Click the box above for the PDF of the below manifesto.



28 January 2026




The system of political parties, and of standing for election, has failed. "The meek shall inherit the earth", and there's nothing particularly humble about campaigning for power. The system is an open door for people who are delusional and up to no good. A different system is needed — one where the commonality of the most powerful people is not that they sought power or were chosen by it. Politics shouldn't be led by politics clubs, and even the highest power doesn't have to be anywhere near a full-time occupation. It should be possible that a random soldier, because of the political selection process, finds himself as the world's most powerful man, therefore both above and below his commander, for example. The following process can enable that. A number of people (e.g. 1% of the population) are randomly selected and put into proximity-based groups (e.g. of ten people), and those groups vote for their favourite person in the group, and the favourite people get put into proximity-based groups, and the process continues until there is a final favourite person. In each round, the number of votes someone gets is multiplied by the round number, determining how many points they gain. Once the election is finished, points equal political power. More on this system later. First, I'll dive into criticising the education system and describing a new one.




In South Korea, the country with the world's lowest birth rate and most demanding education culture, suicide has been the leading cause of death in 10–24-year-olds since 2011.


“It takes a village to raise a child” (African proverb), yet schools function comparably to prisons.


Those sub-replacement and declining birth rates aren't going anywhere without meaningful change — which means social system replacement — reinvention, reimagining. It's time to get out of the death spiral. I could list off a bunch of other worsening issues but you probably don't need to be told what you can experience yourself.


Einstein — someone who has done far more good for the world than any power-seeking authoritarian of the day — thought in 1931 that “individuality” is “destroyed” by the treatment of the child as a “dead tool”, "bee" or "ant". He said that ”a community of standardised individuals without personal originality and personal aims would be a poor community without possibilities for development”. He also said “the worst thing seems to be for a school to principally work with methods of fear, force and artificial authority” as it "destroys the sound sentiments, the sincerity, and the self confidence of the pupil” and “produces the submissive subject”.


The political clubs have been unwilling to make an education system fit for geniuses, so it's no surprise that dumber people have more kids. There's a vicious cycle going on, where survivorship bias keeps the system going despite how ridiculous it really is.


The purpose of having a qualification is to prove one's ability to a potential employer or for an educational opportunity. Ability is not limited to exam results, so individualised certification should be possible. A six-year-old should also be allowed to sit the same exam as a sixteen-year-old, for example, given that education generally doesn't need to be age-restricted, but should build on existing ability and knowledge. Appropriate education cannot be offered by a school that hasn't checked the existing abilities and knowledge of its students.


To make the fundamental insanity of the system clear to any privileged doubters, here's an experience of mine:

It's 2006 and I've started primary school. Probably on my first day, I'm given a colouring-in task, which I take as insulting to my intelligence. I'm waiting to be asked what I want to learn. I want to learn about computers and programming. I'm not asked what I want to do. 

To make the colouring-in a challenge, I do it perfectly, not going over the lines as people around me are. I rotate the page to make it easier for me to colour perfectly. The teacher shouts at me for that.

Clearly, then, the education system is not about "life chances". It's about being a cog in someone else's self-serving machine. Only in a system centered on obedience is this possible. Politicians must be very two-dimensional characters if they seriously think it's maximising life chances. Ah, yes, maximising chances by minimising choices 🤣 Coercive monopolies don't work for the people under them.


Schools should be sure to teach what's actually important or extremely useful to teach all children. Algebra is important, sure, and English literature can be interesting and educational, but it's more important that all children know how roads and intersections work before they start cycling, which is another generally more important skill than advanced maths or artistry or playing the flute. Knowing about health issues and risks, and what the symbols on product packaging mean, is more important than knowing about knights and castles. In the UK, under the "Labour" party, schools have art, drama and music classes, but not building, driving or street cleansing ones. It is standard that political parties' names hold no weight. The "Liberal Democrats" "fight" for a “fair, free and open society” where nobody is “enslaved by conformity”, yet they believe in the same education system as the "Conservatives". Politicians support the existing system because they either can't or aren't willing to think outside of the box that already exists — despite all the suffering in our world, which politicians treat as normal. The world would be a much healthier and happier place if politicians spent five hours a week thinking and researching towards making life better for their people, instead of the five seconds a month it seems they currently do. I have been fooled by the political parties, and I am fooled no longer. The system has defeated many people. Luckily, it hasn't completely defeated me, otherwise I'd be dead or in jail or thinking there's something wrong with me, instead of thinking there's something wrong with the system. I definitely have reason to regret not raging as soon as school spent other people's stolen money to mess me about, instead of letting the system chew me up and spit me out like it did, wrecking me physically, socially, educationally and economically. If we can move into the future with my much more intelligent system, I won't regret. The politicians love to use violent metaphors, so, I say, a steamrolling is in order. The answer to the question of why humans put up with being treated terribly, instead of kicking off at every little bullshit, is probably summed up by this quote from Alexander the Great: "I am not afraid of an army of lions led by a sheep; I am afraid of an army of sheep led by a lion".


