: lawyer with 15 experience in a very specific legal field. Examples PhD in batteries, PhD in genomics.\nApplication A large proportion of EF cohort members come from direct recommendations or from being headhunted by the EF team. It probably makes sense to check who of your LinkedIn connections went through EF and ask for a recommendation. Being recommended or being headhunted has a strong correlation with being successful at EF.\nEF describes what they’re looking for here. At the same time many of these terms seem rather vague. After all, everyone says they want ‘out-of-the-box’ thinking (whatever that means 🤷 ), and EF’s framework does not challenge convention in many ways. EF is set up on a rigid paradigm that seems optimised for reducing bad companies, at the cost of sometimes foregoing truly great companies. This is natural, every heuristic has to balance false negatives and false positives.\nSo let me go through the ones that I noticed at EF:\nFollowership: weird way of saying leadership. You rally people behind a cause. Maybe at your university you ran a charity fashion show, maybe you have a large newsletter. Clarity of thought. This is crucial. Here are a few tips: Use short sentences. This forces you to pick words clearly. Structure paragraphs clearly. Make your point, then present the evidence, then end with what this means for the overall story you’re telling. Before responding to a question, rephrase it and ask whether you’ve understood it correctly. Technicality \u0026amp; Commerciality: this is the basic for EF’s CEO/CTO dichotomy. You should know how to build or how to sell and manage. Maybe you won several hackathons or you were a dev at some important company. Maybe you worked at McKinsey or went to a fancy business school. If you have friends who were at EF, ask them to share their application. If you don’t, just do some cold outreach to EF alumni. There will be many keen to help. Ask them to share theirs and review yours.\nKonrad’s EF Application Berlin 2021\nThis is my application. My application is not very relevant to everyone, because I had founded a startup before. Following this application, I had three 30 minute interviews. In these interviews, pretty much the same questions were repeated and my answers were pretty much the same. First, I thought this was to check for consistency (to weed out liars), but I know think it’s because the EF team doesn’t have time to prepare for these calls. They seem to be back-to-back, so you start the conversation from scratch each time (that seems to be a general problem at EF). These conversations were very casual and we just went a bit deeper into the points we talked about.\nHow selective is EF? I think the level is generally lower for technical founders as in my cohort there were several with zero or very little professional experience. These shined by showing leadership at university and studying at fancy universities for computer science. At the same time, business cofounders were generally more senior with several years at consulting firms or having built a company before. This makes sense given the shortage of tech talent.\nThe statistics they share are vanity metrics. Here is a pipeline I saw:\n9000 conversations 600 applications 184 interviews 48 selected That makes it look like a 0.05% conversion rate, which sounds super exclusive. However, what does conversations mean and why is the drop-off so high? I assume it’s something like looking at a profile. Applications are not a strong selection either, the application takes \u0026lt;1h to complete, so I assume the quality will all over the place. I would say that the application-to-interview ratio is more relevant. That being said, I know of some surprising rejections and some surprising acceptances.\nKeep in mind that these funnel shapes are never a good metric for selectiveness, because it doesn’t say anything about prequalification.\nThe EF Framework I can’t write about Antler, because Idk. I invite anyone who has been through Antler, to collaborate on this section.\nYou can read all about it here. This is a great reading list and along with YC content will teach you a lot about startups. Don’t binge read this this stuff, some seemingly simple concepts really need time to digest. E.g. scalability is a concept that seems simple but one could write books about.\nI’ll very briefly go over one that is only briefly mentioned in this database, but that I find very useful.\nNon-obvious beliefs EF encourages you to think about your past experiences and what insights you’ve gained there. Ideally, you should made some surprising insights, that someone who is vaguely acquainted with the field would have no way of knowing after a few minutes of research.\nExamples of obvious beliefs:\nMillennials have a hard time buying real estate Teenagers spend too much time on their phones Genomics is getting cheaper (even if you didn’t know, anyone can google this in 1 minute) Examples of non-obvious beliefs:\nGood teachers are coaches primarily, there is little knowledge transfer 2nd hand batteries would be 50% more efficient, if they were stored properly Many defi products fail, because it’s very hard to properly test solidity code Essentially, don’t work on stuff that you have no experience in. It’s usually very hard to solve a problem better than the people who face them every day.\nThe idea here is to weed out all the startup ideas that everyone has or can have. People tend to come up with similar ideas all the time. Airbnb for X, fractional real estate investing, investing with friends, an AI that does legal work, etc. Usually, these are bad startup ideas.\nIt takes some experience in the space to realise the real paint points. E.g. AI can do legal work but it’s more about document meta data and tags, than figuring out the legal logic. It takes some legal experience to know this.\nTimelines I can’t write about Antler, because Idk. I invite anyone who has been through Antler, to collaborate on this section.\nThroughout the program, there are many workshops and talks. These are your usual accelerator talk. Sometimes great, usually mediocre. IRL talks are not a good way to learn, but workshops definitely are. Make sure to attend workshops. Talks you can probably skip and just read. YC has great content. Reading EF materials carefully and deeply is also a great source of knowledge and thought.\nWhat Happens at EF Month 0 This is a crucial month that EF haven’t told us anything about. Essentially, EF released the cohort list sometime early March, but did not tell you that’s when the programme starts. They claim it starts with Week 1 (two weeks after Kick-Off weekend). This is not true. As soon as the cohort lists are released, people start networking. This is a very important time because people are already chatting and gauging with whom to work.\nKick-Off Weekend We had some speed networking sessions, some dinners and drinks. It was an intense weekend and you had the opportunity to meet 50 people or so. Definitely a must-go.\nThe two weeks before the start of the programme and the kick-off weekend are crucial as well. People meet up for coffees and drinks and already start networking intensively.\nMonths 1-2 Team formation time. It’s like love island, but for founders. People team up, split up, there is drama. They make you do weird posts on Slack where you compliment each other and give reasons for your break ups, then people react with weird emojis to them. It’s all a bit bizarre but seems to work.\nMonth 3 A lot of people seem to be in a rush to co-found at this point. It’s a little bit of a last attempt at EF. It’s important not to get overwhelmed by FOMO.\nIC Essentially, you pitch to the investment committee. They seem to judge you primarily on the data they gathered before the IC, on your progress etc. Keep in mind that they gather A LOT of data.\nI am not qualified to talk about IC, because I dropped out before and found a co-founder outside of the program. I can’t write about IC, because Idk. I invite anyone who has been through IC (both accepted/rejected), to collaborate on this section.