Let’s Design A School, Part 3.1: Bringing it all together with the Sieve Model

Link post

In part 1, we laid out the social services model of a school.

In part 2, we described a new educational model of a school.

In part 3, we’re going to combine them.

Different Schools, Different Problems

The hardest part of designing a public school is that you’re trying to create a one-size-fits-all solution to an array of vastly different problems.

In wealthier districts, where parents are more involved, discipline is less of an issue. More importantly, things like making sure children get fed is less of an issue. The school can focus on education.

In poorer districts, the school often can’t do a whole lot of educating, because the students aren’t in a state conducive to learning. Their parents might be uninvolved for any number of reasons; their home lives might be fine or they might be abusive. Nothing can be taken for granted about the health or well-being of any student who walks in the door.

Urban districts will face different issues from rural ones.

Things like weather and climate can play a role—some schools may need to close multiple times a year for snow days; others might need heavy air conditioning to be habitable.

In all cases, students with physical and mental conditions need to be accommodated. Parental participation, while appreciated, cannot be counted on. Transportation must be provided.

And hopefully some education should take place.

Making It Through The Sieve

In part 1, we introduced the Sieve Model of school-as-social-services. Students enter the building, and are filtered into different paths depending on their physical and mental well-being, climbing Maslow’s hierarchy as they go. If they need to eat, they eat; if they need professional help, they get it.

With school-as-social-services, students who make it through the sieves are mostly left to their own devices. They’re still supervised by adults, but they’re free to play or eat or be at a computer about as much as they want. The children were safe and fed; our goals didn’t exceed the basics.

Now, with a model of school-as-education to work with, we have a curriculum for students who are doing okay. First they’re taught literacy and numeracy, then their core civilizational requirements and survey courses, and lastly core adulting requirements and self-study. By the time a student’s graduated, we should have a reasonable confidence that they’ve been either a) given the tools to make their own way in the world, or b) at the very least kept fed and safe and healthy during their formative years.

Flexibility In Design

We’ll solve the issue of each school having vastly different needs by letting each school strike its own balance between the two models. Given the finite resources they have and the student population they’re working with, school administration and other stakeholders will decide on a year-by-year basis how much of their budget is allocated to each of the different sieves and curricula.

For schools with student populations in need of more assistance, more of the budget can flow to counselors, food, and social workers. For schools whose students are mostly ready and able to learn, they can spend their money on teachers and the curriculum.

A few points:

  • The entirety of the curricula—lectures, online classes, etc. (basically everything except in-person work) should be available online, to anyone, for free. While creating this is a significant capital expense, once created it takes a lot off the shoulders of individual schools and teachers, especially the ones with less resources to devote to the curricula. This also makes it available in other countries, at military bases, and for homeschooled children.

  • Teachers, in this environment, spend most of their time on in-person activities with students, answering questions, and helping students with self-study. What they don’t do is act as guidance counselors, social workers, or therapists. Because the school already has those.

  • Structure (assigned times for classes, lunch, or other activities) can be provided (and perhaps should be the default), but it doesn’t have to be: there aren’t many parts of this system that absolutely have to be completed by any particular date or on any particular schedule. This should be helpful for students who take care of younger siblings or work. Simultaneously, there will be adults in the building over a wide range of time, which means that parents can still get the same babysitting function out of the new system that they get with the existing one.

Compare and Contrast

It’s easy to dream up a new system and declare that it’s better than the old one. It’s somewhat harder to carefully think about what problems one is trying to solve, and comparing one’s solution to the existing system on the merits of how each solves the problems and performs the functions they’re supposed to solve and perform.

Which is a long-winded way of saying that we need to check to see if our system is actually any better than the existing one or not.

The Problem(s)

What is the point of this whole ‘school’ thing, anyway?

The answer to that question has changed over time. Answers include:

  • Educating the populace, on the basis that an educated populace is a necessary ingredient for a functioning republic

  • Assimilating children, especially immigrant children or children of immigrants, into American culture

  • Preparing children for adulthood

  • Creating productive workers for a factory (or other economically productive work)

  • Babysitting children so their parents can work

  • Building a STEM workforce for national security or pride reasons

  • Helping children in abusive or unsafe situations

  • Employing teachers

To be clear, these aren’t necessarily the answers I would give, just the ones that I’ve seen used before.

The Solution: Existing System

How well does the existing system solve those problems?

Well, it does a decent job babysitting children so their parents can work. It employs plenty of teachers. It seems to do a pretty good job assimilating children. When factory work was a large percentage of jobs, it produced people who could do that work.

As for the rest of it, well…

Some nonzero amount of education takes place, that much isn’t debatable. What is debatable is how effective, useful, and relevant that education is. Granted, COVID had large deleterious effects and was pretty recent, but the statistics aren’t pretty even without it.