The 20th century brought secondary education expansion (1950s–1960s in many countries) and mass higher education (post-WWII, accelerating in the 1960s-1970s).


Given that school does undeniably make children "submissive subject"s, as Albert Einstein said, it should come as no surprise that, in the so-called developed world, sperm counts and testosterone levels in young men have fallen roughly 1% per year for decades. Western men had, on average, 113 million sperm per millilitre of semen in 1940, 99 million in 1973, 66 million in 1990 and 47 million in 2011. If the sperm decline trend were to continue, the average sperm count would reach zero by 2045. In other words, big political/social change is coming by 2045. In 2045, many old ways and old groups will be history. It will be a very different world. Hopefully, not the very difficult world the establishment is leading us to.


From the Netflix documentary "Inside the Mind of a Cat":


"Not every cat can do special tricks. You must observe the cat to understand what she is capable of, and teach them things based on their strengths."


"When animals are born, they're born with a temperament. That temperament is the initial building blocks for that personality that's going to develop over their lifetime."


"All of our cats have different personalities and you need to adjust to them, so, you cannot make them adjust to you — it's kind of not their job, it's yours. So, you need to pick what works for them better."


"It's like with humans. Some are great at computers. Some can be great ballerinas. You wouldn't teach computers to a ballerina."


"The Savitskys don't force their performers to do these amazing stunts. Like a sculptor who finds the sculpture within the rock, they let each cat's potential reveal itself."


Humans being educated for their own lives should be afforded this level of humanity with which cats are trained for humans' entertainment. The trainers love their cats. Schools should love their humans. Instead, they are so distant from each child that their suffering and wasted potential are invisible. Schools are just focused on obeying the orders of the out-of-touch elite, when they should be organising around the complicated reality that is 7 million years of human evolution.


Those who justify the education system are the 21st-century equivalent of those who justified slavery by saying that black people are suited to it. In recent decades, they mostly stopped hitting children in schools, making schooling less visibly comparable to slavery. Education obviously isn't a "right" if it's compulsory. That's ridiculous.


In justifying the system, they use correlations (e.g., school attendance is linked to earning more money, and to greater executive functioning after a year of school), even though lots of other data can easily discredit it. "We have to remember that what we observe is not nature herself, but nature exposed to our method of questioning." — Werner Heisenberg


Since what is truly desired is not schooling for its own sake, but rather the outcomes associated with it, the school system itself should not be based on maximising "attainment" within the system (or obedience to a programme), but rather should be directly concerned with maximising these desirable outcomes. A hierarchy of desirables — starting with health — will go a long way in creating an education plan that is maximally based in reality. All subjective and objective life experiences should be considered associated with one's school experience (or lack thereof), making the system maximally accountable, adaptable and optimisable.


Nevermind attendance, and academic grades are only a piece of the puzzle. Schools (their management) should instead be judged by many long-term and short-term metrics, including the reported happiness and successes of former students, and the dental health and memorisation abilities of current students, to name a few. There should also be a feedback mechanism, so the feedback of children and parents truly shapes the way state education is done. I'll get to that.


I said "students", but that's not quite right, given I said "nevermind attendance". Regardless of attendance, each child should have a school that is their "responsible school", which is judged on data regarding that child. The only thing that should be compulsory for the child is a yearly welfare check from the education regulator, taking place in their responsible school. The welfare check should be the start of a yearly data collection process by the regulator, in which children are offered payment to participate in health checks and a demonstration of their knowledge and skills: the child chooses from a list of available tests to take, and can also demonstrate their abilities open-endedly.