\nThings I didn’t know but wished I knew Timeline EF does not communicate the point of the start of the program very well. It’s crucial to understand that the program starts the very moment they release the candidate profiles. People will immediately start networking and prequalifying teams. This effectively means that the programme starts roughly 1 month before the official start and 2 weeks before the Kick-Off Weekend.\nPeople who join late are at a huge disadvantage, so don’t let them tell you that it doesn’t matter if you’re late. It’s false. In my case, I was heavily involved in humanitarian aid for Ukrainian refugees, the largest migration crisis since WW2, and was not in a mental or logistical place to start the program. However, I was assured that the start would not matter too much. It does matter a lot. There were other people who joined very late and they did not find good matches.\nThe first weeks are absolutely crucial. Do all your self-discovery before the program. I should have asked myself things like:\nwho do I want to work with what are the topics that excite me what are the values I care about who do I want to hang out with Pre-investment \u0026amp; Post-investment / Advisors vs VCs What do I mean by this distinction?\nAdvisors \u0026amp; angels: specialists or more senior execs who have some skin in the game with your company, usually equity.\nVC investors: professionals who sift through thousands of companies looking for investments that will return 100X.\nThe big difference is the kind of things you talk about to both. VCs don’t have skin in the game before they invest, they have no reason to help you before they invest. There is a careful balance of telling them about problems and actually making things look investable. Everyone knows that you’re likely to pivot a million times and that you know very little. However, if already at pitching stage there are major problems, such as a very small total addressable market, or a bad team-company fit, then you’re not investable.\nOnce they invest however, incentives change drastically and now both of you have a big interest in helping each other. After the investment VCs have skin in the game and if they’re able to increase your valuation by 2X, they will be 2X better off.\nFor instance, sharing any kind of issues such as doubts about team fit, mental health or family issues pre-IC will be extremely detrimental to whether you’re investable. This is super hard at EF, because you meet the team and become friends with them, so it’s natural to share worries etc. However, remember not to treat them as advisors or mentors before they invest.\nTransparency Expect that everything you share with EF team members, will end up in their databases, so treat them as bad HR interviews. This is crappy, because in 1:1 conversations, there is a reasonable expectation of privacy. Expect that everything that EF people can get their hands on, will end up in their database, which they will use to evaluate whether you are investable.\nIn other words, talk to them as you would to a VC, do not share problems with EF, unless these are problems that an investor should now about. So, for instance, do not talk about emotional worries, mental health, potential disputes with your cofounder. They will ask about these things, you have no obligation to tell them.\nHere is an example scenario. This actually happened to me. EF team member has a 1:1 check in with you and asks “how productive have you been with your co-founder last week?”. Both of you are honest and say that you went off for a holiday and weren’t productive at all. This would be a totally normal response to other founders, angels or a VCs who has already invested, yet to a pre-investment VC that’s a red flag.\nExploration vs Exploitation All choices face the exploration vs exploitation dilemma. Do you spend more time getting a clearer picture of the choices you face, or do you stick to one without knowing\nImagine that you’re looking for gold. You can either keep digging where you last found some small gold veins and get a constant payout of 10-20 nuggets per day, or explore more dirt at 0 nuggets per day, hoping that you will find a better vein.\nHere is a weird thing about choice. Imagine you have three buttons, pressing them gives the following random pay-outs : A → $10-20, B → $12-22, C → $9-14. You can press a total of 100 hundred times and you want to maximise your pay-out. The dominant strategy would be to press a number of times (stats nerds can figure it out exactly), say, 6 times on each, get the averages, and then keep pressing on the one that yields the most. Both computers and humans act the same way here. However, introduce an additional rule: after not pressing a button for 10 presses, it disappears. What would you do? Rationally, you should proceed the same way. After all, after your first 24 rounds, you stopped exploring and start exploiting. However, people don’t like disappearing options and will waste their precious presses on lower-yielding buttons. That’s what people do in dating, cofounder, job, shopping search, etc. People love to hedge their bets. However, for very serious choices, it might be detrimental. You can’t explore 2 co-founder relationships at once.\nAnyway, back to EF. You will face something like 25 CEO/CTO complimentary cofounder candidates. Perhaps 10 of them will be a good fit in terms of your areas of interest. That’s a lot of candidates to go through. You won’t be able to test deep and meaningful work with all of them. You will be out of sync in terms when you’re in the solo founder pool a lot. You’ll get maybe 3-4 shots of trying out a week or two with someone, and perhaps 5-10 chances of trying out a few hours.\nHere are the costs of team switching:\nAnnouncement costs: every time you form a team formally, everyone gets a slack message and people ask you about it. That creates some pressure to stick to it Break up costs: every time you break up, you write a slack message with a mini review (always praising) your ex-would-be-cofounder. This results in everyone asking questions etc. This takes up time. Project management costs. Every time you switch, you have to have conversations about working hours, preferred methods of communication, tool stacks, division of AoRs, setting goals, etc. Emotional cost. It’s a bit like dating, so chances are you start getting emotionally involved in fantasies of success, what your life might look life later etc. Finding the Right Cofounder It’s hard. Don’t rush it.\nCheck out my substack article on co-founder fit questions.\nAlternative ways of finding co-founders:\nYC Startup School Cofounder Matching Hackathons Friends Other Voices 🙏🏼 If you wrote something on EF, please DM me https://twitter.com/0xkkonrad or here https://www.linkedin.com/in/konrad-urban/\nMy journey at Entrepreneur First Berlin\nUNLICENSE, because fuck copyright. You can’t own ideas. Attribute me if you want to. It’d be nice.\n","permalink":"http://kkonrad.com/writing/cofounder-search/ef-antler-honest-guide/","summary":"An honest, opinionated walkthrough of EF and Antler — what they actually are, who goes, the deal, and the things I wish I\u0026rsquo;d known.","title":"Entrepreneur First \u0026 Antler: Konrad's Honest* Guide"},{"content":"Konrad Urban. This is a draft and work in progress. Feedback welcome.\nThe Knowledge Argument argues that non-physical properties in conscious experiences exist. Frank Jackson (1982) constructed a thought experiment featuring Mary. Mary is a brilliant colour scientist and neurophysiologist. All her life, she has been confined to a completely colourless room. She possesses all physical information about colour vision (wavelengths of colours, which neural connections are at work under which conditions etc.). Jackson claims that Mary would learn something new after her release from the room and first experience of colour. More formally,\n(KA P1) Mary knows all the physical facts concerning human colour vision before her release. (KA P2) But there are some facts about human colour vision that Mary does not know before her release. (KA C1) Therefore, there are non-physical facts concerning human colour vision. (Nida-Rümelin, 2009)\nAn influential strategy against the Knowledge Argument rests on the Ability Hypothesis (Lewis, 2004; Nemirow, 2007, 2008). It states that knowing what an experience is like is an ability (e.g. to remember, recognise etc.). If the Ability Hypothesis is true, then Mary does not gain new knowledge but a new ability. This saves physicalism from the Mary case. I would like to show that ability of a different kind is a problem for the Knowledge Argument: Mary is unable to know all relevant physical facts (KA P1 is impossible). I am not defending physicalism, I am only stating that either Mary is impossible and should not be used as a case in either direction, or (if construed such that Mary is possible) the case collapses into a different case, better described by Thomas Nagel\u0026rsquo;s problem of objective/subjective perspective (1974).