There are inner city schools where 93% of 3rd through 8th graders aren’t able to do math on a grade level.

From the Nation’s Report Card:

In 2022, the percentage of fourth-grade public school students performing at or above the NAEP Proficient level in reading was 32 percent nationally.

And even in schools where things are successfully taught, how much of what is taught is used for anything other than passing tests?

Think back to your K-12 school experience: how much of what you learned actually matters, or is used in your day-to-day life?

Are students able to become productive members of society because of schools, or despite them?

While I don’t have the data or the statistical know-how to actually answer that question, my experiences tend towards the following conclusion:

If you were going to succeed academically anyway, existing schools can be a tool to help you do so. If you weren’t going to succeed academically, or are looking for more applicable life skills, existing schools are not going to help you by default.

From what I’ve heard from people who’ve been there, the existing system breaks down completely if the number of social issues (special needs students, behavioral issues, etc.) exceeds a certain threshold, because the existing system isn’t designed to provide social services at scale.

The Solution: New System

First, what the new system doesn’t do: it doesn’t employ nearly as many teachers. This is on purpose, because currently teachers are asked to do several different jobs, whereas in the new system they’ll only be asked to do one.

Instead of teachers, the new system employs more social workers and counselors to do the job of social work and counseling. The additional burden of teaching is taken up by online curricula and resources.

Second, the system does not create productive factory workers, because that isn’t what today’s labor market looks like.

Third, it doesn’t necessarily do the same kind of assimilation the current system is set up to do. Some assimilation will no doubt happen by default, but the new system is far less *cough* systematic about it. (We suspect a great deal of the assimilation is from forcing students to be together for a decade plus, and letting peer pressure do its work, rather than intentional indoctrination.)

Now for what the new system does do.

The new system should provide babysitting for children, same as the old.

It should do a better, more systematic job of helping children who are abused or unsafe or hungry or in distress, because it’s actually designed to do that.

It should do a passable job at preparing children for adulthood and the modern career environment.

It should create a populace at least as educated as the current one.

It should create the opportunities for students with STEM inclinations to pursue them.

Lastly, and in our opinion most importantly, the new system does this flexibly, at scale, and gives students the freedom to advance at their own pace.

Academic Pace

The current system of public schools in the US is designed to support the median student. Those who can advance academically much faster or slower than the median have options, but those options are not really good.

For those behind the median pace, they can be held back a grade or take remedial or otherwise “basic” courses.

Being held back a grade forces the student into an entirely different social environment where they’re older than their peers and judged for being held back. It also burdens the school system with another student-year of costs.

Taking basic or remedial courses might help, but a) it’s an additional burden on educators to design and teach more basic courses, b) it inflicts a stigma upon those in said courses, and c) a basic or remedial course still moves at the pace of its median student, meaning that anyone who learns faster or slower than that still has a problem.

For those ahead of the median pace, they can skip a grade, take advanced (honors, AP, etc.) courses, or receive some kind of private tutoring.

Skipping a grade, like being held back a grade, puts the student into an entirely unfamiliar social environment, because in the existing school system academic and social environments are tied together.

Taking more advanced classes or electives, taking classes at a local community college, or receiving private tutoring can help fast learners, but the experience can still be socially isolating. It also tends to cost more, and not every school or parent has the resources to support such classes, electives, or tutoring.

Furthermore, the pace a student learns academically may not be the same as the pace they learn socially. This is important—it means that being moved forward or back a grade in today’s world always involves a tradeoff between academic and social health. Our new system eliminates this tradeoff by separating the two.

We spent a long time discussing our core requirements, because they’re the most original part of the new design, but the rest of the curriculum is incredibly flexible. Because the entirety of it is available online, students can move at their own pace through the survey courses of phase 2 and into the independent study of phase 3. Students who need more time can have it without any judgement or stigma; students ready for more difficult work will find it available whenever they want it.

Because the new system doesn’t segregate by age, students are free to mingle and remain in their friend groups if they want to, regardless of the pace of their academics. They can agree to meet up for lunch when they want; they can embark on core requirements or independent studies together if they want.

Every student (and their parents/​teachers) can find the balance between academic and social development that’s right for them, because the two are no longer tightly coupled.

Schools with fewer resources can concentrate those resources on the students that need them the most, while the students who can succeed on their own are given all the tools necessary to do so. Exceptional students can be accelerated as fast as they can go, while students that need more time will have it.


In this post we laid out the combination of our academic and social services models of school, and how they merge to create a flexible design for public education. We’ve compared some of the advantages of this design against the current school system.

Next post, we’ll take a look at the budgets, and ask:

Is this actually affordable?