The main purpose of the welfare check is to have all children attend their schools at least once per year, ensuring that they are aware of what their schools offer, feel free to benefit from them and understand their value. As part of the check, it should be compulsory that a child who has spent less than five hours in their school in the past year, spends the minimum five hours.


In order for educational practices to evolve according to feedback and metrics, at least some schools will need to enact unproven strategies. Consider the current educational practice as proven to be destructive. It will do no more harm to enact unproven strategies, especially given that schools will be regulated according to routine feedback and metrics that matter. Each school should have a strategy document that it follows and can only change with the regulator's agreement. Schools (the head teacher, primarily) should develop their strategy documents to suit their premises and other school-specific realities. I explain strategy documents later.


Since it will certainly be beneficial, schools should be open 24/7, so children's potential studying/sporting/etc time is not limited, and so they always have a place to go. Of course, not every part of a school needs to be always open — there are diminishing returns — but it's not as though school facilities are only useful for a portion of the 24-hour day. There should be beds for children who find it convenient or preferable to sleep at school. Schools should host community dinners, and have parts for independent study, for tutoring, and for learning in groups of everything from a few to a few hundred. A booking system should be used and applied to activities as suited, so, for example, if a child wants a seat in an independent study computer room, they need authorisation. Or, if they want to attend a class, they may need to take a test, so the management is sure they're suitable for it. Children and parents should certainly be able to message and book meetings (if not randomly show up) to talk with the management. The booking system should be available for staff, children and parents as an app and website. Children, parents and staff should be able to propose bookings to the management. For example, a group of children might see that the sports hall is not booked for a certain time, and propose a group booking to use it for a certain activity. The system should show what is planned so that people can join in. Each school should also have an official Discord server (for everyone involved, including parents) — Discord being a versatile and broadly used communication platform that has even been used by Russians coordinating their invasion of Ukraine, so you can't say it's unsuitable.


With this school system, different children can live highly varied lifestyles and truly be themselves. Adults are allowed to live their own way, and are expected to. Anyone who expects children to live in one way is badly mistaken, yet this is what politicians have pretty much done to humanity. 


A kid might basically eat, sleep and spend all their time on school premises, and their parents might go to school to see their child, more than their child goes home. Another kid might never attend school, besides the yearly checks, and they will thrive just fine without the intrusion of the state, which would be a waste of taxpayer money, to say the least.


Schools should avoid finding themselves burdening the taxpayer with costs for raising children that nobody — not even the offspring themselves when they are adults — is willing to cover.

Therefore, reasonable responsibility has to be put on parents, with responsibility increasing for each additional child.

Schools should seek to fund things through donations, which could both limit the need for taxation and extend what they can offer.


Headteachers should be on a social network where they can communicate their ideas and experiences with other head teachers as well as the regulator. If a school implements a new idea which is found to lead to better data (feedback and metrics) across schools, then the pioneers should be financially rewarded, including the staff who first enacted the policy. 


The education system needs both innovation and stability. Freedom (of policy) is needed for innovation, and restriction is needed for stability. The restriction is needed in order to lock in successful policies once they have been discovered through experimentation. A history of unsuccessful ideas should be kept, so that mistakes aren't repeated. Everything that a school does should be recorded, so that actions can be linked to data. This means that all school employees need to keep a journal of what they do as part of their job.


Regarding the feedback mechanism: there should be a website (of the regulator) where someone can log in and give feedback (in the form of ratings and comments) regarding a school they have attended (but attendance is not quite the right word). Parents should also be able to give feedback on any schools their children have attended. Each child will have one default school, which they can give feedback on, which is their responsible school, but they may interact with other schools, particularly during multi-school events. Each time a child interacts with a school, it gets added to the list of schools they can give feedback on. The regulator should, every Sunday, prompt everybody (in the form of an app notification or an email) who has a responsible school to give feedback on it, including parents, though feedback may be given anytime. It should be possible to give feedback on the school as a whole and also on individual staff members, including management. Pay should be influenced by feedback. A staff member with an average rating of 10 should earn twice as much as one with an average rating of 5. Whole-school ratings should primarily influence the management's pay, and only somewhat influence staff pay. Feedback should be timeframe-based, meaning that you can give an all-time review or feedback that applies to the past year, month, week or day. It should be possible to give a rating without a comment, a comment without a rating, and specifically a suggestion. You could, for example, send feedback which is about a particular staff member, about the past week, including a rating, including a comment, and check a box to indicate that your feedback includes a suggestion. You should also be able to indicate who you want to be able to see your feedback — for example, only the regulator, or everyone, including the general public, or somewhere in between. Once someone has experienced a school or otherwise interacted with the activities of its staff, they should forever be able to provide feedback on their experience, and should be periodically prompted (e.g. every five years, if they are of middle age) to do so. The feedback service should use facial recognition checks to reduce the potential for non-genuine feedback.