\nA lot of attention has been directed at what Mary\u0026rsquo;s case entails and how misguided our initial intuitions are. For instance, Daniel Dennett argues that Mary\u0026rsquo;s is not readily imaginable, and that Mary\u0026rsquo;s case is bound to misguide us because of that (Dennett, 1993, p.399; Robinson, 1993). He also argues that a proper understanding of Mary would lead us to believe that Mary already had been able to imagine everything about colours before her release (Dennett, 2007). George Graham and Terence Hogan (2000) claim that she would not be able to imagine everything, leaving some unimagined phenomenal residue the content of which would be new to her after being exposed to colour. The defining point of the debate is to be able to tell what Mary is able to know before her release. I agree with Dennett\u0026rsquo;s preliminary analysis that Mary\u0026rsquo;s case is not readily imaginable. I agree that the case is constructed to mislead intuitions: Jackson\u0026rsquo;s misleading \u0026ldquo;brilliant scientist\u0026rdquo; is actually Super Mary, at least a demi-god, with abilities far beyond humans'.\nExperimentally describing Mary\u0026rsquo;s case in a different terminology, such as demi-god, daemon or Super Mary, would be illuminating. In fact, I am convinced that if the Jackson\u0026rsquo;s version used such descriptors instead of human(oid) Mary, initial intuitions would not be anti-physicalist. As much fun \u0026ldquo;Super Mary\u0026rdquo; would be, it is probably enough to mention this and stick to her original name, but disentangle what Jackson demands of us. My motivation is different from Dennett\u0026rsquo;s. I hope to show that Mary\u0026rsquo;s case is not readily imaginable because knowing all relevant physical facts is conceptually impossible. The brains Mary can know all physical facts about are necessarily simpler than hers, which collapses the case into Nagel\u0026rsquo;s problem of perspectives.\nFirst, I will show that \u0026ldquo;all physical facts relevant to colour vision\u0026rdquo; either must include all facts about Mary\u0026rsquo;s brain state when experiencing colour, or the thought experiment is unwarranted as it collapses into Nagel\u0026rsquo;s qualia problem (1). Then, I hope to show that Mary must be able to operate on the facts she knows in a specific way that I call emulation (1.2). Then, I will explain why an informationally poorer system cannot emulate (perfectly simulate) an informationally richer system (2). Finally, I will show that Mary\u0026rsquo;s brain cannot emulate itself, because it would have to be informationally richer than itself (3). I shall conclude that: either the Mary case is conceptually impossible and therefore should have no weight in supporting physicalism or anti-physicalism, or it collapses into Nagel\u0026rsquo;s qualia problem.\n1 Knowledge Mary investigating simpler brains (such as human brains) would prevent the Knowledge Argument from working. Mary must be able to emulate her brain because it is Mary\u0026rsquo;s phenomenal experience that she is after, not that of some simpler brain. Simply put, if Mary were to analyse only her visual cortex, she would not be analysing Mary\u0026rsquo;s experience. On a dialectical level, if she were after the phenomenal experience of a simpler brain, the whole case would collapse into Thomas Nagel\u0026rsquo;s subjective/objective gap in \u0026ldquo;What Is it Like to Be a Bat?\u0026rdquo; (1974). This posits that the jump from subjective to objective is impossible. Conversely, if Mary were a non-human demi-god with full physical knowledge about human brains, her question would be about what it\u0026rsquo;s like to see red for humans, not whether there are new (physical or non-physical) facts. After all, Jackson asks whether pre-release Mary would know what it\u0026rsquo;s like to see red. He does not ask whether she knows what it\u0026rsquo;s like for something else (e.g. a simpler brain) to see red. This is not a problem for the physicalist, but it makes Mary dialectically superfluous (and more confusing) as a case separate to Nagel\u0026rsquo;s problem of perspective.\nA key term in my discussion is \u0026ldquo;emulation\u0026rdquo;, a term I borrow from computing. Let an emulation be perfect simulations or perfect mental representations of object. An emulated brain is not just a model of a brain, because models necessarily simplify (Cartwright, Shomar, \u0026amp; Suárez, 1995). An emulated brain is not just a simulation, because simulations can use representational shortcuts (e.g. a simulation of a billiard game does not need to contain representations on the quantum level given that observable billiard-relevant results are equivalent).\nI avoid familiar terms like \u0026ldquo;imagining\u0026rdquo;, because Mary\u0026rsquo;s mental processes are unfamiliar to us. We usually approximate, model and abstract, losing data in favour of comprehension. For instance, redness could be described by some cultural model (\u0026ldquo;a symbol of aggression\u0026rdquo;) or some physical model (\u0026ldquo;wavelengths of 620-700nm\u0026rdquo;). Neither of these theories can cover the content of the other. Mary, by design, never faces this payoff, because her knowledge is full. In Parts 1.1–2, I will elaborate on what knowledge Mary must possess for the Knowledge Argument to make sense.\n1.1 The Scope of Relevant Knowledge Mary is defined to know all physical facts relevant to colour vision. The key word is \u0026ldquo;relevant\u0026rdquo;. Depending on commitments in metaphysics and theory of relevancy, Mary must have, minimally, knowledge about Mary\u0026rsquo;s brain states when exposed to colours, or maximally, knowledge about all physical facts including possible worlds.\nThe minimum cannot be reduced, because the Mary case needs at least this result for its intended anti-physicalist intuitions. It is about the phenomenal experience of being someone like Mary and seeing colour for the first time. Despite all her factual knowledge, she was unable to imagine what it\u0026rsquo;s like. This minimum is of course implausibly optimistic, because it is unlikely that brain states alone fix phenomenal experiences, if they exist. Furthermore, counterfactual theories of causation such as David Lewis\u0026rsquo; (Menzies, 2009) require possible worlds analysis. Under such commitments, Mary must also know about them. For my argument, we only must accept the plausible premise Mary must at least know the state of her brain when experiencing colours. However, most likely, she must know much more.\n1.2 Knowledge and Emulation For full knowledge, all relevant facts must be known — a truism underappreciated ever since Dennett\u0026rsquo;s case (Dennett, 1993, p.399–401). If Mary had only physical (or encyclopaedic) knowledge without full understanding, the Knowledge Argument would not work. The physicalist could easily argue that her \u0026lsquo;phenomenal\u0026rsquo; experience are just emergent physical properties that are beyond Mary\u0026rsquo;s comprehension (rendering KA C1 a disjunctive: there exist physical facts or facts inaccessible to Mary). For instance, knowing all the coordinates of all trees in Europe does little to establish knowledge of the Bialowieza forest. This would require the cognitive power to map such a list and applying auxiliary knowledge of how forests emerge from trees. For full knowledge of the forest, one would have to effectively reproduce the forest in their mind. This is what I refer to as emulation.