A child should have a responsible school from the moment they are conceived, meaning that the school interacts with the mother, and that data on the new human counts towards the success (propagation) or failure (disuse) of the school's strategy document. In reality, it will be parts of documents that will evolve, like genetic strains. Which school is a child's "responsible" school is likely to change as they age. Data can be turned into points or percentages for categories and subcategories. For example, one school might have a "strength" score of 150 while another has a "strength" score of 100, because the "responsibles" of the former are physically stronger. Schools and strategies aren't the same thing, though, and it must be figured out what data are more closely related to individual staff and other environmental variations, and what data are more closely related to strategy documents. A strategy document will also have a strength score that can change even though it isn't being used anymore, because strength data will continue to be collected from those who experienced the strategy. Of course, this is made complicated if a school's strategy document is frequently changing. Again, in reality, data would be found to be related to strategies and sets of strategies. The document is one of "strategies", which are just "things" of the school, which I'll explain.


Schools should cooperate to provide a wider range of facilities and opportunities. What I mean by this is, for example, a child's responsible school is the closest to their house, but since a few other schools are also very accessible, they can help each other. For example, school A accepts school B's responsibles to use a climbing wall, and in return, school B accepts school A's responsibles to use their trampoline park. This gives each school's responsibles access to both a local climbing wall and trampoline park, whereas they may have otherwise had access to only one or neither of these. It's a win-win. 


A strategy document will include currently implemented things, works-in-progress, a realistic five-year improvement plan, and overbudget ambitions/ideals. Said "things" are 1. physical things and 2. practices. Physical things should be divided into "fixed" and "movable" things. For example, carpeting, building layout and material and the tennis court go in the "fixed" section, while indoor plants, books, tables, screens etc. go under "movable". If one or two people can simply carry or push something to another part of the school, it should go in the "movable" section, including location strategy. Practices should be divided into "internal" and "external". "Internal" practices would be the management's hiring practices, routine meetings, who the management and staff are and how they interact and what chain of command there should be, maintenance etc. While each school should have one head teacher, they will certainly appoint management. There needs to be at least three people that the regulator knows are the management. What powers the management has depends on their strategy document. The management should generally not work from home, but should be physically present in their school, whereas the regulator can more so work from anywhere, and randomly visits schools to check on them, including asking questions, doing fire safety inspections etc., taking notes and comparing the information gathered to the school's strategy document. "External practices" are the equivalent of "customer-facing" activities. For example, handwriting lessons, farm visits, how often children should be messaged, what parts of the school are for what and how each part of the school is managed in terms of their rules and purpose. A new staff member would have to read the strategy document (though this depends on their role) in order to understand how the school works, what goes where and what to do etc. Online documents have a suggestion feature — the school management can use this to propose edits to their document, pending approval from the regulator. The regulator should be full of elected officials, or at least well overseen by them. The sections and subsections of a strategy document should be as follows. The four top-level sections are "currently implemented", "works-in-progress", "realistic five-year improvement plan" and "overbudget ambitions/ideals". Nested within each of these sections is "physical things" and "practices". Nested within each "physical things" subsection are "fixed" and "movable" subsections. Nested within each "practices" subsection are "internal" and "external" subsections. Furthermore, each bottom-level subsection can have an "individual-specific" subsection for comments and practices relating to individual people. For example, in "realistic five-year improvement plan", in "practices", in "external", in "individual-specific" can be a plan regarding a specific child. While the strategy document should list currently implemented movables, it shouldn't be as detailed and specific as an inventory. Each school should keep an inventory of its property, in which individual products and items, and their condition, are recorded and updated. This means that the strategy document wouldn't be updated with every minor change to the items in the school. 