\n2 Emulation Capacities A key premise of my argument is that an informationally poorer system cannot emulate an informationally richer system. Imagine a binary two-node net where every second nodes randomise their states. This allows for four two-bit states, A={{0,0},{0,1},{1,0},{1,0}}. At any second, the network only produces two bits (e.g. {0,1}). However, to emulate the network one needs information beyond two bits and beyond A (8 bits). The emulator must at least include the procedures of the emulated network. Because of these auxiliaries the emulator is informationally richer than the emulated network (call this the Auxiliaries Assumption). Depending on one\u0026rsquo;s metaphysical commitments (1.1), the auxiliaries can be a much larger set than S, making my case even more intuitive.\nBut what does this have to do with Mary? Let us use an implausibly minimal example to make the case. Let Mary-Emulator be the system capable of emulating Mary\u0026rsquo;s brain. I hope to show that Mary-Emulator must be informationally richer than Mary\u0026rsquo;s brain. Recall (1.1–2), Mary\u0026rsquo;s knowledge must at least encompass knowledge of the state S, i.e. the state her brain would be in if she were to experience colours. S can be produced by Mary-Emulator, just as elements of A were produced by our previous emulator. To do that, Mary-Emulator must invoke the same procedures that guide colour vision in Mary\u0026rsquo;s world and apply them to an emulated version of Mary\u0026rsquo;s brain.\nSo far, the Knowledge Argument remains undisturbed by the Mary-Emulator. After all, Mary needs something like the Mary-Emulator to have full knowledge of all physical facts relevant to colour vision. However, what the Knowledge Argument needs is to let Mary-Emulator be Mary\u0026rsquo;s brain. Otherwise, as discussed previously, Mary would be investigating a different brain, which collapses the case into Nagel\u0026rsquo;s perspective problem.\nNote that my argument is neutral about multiple realizability, the idea that mental kinds can be realised by various physical kinds (Bickle, 2016). Brains in my argument could refer to brain tokens but also to types realised in different ways as tokens. Even when I say \u0026ldquo;Mary\u0026rsquo;s brain\u0026rdquo; I can refer to any system that realises Mary\u0026rsquo;s brain; depending on commitments this could be fixed by what it\u0026rsquo;s like to be Mary, brain-related properties, powers etc. What matters for my argument is the informational richness of the brain, i.e. how much information there is about it metaphysically (not how much it stores). My argument relies on the assumption that for any brain, the informational richness is always the same (call this the Equality Assumption). So, whether realised in silicone, neurons or magnets, whether largely active or inactive, the information richness about whatever makes it a brain is equal.\nThe Equality Assumption implies that the informational richness of all of Mary\u0026rsquo;s brain is equal. It may seem that some states contain more information. For example, say, in state B1, Mary\u0026rsquo;s brain is mostly inactive and state B2 it is very active. It might seem that B1 is informationally poorer, but that is only because B1 could be described using less information, not because there is less information about it. E.g. set C={1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16 … 99} could be described as C = {x|x∈N, x \u0026lt; 100}, which is much shorter. Yet, both sets are the same. The mistake here is conflating description with content. I claim that set of carbon atoms has the same amount of information no matter whether it is arranged into a neat crystal that is elegantly describable or chaotically arranged. In both cases, there is still information about every atom. Conversely, there is information about every neuron.\n3 Mary-Emulator\u0026rsquo;s Capacities Mary-Emulator cannot be contained in Mary\u0026rsquo;s brain, because it is informationally richer. Recall, minimally, Mary-Emulator must produce a brain in S. Producing S can be done either by externally constructing or by adopting S. These distinguish the hotly debated third person (objective/perspectival) and first person (subjective/sympathetic) perspectives, i.e. Nagel\u0026rsquo;s objective/subjective distinction. Perspectival emulation views the brain from an external perspective, not unlike a scientist. Sympathetic emulation simply puts the emulator in the emulated state with the information that it is an emulation. According to Dennett, Mary can emulate sympathetically and be unsurprised by colour after her release as she had a priori access to the experience (2007). Regardless of whether these sympathetic emulations are possible, I will deny that any emulations of any of Mary\u0026rsquo;s brain states are possible for Mary.\nIn both cases, Mary-Emulator\u0026rsquo;s information must contain S plus all the relevant auxiliary facts about emulating (call this the Auxiliaries Assumption), e.g. that it is an emulator. Without these facts, the sympathetic emulator would simply be a copy, whilst the perspectival emulator would not know how to interpret the data.\nMary\u0026rsquo;s brain cannot be the Mary-Emulator, because it would be a set of its size plus something. I will show that Mary\u0026rsquo;s brain cannot be Mary-Emulator by a reductio ad absurdum. Were Mary\u0026rsquo;s brain to be Mary-Emulator, it would be in the state of Emulating-Mary\u0026rsquo;s-Brain. Yet, S is a brain state of Mary\u0026rsquo;s already. Both, S and any other Mary brain states (including Emulating-Mary\u0026rsquo;s-Brain-State), are informationally equally rich, because both are confined Mary\u0026rsquo;s brain and thus there is as much information about them. More generally, S and any other brain state (including any state when she emulates her brain) of Mary\u0026rsquo;s are informationally equally rich. Yet, we have shown that emulator must contain more information than the emulated system, which leads to a contradiction (with S and Emulating-Mary\u0026rsquo;s-Brain being informationally equally rich). More colloquially, Mary cannot create a conceptual copy (emulation) of her brain, because such a copy would require more brain than Mary has.\n4 Conclusion I hope to have shown that Mary cannot emulate her own brain. After all, even a minimally conceived set of Mary\u0026rsquo;s knowledge must include all physical facts about Mary\u0026rsquo;s brain. To fully know Mary must emulate her own brain. Yet, an informationally poorer system cannot emulate an informationally richer system, and a system emulating Mary\u0026rsquo;s brain is informationally richer than Mary\u0026rsquo;s brain. It follows that Mary cannot know all the relevant physical facts (rendering KA P1 impossible). However, if Mary were to look at simpler brains to make the case work, the thought experiment would add nothing new to Nagel\u0026rsquo;s subjective/objective problem, making it a better pivot for the debate. Unfortunately, we will never hear whether (Super) Mary was surprised to see red, because she is conceptually impossible.\n5 Bibliography Bickle, J. (2016). Multiple Realizability. Stanford Encyclopedia of Philosophy, (Spring).\nCartwright, N., Shomar, T., \u0026amp; Suárez, M. (1995). The Tool‐Box of Science. In Theories and Models in Scientific Processes [Poznań Studies in the Philosophy of the Sciences and the Humanities 44] (pp. 137–149).\nDennett, D. (1993). Consciousness Explained. London: Penguin.\nDennett, D. (2007). What RoboMary Knows. In Phenomenal Concepts and Phenomenal Knowledge: New Essays on Consciousness and Physicalism. https://doi.org/10.1093/acprof:oso/9780195171655.003.0001\nGraham, G., \u0026amp; Horgan, T. (2000). Mary Mary, Quite Contrary. Philosophical Studies, 99(1), 59–87.