When a new school is built, the regulator should at least cobble together a strategy document for it, made using successful strategies. The school itself should be built according to connections found between existing schools' fixed physical things and their data (feedback and metrics). For example, it might be found that yellow paint is better for maths classrooms. Countless correlations can be found, and different successful strategies may emerge, allowing for different types of schools, much like how evolution results in different adaptations depending on the environment. A school may require a strategy document change if a generally successful strategy isn't successful as expected for that school. The regulator may have some schools update their strategy documents quickly, according to what appears to be the latest great innovation, while having other schools play the long game, waiting for more long-term data, ensuring that strategies are adequately explored. 


When feedback is sought from adults, regarding the schools they had interactions with in their childhood, that should be accompanied by seeking metrics. "Metrics", like literacy, and "feedback", like review scores, are "data", in this context. Schools are "responsible" for certain children, but it's not like a school should ban people once they reach eighteen, especially if that would not only backfire regarding data from them, but also not even be related to better data from the children of the school.


Do not doubt. If the financial incentive of the feedback system were to guide staff behaviour in a way that undermines the purpose of other systems, leading to worse meaningful metrics, then financial incentives can be more significantly applied to metrics. It is certain that all children will learn basic literacy, numeracy and beyond, in this system. Even if most children are idiotic short-term thinkers who wouldn't even learn basic skills on offer by schools, parents can easily teach them anyway, and schools can easily incentivise children to learn. Even under the assumption that children only want to play, they will learn basic skills through play anyway — that is, when they are truly enjoying themselves due to not being downtrodden by the very system that is supposedly for raising them up. Short-term pleasure-seeking does not exist irrespective of social conditions. An experiment found that unhappy rats in a cage, alone with no fun, end up drinking cocaine water available to them, while rats in a "rat park" do not. Cracking down on children's social media use, instead of raising them up properly, is a current example of using tyranny to solve problems created by tyranny. I only had to watch a YouTube playlist on European history to learn that this has happened a lot in the past. The oppression of the education system in fact shoots itself in the foot, leading to children who need more learning support and are more dependent on the state.


Financial incentives are something for the regulator to play with. How staff and management pay are affected by data, and how much financial and other incentives may be used for data collection, and how often data is sought, can be changed to see what works better. So, that's another layer of strategizing. That is because, for example, seeking data from each child more often than yearly may be worth more than it costs, or spreading out the data collection might be more effective than proposing all the testing at once. Data collected from the regulator can be turned straight into data available for employers and other educational opportunities.


The practices described don't have to stop at schools. They can be used for community centres, generally. Ideally, everyone will have one or more community centres that cater to them, meaning that, for example, someone who turns sixteen or eighteen and finds themselves with limited access to the school they attended, will, at the same time, have expanded access to a different place and other opportunities. There will be natural separation of age groups and of the sexes, and, to a lesser degree, other demographics, but it should be natural, according to what works, which is what produces better data. So, instead of just having mixed-sex schools and single-sex schools just because people thought they're good ideas, you'll have mixed-sex and single-sex stuff based on what actually produces the most bang for the buck. So, no more ideological laziness. Power actually has to serve the people. To serve both the individual and the community is not necessarily easy, but that's what must be done.


On the one hand, staff and management should be financially rewarded for positive data (regarding their responsibles), and on the other hand, everything is relative, which means that rewarding positive data could backfire by encouraging keeping strategies secret that lead to better data. "Positive" data is really just "better" data, and it's easy to see that someone might try to remain "better" than others for the money. It shouldn't be that keeping an effective strategy secret is more financially appealing than properly documenting it.


It may be that most of the evolution of strategies will come from staff taking the initiative due to reward-seeking, without knowing if their strategy will work, and then, due to their journalling, their unplanned strategies can then find themselves in strategy documents. It may be that, rather than the management's strategy document informing staff actions, it is staff actions, as written in their journals, that populate the strategy document. In reality, it will be a mix.

Staff autonomy, or lack thereof, is something for the "internal practices" part of the document to explain. Increased staff autonomy does not mean a decrease in strategizing, though. It means that strategizing is less centralised. Low staff autonomy would mean the management has a lot of work to do. High staff autonomy would also mean the management has a lot of work to do, because they'll have to be vetting and figuring out what the staff are up to. 

Although there should be an official head of a school, and official management in the eyes of the regulator, the actual practiced management structure of a school will be more complicated and natural than that, and this is what the "internal practices" part of a strategy document is about.


A school's strategy document should be, within reason, readable by those involved (parents, kids, etc.), so that they can verify its truthfulness, gain insight, make suggestions and offer to contribute to plans. The main or only thing requiring privacy by default would be things in the "individual-specific" subsections.