\nJackson, F. (1982). Epiphenomenal Qualia. The Philosophical Quarterly, 32(127), 127. https://doi.org/10.2307/2960077\nLewis, D. (2004). What Experience Teaches. There\u0026rsquo;s Something about Mary: Essays on Phenomenal Consciousness, 77–103. https://doi.org/10.1017/CBO9780511625343.018\nMenzies, P. (2009). Counterfactual Theories of Causation. The Stanford Encyclopedia of Philosophy, 2011(Dec. 16), 1–47. Retrieved from http://plato.stanford.edu/archives/fall2009/entries/causation-counterfactual/\nNagel, T. (1974). What Is It Like to Be a Bat? The Philosophical Review, 83(4), 435. https://doi.org/10.2307/2183914\nNemirow, L. (2007). A Defense of the Ability Hypothesis. In Phenomenal Concepts and Phenomenal Knowledge: New Essays on Consciousness and Physicalism. https://doi.org/10.1093/acprof:oso/9780195171655.003.0002\nNemirow, L. (2008). Physicalism and the Cognitive Role of Acquaintance. In Mind and Cognition: An Anthology (pp. 490–499).\nNida-Rümelin, M. (2009). qualia: the knowledge argument. In Stanford Encyclopedia of Philosophy (pp. 1–19).\nRobinson, H. (1993). Dennett on the knowledge argument. Analysis (United Kingdom), 53(3), 174–177. https://doi.org/10.1093/analys/53.3.174\n","permalink":"http://kkonrad.com/writing/philosophy/super-mary-mission-impossible/","summary":"The Knowledge Argument\u0026rsquo;s Mary is conceptually impossible — any system capable of emulating her brain must be informationally richer than her brain, making KA P1 self-defeating.","title":"Super Mary: Mission Impossible"},{"content":"Konrad Urban\nAbstract The paper assumes that conformism is an appropriate response to peer disagreement in idealised cases. It develops a dynamic model that allows to approximate from ordinary and rough cases towards idealised cases (given same cognitive ability) by expanding the evidential set of the doxastic agents through disclosure of evidence. This model is applied to five frequently cited cases that intuitively demand either a conformist response or a response of steadfastness. It succeeds to adequately explain why in the given cases the response should be conformist or steadfast.\nThe Motivation There is no universal response to all cases of disagreement. Traditionally, two responses to cases of peer disagreement have been given: conformism or non-conformism. Conformism (or conciliationism or equal weight view): peers who disagree about a proposition should lower their confidence or suspend judgement (see Feldman 2007; Elga 2007; Christensen 2009). Non-conformism (non-conciliationist or steadfast views), on which there is no such requirement (see Kelly 2005; Enoch 2010; Foley 2001)). Arguably, conformism provides a good account in scenarios with ideal epistemic peers but fails in other cases. This paper will defend conformism in these other cases when it is possible to approximate rough peers into ideal peers.\nIn short, you have an epistemic duty to have informed believes, so you have an epistemic duty to collect evidence. To be able to weigh the evidence of peer disagreement you should understand the disagreement. Only once you understand the disagreement, you can react appropriately.\nLimitations In this paper, I shall take a weak understanding of conformism. Conformism requires to suspend one\u0026rsquo;s belief or lower one\u0026rsquo;s confidence in that belief when faced with peer disagreement. In other words, one cannot rationally hold steadfast to one\u0026rsquo;s belief (or confidence in it) when faced with peer disagreement.\nThe literature varies on the meaning of \u0026ldquo;epistemic peer\u0026rdquo;. Generally, epistemic peers have similar cognitive abilities and similar evidence for the relevant belief. Some scholars put strict limits on this and a distinction between idealised and ordinary cases is often made1. In idealised cases, evidence is identical and cognitive ability is equal2. Ordinary cases allow for vague identity and equality. Ideal cases are rare or non-existent (King 2012). A classic case involves two epistemic peers getting two different results when trying to split the bill in a restaurant (Christensen 2009). Their relevant evidence (the bill) is the same and their cognitive ability is assumed to be the same, yet they have found two distinct solutions to this simple arithmetic problem. Should they suspend or lower confidence in their belief (conformism), or should they hold onto it (steadfastness)?\nIn this paper, I assume that uniqueness thesis is true: given a reliable method of reasoning for all rational doxastic agents, the same set of evidence must produce the same belief (Feldman 2007). It can be one of the strongest motivations for conformism3. Given the uniqueness thesis, peer disagreement (i.e. non-unique solution given the same evidential set) is strong evidence for one of the agents having reasoned badly. Given this evidence about both agents\u0026rsquo; (as this evidence is symmetric) bad reasoning, one must suspend belief or lower one\u0026rsquo;s credence in that belief.\nFour Cases I shall briefly discuss four scenarios that reoccur in the literature on peer disagreement. I will take the intuitions on the appropriate attitude as granted. Conformism as it stands fails to generalise onto some of these cases:\nRestaurant with Mental Maths: Two epistemic peers want to split a restaurant bill. They both arrive at different results using mental maths (Christensen 2009).\nRestaurant with Calculator: Same as above but calculator is used (Christensen 2009).\nHorse Race case: We are both at a horse race. I see the horse pass the finish line first. You see the horse pass second. We both paid equal attention (Elga 2007).\nDavid Lewis case: David Lewis is my epistemic superior (he has better cognitive ability and has more relevant evidence). He believes in real possible worlds and I don\u0026rsquo;t (Frances 2010).\nI have included the David Lewis case in the discussion about peer disagreement because it is a stronger case than peer disagreement. If a peer\u0026rsquo;s disagreement is evidence for unreliable reasoning (or incomplete evidence – as can could be in this case), then it is even stronger in the case of peers.\nIn the literature, these cases were brought up to argue for either steadfastness or conformism. The Mental Math and Horse Race cases were brought up to argue for conformism – if the results of mental math or observation are different, then surely at least one agent has made an error and thus belief should be suspended, or credence lowered. The Calculator case was brought up as a counter-example to this, but was heavily debated – after all, I can rule out that I am not dishonest, but I can\u0026rsquo;t rule out that you are not, so I should trust myself more (cf. Christensen 2009). The Horse Race case is meant to support the steadfast attitude, claiming that my sight is more direct evidence than Alice\u0026rsquo;s claim about her sight. The David Lewis case was brought up to show that extending the conformist stance to all beliefs was too demanding or counter-intuitive. After all, one would have to lower credence or suspend belief about all controversial matters where peers held incompatible views, e.g. in religion, philosophy or politics. Let\u0026rsquo;s accept that three of the cases support either the steadfast view or the conformist view.\nIn more recent literature, this opposition has been broken into several mixed views incorporating the above-mentioned intuitions into one ruleset or denying universalisation (Goldman and Blanchard 2016; Matheson 2015). Instead of discussing their failings, I construct an alternative explanatory and normative model. It should fully explain the intuitions of the abovementioned cases. Its downside is that it is – although more unified – more demanding on the agents than some of the other solutions.\nTowards a Dynamic Model Some literature suggests that agents ought to fully disclose their evidence (Bergmann 2009). One reason is evaluating whether they have reasoned using the same set of evidence (to evaluate peerage). Another reason is locating potential errors in the reasoning itself. If the set of evidence changes after full disclosure, then it allows for a dynamic model of peer disagreement. Of course, in such a case the agents were only rough (not ideal) peers, but they can move towards idealised peerage by sharing their evidence. My dynamic model will hopefully give a procedure for cases of disagreement and tell us whether (and when) conformism should be abandoned.\nSo far, almost all models of disagreement have been static, that is, they required or allowed for a static equilibrium without changing circumstance. Hopefully, it will become evident that this is why some fail to generalise. There have been some attempts at a dynamic model by Douven (2010)4. I shall go in a similar direction, but with a stronger emphasis on locating the disagreement in the web of evidence and belief. The model will show when conformism is appropriate, controlling for what counts as peerage.\nDisclosure and Entanglement The concept that beliefs are entangled has been mentioned5, but it has not been appropriately applied to evidence. The literature on peer disagreement so far has assumed that a given set of evidence entails a belief, and revolved peerage and reliable entailment or what counts as belief (Matheson 2015; Goldman and Blanchard 2016). Little attention has been paid to the entanglement of evidence and belief. Drawing a parallel from the holistic science thesis (Stanford 2013), evidence is not simple (see Moffett 2007 for an attack on conformism in this spirit). It can only point towards a belief if it is interpreted with the use of auxiliary beliefs. For instance, my evidence of the restaurant bill will point to my result only with the use of my auxiliary belief about basic arithmetic – for which I have evidence. Such beliefs are also entangled with other beliefs and so on. In a simple case like the restaurant bill, both agents have reasons to belief that their auxiliary beliefs will be the same. For beliefs that are result of very complex deliberation, it is unlikely that doxastic agents will share auxiliary beliefs – even in cases of agreement.\nIf this is indeed how evidence and beliefs work, then disagreement does not challenge a particular belief, e.g. the reality of possible worlds, but a whole web of entangled beliefs. The disagreement itself does not reveal the location in that web. Why does this matter? If two agents work under different auxiliary beliefs, then they are not ideal peers and their disagreement reveals little about the proposition at hand. For instance, if I believe in solipsism and David Lewis beliefs that people exist, then the fact that we disagree about the proposition that real possible worlds exist is no evidence about said proposition. Similarly, say, we have a different understanding of addition and we disagree about the restaurant bill. If we don\u0026rsquo;t know about each other\u0026rsquo;s auxiliary beliefs, then all the disagreement does is show that our entangled webs of relevant beliefs are non-identical. If we know the auxiliary beliefs are different, then the discovery of disagreement seems to give no reason to modify the disagreed upon belief or its credence. If anything, it invites an evaluation of the auxiliary beliefs.\nRounds of Disclosure We have shown that beliefs are entangled, and that disagreement is a challenge to the web of beliefs, not to the isolated belief that manifested the disagreement. This means that the precise location of the disagreement remains initially unknown. When agents disclose their evidence and their auxiliary beliefs, they see the relevant disagreement travel (disagreement overall expands, but the relevant disagreement for each round travels). In the aforementioned example, it travels from the disagreement about the restaurant bill split to the disagreement about how arithmetic and addition work. Rounds of disclosing evidence and beliefs should continue, letting the initial disagreement travel and locating new disagreements (in which case the discussion branches off into a new series of rounds). Rounds continue until either agreement is reached or disclosure cannot continue because of fundamental beliefs (beliefs not based on evidence or reasoning, such as intuition-based beliefs). The conditions for agreement are dictated by more general rules of epistemology, not necessarily specific to social epistemology. In the case of fundamental beliefs, they can of course be irrational, but, again, this is relevant to other areas of epistemology.\nMany rounds can be required as these webs of beliefs and evidence can be extremely deep. So far, we have held cognitive ability constant and equal for peers. This means that both peers will reach a similar level of depth before entanglement overwhelms their brainpower. When the web is too deep, it becomes easy make errors and there will be a point where an error is more likely than furthering understanding about the disagreement. Because the agents know that they will be able to make a common judgement when to cease disclosing. At this point, they can give up disclosing and hold steadfast to the belief from the most recent round.\nThe ever-increasing complexity of system can be demanding in terms of time and brainpower. This means that other considerations will often override the epistemic duty to locate the disagreement at fundamental level. For instance, perhaps I should not continue rounds of disclosure about a political belief at a dinner party as to not bore everyone. Other duties can override epistemic duty.\nThis model allows for modifying beliefs with each round – not because of conformism as such but because of expanded evidence (e.g. in the form of testimony). This means that the disagreement can travel in an ever-expanding web, but also that new disagreements can be uncovered. There can be disagreement about when to stop disclosing due to different beliefs in what counts as a fundamental belief. When one party reasonably makes the claim that it has reached a fundamental belief, rounds end. If the disagreement about the fundamental belief explains the disagreements from the previous round, then both parties can hold steadfast to their beliefs from the previous round. A crucial part of this model is that agents can often foresee the rounds, especially in simple scenarios.\nOverall, the conformist attitude is required in scenarios of ideal peerage. In all rough cases (most, if not all, real world examples), the requirement of dynamic disclosure should be seen as an approximation towards ideal peerage, except in cases where rounds cease due to fundamental beliefs. Generally, this principle can be understood as \u0026ldquo;you shouldn\u0026rsquo;t use disagreement as evidence if you don\u0026rsquo;t understand the (origin of) the disagreement\u0026rdquo;.\nWe generally have a feel for how entangled a certain belief is. For instance, we know that in areas where disagreement is common, e.g. politics or philosophy, beliefs are very entangled, so much disclosing must be done. In contrast, in areas where disagreement is rarer, e.g. arithmetic, there rarely is much additional information gained from disclosing auxiliary beliefs.\nNow, let\u0026rsquo;s test this model against the cases.\nRestaurant with Mental Maths: Here, the evidence, i.e. the bill, is disclosed to begin with. What needs to be disclosed are the auxiliary beliefs, e.g. about arithmetic. If they are identical, as they probably should be, then it is unconceivable that at least one of the parties has not erred in their calculation. The agents who erred (which must become obvious to them after disclosure) must modify their belief. Because the case is symmetrical and agents can foresee that disclosure will locate the error, they ought to lower their confidence or suspend judgement. Conformism is applicable (for a different reason than in the literature so far).\nRestaurant with Calculator: Same as above, but auxiliary beliefs include how to use of calculator.\nHorse Race case: There are few auxiliary beliefs here (e.g. that we are both sober and that our eyes work). If the agent believes that sense perception is fundamental, then steadfastness is permissible. If the agents disagree about these auxiliaries (e.g. one thinks the other is on drugs), then the disagreement travels to that belief – conformism is applicable.\nDavid Lewis case: The case of real possible worlds is too deeply entangled to make it possible to approximate ideal peerage. Because I can\u0026rsquo;t even establish whether we have similar auxiliary beliefs, I cannot locate the actual source of our disagreement. This lets me stay steadfast in my beliefs.