Like how the strat doc should not list items with the detail of an inventory, it should not contain detailed lesson or event plans etc. These should be linked to, and can be kept private as suited.

I think that the document suggesting feature would not be suited to children and parents suggesting edits, given the volume. 

There should be a suggestion channel as follows: you first select if what you want is a removal, addition or a mix. Then, you select the type, going along the strategy document tree, starting from "currently implemented", "works-in-progress", "realistic five-year improvement plan" and "overbudget ambitions/ideals". This is optional and so that your suggestion can be categorised for efficiency. For example, you might read the strategy document and want to edit a five-year improvement plan regarding a fixed physical thing. Or, you might want to suggest an addition to currently implemented movable things. You should be able to provide an estimated cost for the implementation of your suggestion, and can select if you are willing to pay for it yourself, or pool money towards it, or provide it yourself. You, as a parent or child or anyone, should be able to look through non-private suggestions and offer to pay for or pool money towards them, and/or express approval or disapproval (like/dislike). The same should go for the strategy document itself. So, you can interact with suggestions to the strategy document and you can interact with the actual document: liking, disliking, commenting and contribution options. When reading a school's strategy document/suggestions in "interaction mode", the text/media should be in auto-generated interaction sections ready for you to interact with, and you should also be able to freely highlight parts to apply interaction to. The interaction types are in three broad categories: approval, communication, and contribution. Approval: liking/disliking. Communication: commenting and discussing in the Reddit/Facebook format, which can lead on to the Discord format. Contribution: offering to either do it yourself or pay for it yourself, or do it as part of a group or pay for it as part of a group. In interaction mode, you should, of course, be able to see others' interactions.


This strat doc and suggestion interaction system means that, for example, one parent might suggest doing something for the school, and then people like it, and then someone suggests to pay for it, and then the management/regulator approves it, and then, ta-da, you have an actual school community.

An example of a practice which would probably count as "internal" is cleaning the toilets. The exact procedure for cleaning the toilets would be linked to from the strategy document. In Japan, the kids clean the school themselves, I heard. With this document system, children can read and understand the functions of the school and a kid could use the interaction system to offer to do that practice, for example.

Another example: A company might see that the management plans to resurface something, and offer to do it cheaply for brand recognition.

The benefit of a school's document and suggestions being broadly visible to the public is that people from far and wide could offer big contributions. Many small contributions will come from the school community itself, but occasional big contributions can come from farther away, for example, from rich areas to poor areas. It's much nicer to donate to something specific that you like than it is to pay tax to a soulless government for many things that you do like as well as many things that you don't. The farther away people are, the less they're relevant to a school, unless, of course, they're offering something rare. So, a channel needs to be open for outsider contribution.




On the aforementioned political system: 


Someone's points after an election will be their "voting power" until the next election, which is the amount of weight they can throw behind a political proposal. Favourites will be able to pool their points for and against proposals. With enough support and not enough disapproval, a proposal can be implemented.


Regarding the election process: The groups are facilitated to communicate and meet up before voting. Group members participating in communication are all expected to vote, so they are disqualified if they decide not to. The randomly selected people are automatically put into groups, which means someone could be voted into power, in their absence, without even knowing they had been randomly selected. By default, information on them is given to their group members, so that they can make informed decisions to vote for them. Regarding group size: I imagine 10 people per group as the default. A group size of, for example, 100, would require more meetup time. Whereas a 10-person group could do with a 15-minute meet, a large group would spend hours together, likely eating together and going through a sequence of social activities.


Elections will be a process of finding, connecting and filtering people. The more people that are randomly selected, the better, because the bigger the election process will be, relative to the population size, and therefore, the more that useful ideas and people will be drawn into power. No segment of the population (e.g. children, prisoners) should be excluded.


Given that the voting groups are proximity-based, the election process starts local and then zooms out and out, connecting people from farther away. Someone may, for example, have "their" street favourite, neighbourhood favourite, town favourite, regional favourite, etc. Local governments can be made out of this, and there can be a hierarchy of governments. A favourite will primarily concern themselves with the area that voted for them.


Using their political responsibility, officials may delegate a lot. It is up to them how they want to approach their authority. They could perhaps remain anonymous and make big decisions without even their family knowing, and take a laid-back approach, only chiming in occasionally. Or, they could do the opposite, and gain a following, and spend a lot of time meeting people and involving themselves with things.