\nConclusion Hopefully, the dynamic model gave a convincing account on each of the four cases. Disagreements are often deeply entangled and to take a rational stance on them, one ought to disentangle them to the most fundamental level. Because agents can often reasonably foresee how likely it is whether a disentanglement of disagreement will reach a fundamental point, they can reasonably hold steadfast in so far as their fundamental belief entails the disagreed upon belief. This will be often the case in philosophy or politics and rarely the case in simple mathematics. If one does not explore the entanglement of the belief, then one cannot treat the disagreement as good evidence. For instance, in cases where it is impractical to continue disclosing, the disagreement itself is not good evidence, because it is poorly understood.\nWhat remains to be done is to try to generalise the model to more cases. The greatest challenge to this model is that its applicability is belief-laden: on the theoretical level it remains neutral as to its methodology (e.g. how to unravel entanglement or how auxiliary beliefs help the relevant belief), but on a practical level these questions may become central to applications.\nBibliography Bergmann, Michael. 2009. \u0026ldquo;Rational Disagreement after Full Disclosure.\u0026rdquo; Episteme: A Journal of Social Epistemology 6 (3):336–353.\nCevolani, Gustavo. 2014. \u0026ldquo;Truth Approximation, Belief Merging, and Peer Disagreement.\u0026rdquo; Synthese 191 (11). Springer:2383–2401.\nChristensen, David. 2009. \u0026ldquo;Disagreement as Evidence: The Epistemology of Controversy.\u0026rdquo; Philosophy Compass 4 (5). Wiley Online Library:756–67.\nDouven, I. 2010. \u0026ldquo;Simulating Peer Disagreements.\u0026rdquo; Studies in History and Philosophy of Science Part A 41 (2). https://doi.org/10.1016/S0039-3681(10)00025-7.\nElga, Adam. 2007. \u0026ldquo;Reflection and Disagreement.\u0026rdquo; Nous 41 (3):478–502. https://doi.org/10.1111/j.1468-0068.2007.00656.x.\nEnoch, David. 2010. \u0026ldquo;Not Just a Truthometer: Taking Oneself Seriously (but Not Too Seriously) in Cases of Peer Disagreement.\u0026rdquo; Mind 119 (476). Oxford University Press:953–97.\nFeldman, Richard. 2007. \u0026ldquo;Reasonable Religious Disagreements.\u0026rdquo; In Philosophers Without Gods, Meditations on Atheism and the Secular Life, 194–214.\nFoley, Richard. 2001. Intellectual Trust in Oneself and Others. Cambridge University Press. https://doi.org/10.1215/00318108-112-4-586.\nFrances, Bryan. 2010. \u0026ldquo;The Reflective Epistemic Renegade.\u0026rdquo; Philosophy and Phenomenological Research 81 (2):419–63. https://doi.org/10.1111/j.1933-1592.2010.00372.x.\nGoldman, Alvin, and Thomas Blanchard. 2016. \u0026ldquo;Social Epistemology.\u0026rdquo; The Stanford Encyclopedia of Philosophy.\nKelly, Thomas. 2005. \u0026ldquo;The Epistemic Significance of Disagreement.\u0026rdquo; Oxford Studies in Epistemology 1 (3):169–96. https://doi.org/10.1007/s13398-014-0173-7.2.\n———. 2010. \u0026ldquo;Peer Disagreement and Higher-Order Evidence.\u0026rdquo; In Disagreement. https://doi.org/10.1093/acprof:oso/9780199226078.003.0007.\nKing, Nathan L. 2012. \u0026ldquo;Disagreement: What\u0026rsquo;s the Problem? Or a Good Peer Is Hard to Find.\u0026rdquo; Philosophy and Phenomenological Research 85 (2). Wiley Online Library:249–72.\nMatheson, Jonathan. 2015. \u0026ldquo;Disagreement and Epistemic Peers.\u0026rdquo; Oxford Handbooks Online. 2015. http://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780199935314.001.0001/oxfordhb-9780199935314-e-13.\nMoffett, Marc. 2007. \u0026ldquo;Reasonable Disagreement and Rational Group Inquiry.\u0026rdquo; Episteme 4 (3):352–67. https://doi.org/10.3366/E1742360007000135.\nStanford, Kyle. 2013. \u0026ldquo;Underdetermination of Scientific Theory.\u0026rdquo; Stanford Encyclopedia of Philosophy, 1–17.\nSee Goldman and Blanchard 2016; Matheson 2015. Enoch (2010, p.996) for a reliability-based approach.\u0026#160;\u0026#x21a9;\u0026#xfe0e;\nArguably, there is only one case where evidence is identical, myself. It\u0026rsquo;s irrational to hold a belief in p∧¬p. If I encounter an irrational belief, I should see what went wrong.\u0026#160;\u0026#x21a9;\u0026#xfe0e;\nFor a counter-argument see Kelly (2010).\u0026#160;\u0026#x21a9;\u0026#xfe0e;\nFurthermore, there is a wealth of literature in a similar spirit that focuses on truth approximation and belief merging (Cevolani 2014).\u0026#160;\u0026#x21a9;\u0026#xfe0e;\nElga (2007) calls this entanglement \u0026ldquo;knots\u0026rdquo;. Enoch (2010) also argues against seeing disagreement as point-like and agents as \u0026ldquo;truthometers\u0026rdquo;.\u0026#160;\u0026#x21a9;\u0026#xfe0e;\n","permalink":"http://kkonrad.com/writing/philosophy/conformism-locating-disagreement/","summary":"Defends conformism via a dynamic disclosure model: agents should disentangle their web of entangled beliefs before treating peer disagreement as evidence, approximating rough peers toward ideal peerage.","title":"Conformism: Towards a Dynamic Model of Locating Disagreement"},{"content":"Dangerously low leverage ratios were one of the main causes of the 2008 financial crisis. The post-crisis regulatory response — Basel III and its EU implementation CRD IV — was supposed to fix this. It hasn\u0026rsquo;t.\nThis paper argues that most major UK and European banks remain perilously thinly capitalised, with leverage ratios averaging between 3% and 5% across systemically important institutions. Each pound or euro of equity supports between twenty and thirty-three pounds or euros of assets. Even very small falls in asset value could wipe out the equity of these banks — and trigger another round of taxpayer bailouts.\nKey findings:\nOfficial stress tests are flawed by design. Both the ECB and Bank of England stress tests rely on risk-weighted asset measures that banks have strong incentives to game, and on scenarios too mild to reveal true vulnerability. A stress test using actual asset value drops from the 2008 crisis shows that, without assistance, all major European SIFIs would see leverage ratios fall to 1.5–1.7% — well below the level at which Lehman Brothers failed. Germany and France, not the periphery, are most at risk. Contrary to the ECB\u0026rsquo;s narrative, French and German banks are the worst capitalised by the leverage ratio metric. The ECB\u0026rsquo;s preferred CET1/RWA metric flatters core Eurozone banks relative to peripheral ones. Implicit subsidies distort incentives. The too-big-to-fail guarantee, tax deductibility of debt interest, and deposit insurance together create a system where profits are private but losses are socialised. Without these subsidies, there is no evidence of economies of scale at bank sizes above $100bn. The fix is simple but resisted. Academic consensus suggests leverage ratio requirements of 15–30% — far above the Basel III minimum of 3%. Banks resist this because it would transfer risk back to shareholders from taxpayers. The EU\u0026rsquo;s regulatory structure makes it difficult for member states to impose stricter requirements unilaterally. Written under the supervision of Professor Kevin Dowd (Durham University) and published by the UKIP Parliamentary Resource Unit.\nRead the full paper (PDF)\n","permalink":"http://kkonrad.com/writing/policy/european-banks-not-fixed-yet/","summary":"A stress test of European systemically important banks shows they remain perilously undercapitalised — and that regulators and official stress tests obscure this.","title":"European Banks: Not Fixed Yet"},{"content":"Blues I co-founded:\nOpen Blues — a yearly festival that attracts guests from 5 continents. 100% DIY and volunteer-run. Honey Honey Blues (sunset) — weekly blues classes in Warsaw, Poland. I usually teach with Kate Phellini. NoName Blues (sunset) — weekly blues parties in Warsaw, Poland. Film I worked as the 2nd director for Oko i Ucho on a peak-time TV documentary film series with a viewership of around 0.5M. The series is about the influence of politics on architecture and urban landscapes. Elegy for Dunelm House — Producer \u0026amp; Co-Director. Culture Y Award for Best Film at St Andrews University. Radowid Silk — Producer. Journalism I worked as a fixer for Ángela Rodicio and her team during the first days of the Russian invasion of Ukraine. I worked as a photographer for Süddeutsche Zeitung. ","permalink":"http://kkonrad.com/misc/","summary":"\u003ch2 id=\"blues\"\u003eBlues\u003c/h2\u003e\n\u003cp\u003eI co-founded:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e\u003ca href=\"http://openblues.pl\"\u003eOpen Blues\u003c/a\u003e — a yearly festival that attracts guests from 5 continents. 100% DIY and volunteer-run.\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://honeyhoneyblues.com/\"\u003eHoney Honey Blues\u003c/a\u003e (sunset) — weekly blues classes in Warsaw, Poland. I usually teach with \u003ca href=\"https://katephellini.com/about\"\u003eKate Phellini\u003c/a\u003e.\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://www.facebook.com/nnblues/\"\u003eNoName Blues\u003c/a\u003e (sunset) — weekly blues parties in Warsaw, Poland.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch2 id=\"film\"\u003eFilm\u003c/h2\u003e\n\u003cul\u003e\n\u003cli\u003eI worked as the 2nd director for \u003ca href=\"http://www.okoiucho.pl/\"\u003eOko i Ucho\u003c/a\u003e on a peak-time \u003ca href=\"https://niepodlegla.gov.pl/aktualnosci/rzeczpospolita-nieodbudowana-serial-o-ii-rp-ktorej-juz-nie-ma/\"\u003eTV documentary film series\u003c/a\u003e with a viewership of around 0.5M. The series is about the influence of politics on architecture and urban landscapes.\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://vimeo.com/262599625\"\u003eElegy for Dunelm House\u003c/a\u003e — Producer \u0026amp; Co-Director. Culture Y Award for Best Film at St Andrews University.\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://vimeo.com/manage/videos/254476400\"\u003eRadowid Silk\u003c/a\u003e — Producer.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch2 id=\"journalism\"\u003eJournalism\u003c/h2\u003e\n\u003cul\u003e\n\u003cli\u003eI worked as a fixer for \u003ca href=\"https://en.wikipedia.org/wiki/%C3%81ngela_Rodicio\"\u003eÁngela Rodicio\u003c/a\u003e and her team during the first days of the Russian invasion of Ukraine.\u003c/li\u003e\n\u003cli\u003eI worked as a photographer for \u003cem\u003eSüddeutsche Zeitung\u003c/em\u003e.\u003c/li\u003e\n\u003c/ul\u003e","title":"Misc"},{"content":"Past EthDenver — p2p electronic cash or just new middlemen? DevConnect 2025 — Eth/Acc — Peanut Pitch DevConnect 2025 — Will We Ever Have True P2P Digital Cash? — Onchain Banking \u0026amp; Fintech When Help Hurts: Why a Friendly Government is a Threat to Crypto — Protocol Berlin Wen p2p Electronic Cash System? — Devcon 7 SEA Chain abstraction? But I want to know where my tokens are — Devcon 7 SEA How can wallets deal with the multichain nightmare? — DevCon Wallet Uncon Why wallets suck for payments — DevCon Wallet Uncon Will we ever have a p2p digital cash system? — EthCC Payments Cafe The Next Wave of web3 Applications (panel) — EthCC Infra Breakfast Please stop obsessing about blockchain! — EthDenver 2023, The Next Billion Summit Ambire Mini-Summit \u0026ldquo;The Next Billion\u0026rdquo; Vision (panel) — EthDenver 2023 The evolution of Layer-2 scaling solutions (panel) — CryptoSummit, Zurich What can we learn from CEXes? — Pretzel DAO, Munich On the Future of DeFi — Full Node, Berlin No one should have to care about blockchain — Lift99 DAO, Berlin Permanence and storing memories — IPFS Camp, Lisbon Community Building — Let\u0026rsquo;s Phi A Critique of Proceduralism — Advanced Research Computing, Durham University Free Education — Durham University Union Society On MVPs — University of St Andrews IP is not property — Polish Philosophical Congress Intellectual \u0026ldquo;property\u0026rdquo; — European Conference for Analytic Philosophy Global and Local Exclusivity in Intellectual Property — GAP9 Osnabrück Podcasts Lift99 DAO Human Story Stable Stories Web3 on Fire Krypto Rozmowy Goodnews ","permalink":"http://kkonrad.com/talks/","summary":"\u003ch2 id=\"past\"\u003ePast\u003c/h2\u003e\n\u003cul\u003e\n\u003cli\u003eEthDenver — \u003cem\u003ep2p electronic cash or just new middlemen?\u003c/em\u003e\u003c/li\u003e\n\u003cli\u003eDevConnect 2025 — Eth/Acc — \u003ca href=\"https://www.youtube.com/watch?v=OkKtOzZkR-o\"\u003ePeanut Pitch\u003c/a\u003e\u003c/li\u003e\n\u003cli\u003eDevConnect 2025 — \u003cem\u003eWill We Ever Have True P2P Digital Cash?\u003c/em\u003e — Onchain Banking \u0026amp; Fintech\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://watch.protocol.berlin/65a90bf47932ebe436ba9351/watch?session=6851404d4ac43bf73d22ddb8\"\u003eWhen Help Hurts: Why a Friendly Government is a Threat to Crypto\u003c/a\u003e — Protocol Berlin\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://app.devcon.org/schedule/ZFX3ZF\"\u003eWen p2p Electronic Cash System?\u003c/a\u003e — Devcon 7 SEA\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://www.youtube.com/watch?v=J_VTceUdfgs\"\u003eChain abstraction? But I want to know where my tokens are\u003c/a\u003e — Devcon 7 SEA\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://cfp.walletuncon.org/walletuncon-2023/talk/9QGLLM/\"\u003eHow can wallets deal with the multichain nightmare?\u003c/a\u003e — DevCon Wallet Uncon\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eWhy wallets suck for payments\u003c/em\u003e — DevCon Wallet Uncon\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eWill we ever have a p2p digital cash system?\u003c/em\u003e — EthCC Payments Cafe\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eThe Next Wave of web3 Applications\u003c/em\u003e (panel) — EthCC Infra Breakfast\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://youtu.be/q4NsLmTGNAU?t=9292\"\u003ePlease stop obsessing about blockchain!\u003c/a\u003e — EthDenver 2023, The Next Billion Summit\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://www.youtube.com/watch?v=FFcFb6Arv3M\"\u003eAmbire Mini-Summit \u0026ldquo;The Next Billion\u0026rdquo; Vision\u003c/a\u003e (panel) — EthDenver 2023\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eThe evolution of Layer-2 scaling solutions\u003c/em\u003e (panel) — CryptoSummit, Zurich\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eWhat can we learn from CEXes?\u003c/em\u003e — Pretzel DAO, Munich\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eOn the Future of DeFi\u003c/em\u003e — Full Node, Berlin\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eNo one should have to care about blockchain\u003c/em\u003e — Lift99 DAO, Berlin\u003c/li\u003e\n\u003cli\u003e\u003cem\u003ePermanence and storing memories\u003c/em\u003e — IPFS Camp, Lisbon\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eCommunity Building\u003c/em\u003e — Let\u0026rsquo;s Phi\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eA Critique of Proceduralism\u003c/em\u003e — Advanced Research Computing, Durham University\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eFree Education\u003c/em\u003e — Durham University Union Society\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eOn MVPs\u003c/em\u003e — University of St Andrews\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eIP is not property\u003c/em\u003e — Polish Philosophical Congress\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eIntellectual \u0026ldquo;property\u0026rdquo;\u003c/em\u003e — European Conference for Analytic Philosophy\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eGlobal and Local Exclusivity in Intellectual Property\u003c/em\u003e — GAP9 Osnabrück\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch2 id=\"podcasts\"\u003ePodcasts\u003c/h2\u003e\n\u003cul\u003e\n\u003cli\u003e\u003ca href=\"https://www.youtube.com/watch?v=VHslneMYZUw\u0026amp;t=54s\"\u003eLift99 DAO\u003c/a\u003e\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://www.deezer.com/en/show/3550827\"\u003eHuman Story\u003c/a\u003e\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://open.spotify.com/episode/4KsIm8eFnrzu1Da8CvVqLD\"\u003eStable Stories\u003c/a\u003e\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://open.spotify.com/episode/7uyZ58gkzQdVeT7jgO61me\"\u003eWeb3 on Fire\u003c/a\u003e\u003c/li\u003e\n\u003cli\u003eKrypto Rozmowy\u003c/li\u003e\n\u003cli\u003eGoodnews\u003c/li\u003e\n\u003c/ul\u003e","title":"Talks"}]