I describe two extra practices to maximise the likelihood of great governance:


Trust score: There will be a website up (but not during elections) where, everyone, by default, has a trust score of zero, and you can search someone's name, and you can't see their trust score, nor your own, but you can influence their trust score using your 100 influence points. You could for example, influence person A's trust score by -50, and person B's trust score by +50, and then you have used all your influence points. The only time you can see someone's trust score is if you have the opportunity to vote for them. So, in addition to being facilitated to meet the people in your vote group and seeing official info on them, you can also see their trust score. When you log in to the trust score website, your previous allocation is not shown. You can simply make a new allocation. Therefore, even if someone got your password and logged in to your account, at least they cannot see if you trust them or not, or who you trust/distrust. You could, for example, allocate 40 trust points to person A, 20 points to person B and -40 points to person C. How secret trust scores should be is something to consider.


Governance penalty/reward: Before there is a new election, the public is given the opportunity to vote on if they think the government made things better or worse. If they think the government made things better, then those in government (those with the most voting power) should be financially rewarded in accordance with their voting power and the degree to which the public thinks they made things better. If the public thinks the government made things worse, then they should be punished by time in prison — with those with the most voting power spending the longest time in prison — up to a year — or longer if the public thinks they made things a lot worse. This will deter conning and incentivise actual work.

This major public vote will be multiple choice, as follows:

• I think they made things a lot better

• I think they made things better

• I'm not sure / I think things aren't better or worse

• I think they made things worse

• I think they made things a lot worse




The fixing of the political and educational culture and systems would indirectly fix the military culture and systems. With that said, I have my criticisms and my vision. Starting with criticism:


"Service" and "duty" are suspicious word choices, surely intended to prop up morale. I believe that, like the education culture, the military culture is overly concerned with looking good and producing submission. It's not concerned enough with soldiers being of good health, strength and character. Taxpayer money is wasted on fancy uniforms and medals, singing and dancing, and pomp and pageantry, with surely no regard for whether that's what soldiers even want. I believe that, for all that elaborate dress codes and other practices of arguable utility may "maintain discipline", they equally result in the abuse being passed on, resulting in war crimes, far from disciplined. To coordinate men of high morale is preferable to disciplining men treated as untrustworthy. There's organising, and then there's domineering. Given that the education system is domineering, I have no doubt that the military system is, too. "War is the time of the active and courageous, and not of the clean-shaven." — Yevgeny Prigozhin, in criticising Russian military discipline. Supposedly, such discipline is useful for "creating the mental framework that enables soldiers to follow orders under extreme stress". What extreme stress? Why do they need to be under extreme stress as opposed to having fun? If they're under extreme stress, they shouldn't be "serving" and doing their "duty" in the first place. Someone is not fit for combat if it stresses them out. Being active and courageous makes one much fitter for combat than looking neat. The education system does not enable that kind of activity and courage. It crushes masculinity.


If a state “needs” to conscript, that is because it is abusing its own people — just as a happy family does not need internal coercion to defend itself from a home invasion, for example. It is not impossible to make such a happy family. It is not impossible to make a similarly happy nation.

People who accept conscription are simply so used to big fish eats small fish, and so hollowed by humanity, that slavery is acceptable to them. It is basically slavery. Slavery was gotten rid of in the past, and then, guess what, it clearly wasn't needed. We can do it again.

Also, for a government to advertise its military to its citizens is like a supermarket stealing potential customers’ money to advertise to them. The state should be in alignment with the people such that advertisement is not considered. If a leader has a number of soldiers they want and they aren't getting the desired number, it would be best not to try to manipulate their own people into getting them to join the military for the sake of a higher number.

You don't need to manipulate people into doing what is best for them.

Politicians evidently got so invested in their little game that they forgot to care about people. FAFO


My vision for military activity is one of campaigns — towards world peace, security and advanced civilization — that are joinable not only by professional soldiers. They should wear cameras so that people can learn from their footage and to deter criminality. Returning combatants should have access to living quarters specifically for them. All this should be voluntarily funded. There should be a social media connecting funders to fighters and connecting jobseekers to opportunities.




www.internethub.info  (links to web pages, mainly YouTube channels)




angulardoublehelix@gmail.com — support wanted