I sometimes feel like American education is under a massive collective delusion. You might have heard the phrase "High Quality Instructional Materials," often abbreviated as HQIM. That phrase does not mean what you might think it means.

A bit of history. EdReports is a website for curriculum reviews, founded in 2015. The Common Core standards were relatively new. At the time, many curriculum companies stuck a "Common Core Aligned" sticker on their existing materials while making few actual changes. There was a need to figure out which curricula were in fact Common Core aligned. EdReports filled that gap. That was the original goal: help districts understand whether curricula were truly aligned to the standards.

Somehow, EdReports is still the only major organization reviewing curricula. And somewhere along the line, principals and superintendents all decided that if EdReports gives a curriculum all green (their best rating), then they must be High Quality Instructional Materials. That's what the phrase means to most people. That's all. A perfect score on EdReports is necessary and sufficient to be called High Quality Instructional Materials.1

Plenty of other writers have pointed out issues with the idea of High Quality Instructional Materials. My goal is to give a specific example from a real curriculum to illustrate what this stuff can look like in practice.

An Example

Below is an example from a curriculum that got a perfect score on EdReports.

This is the culmination of a two-lesson sequence on multiplying rational numbers, and it is the first time students have seen multiplication with negative numbers. The two lessons include a few activities introducing the concept to students and doing a quick check for understanding. This curriculum offers teachers some choice so there are multiple ways the teacher can introduce the content. Then, there are two pages of practice.

(I won't reproduce the exact curriculum here or name the program because I don't want to be sued for a copyright violation. If you would like to see it or learn more, feel free to email me.)

The first page begins with these four problems:

1. -8(7)

2. 5(-13)

3. -3.4(3)

4. 4.8(-6)

Then there are four word problems. The word problems are fine; a bit contrived, but so are most word problems. I won't focus on them here.

The second page begins with these four problems:

9. -4(-9)

10. -8(-12)

11. -4.1(5)

12. -12.6(-5)

Then there are three word problems; again, a bit contrived, but fine.

That's the entire set of practice problems for this topic. There's a lesson on division, and then a lesson focused on combining multiple operations, like -5(-8) - 7, though there are only six total multiplications in the lesson. Then there are a few more multiplication problems in the review questions at the end of the unit.

My claim is that, at a very basic level, you cannot call this an example of high-quality curriculum.

So what are the issues?

Not enough practice

It's hard to judge how much practice students need. Curriculum designers want to avoid too much repetitive practice, while also providing enough for students to have a solid understanding of the topic. There is no perfect number. But this is absolutely not enough practice. Multiplying with negatives is an important skill, and most students will need more than 15 total problems.

The numbers are poorly chosen

8 × 7 is typically one of the last multiplication facts that students commit to memory. Very few students have 5 × 13 committed to memory. This matters because, when students are learning a new topic, they should focus on the concept that's new: in this case, whether the answer is positive or negative. When I teach this topic, I ask a ton of questions like -3 × 2, or -4 × (-10). I use easy numbers that students can multiply without having to think, so they can focus their thinking on whether the answer is positive or negative. This curriculum does the exact opposite. Is there a place for questions with bigger numbers and decimals? Absolutely. But not at the beginning of practice, where decimals and bigger numbers pull student attention away from the new concept.

The notation is repetitive

Every question is written in the form 2(-3). That's great notation — it's important students understand what those parentheses mean. But they should also see other notation, like 2 × (-3), and 2·(-3), and (2)(-3). There should be a few problems like “find 3x if x = -5.” Variety helps students use different notation flexibly in the future.

Students can answer questions correctly without understanding the core concept

This is maybe the most important one. On the eight problems on the first page, every problem involves multiplying a positive and a negative, with a negative answer. Students can avoid understanding the concept entirely, and just say "oh the answer must be negative." Then on the second page, every question involves two negatives and a positive answer. A bit of blocked practice is fine at the start, but it should transition as quickly as possible to interleaved practice mixing different problem types together.

The practice isn't very ambitious

There are a bunch of problem types that aren’t present. There are no fractions. There are no problems multiplying more than two numbers, like (-2)(-2)(-2). Those are important problems as students become comfortable with the basics of multiplying two numbers, so they can extend their knowledge to different contexts. For some reason, fractions and more complex problem types appear in the unit review but not in the actual lessons.

The Delusion

These are not high quality instructional materials. Learning to multiply negative numbers is a skill that will come up over and over again, both later in 7th grade math and in future years. It’s worth learning. These materials come up short. The reality is most school leaders making decisions about curriculum don’t have the time to pick through curricula with this level of detail. Many don’t have the expertise to know what they should be looking for in a math curriculum. I understand why they want to farm out some of that decision-making to an organization like EdReports. It’s a shame that EdReports doesn’t do a good job, and “High Quality Instructional Materials” has become synonymous with a high score from EdReports all the same.

As you read this, somewhere there is a principal or math curriculum lead in a meeting with a superintendent or a school board. They’re getting up on their high horse, talking about how important High Quality Instructional Materials are. Maybe they’re talking about how they’ll require teachers to follow their new curriculum with fidelity. And then they go and adopt a curriculum like the one I described above.

A Few Notes

• This is one lesson. You might wonder if I’m cherry-picking. This lesson is particularly poor compared to others from this curriculum. Most lessons do a better job. But it’s not uniquely bad — there are plenty more with similar problems. And multiplying rational numbers is a particularly important concept, so I think it's a fair example to use. If the lesson on complementary angles doesn’t do a great job structuring interleaved practice, life will go on. But you have to get multiplying rational numbers right. If this curriculum can get a perfect score despite lessons like the one I described, something is wrong.

• There are good curricula out there. Not all programs rated by EdReports are as poor as what I'm describing. The issue is that it's very hard to tell with the kind of cursory glance most schools have the resources to give. And the cult of High Quality Instructional Materials means that people assume, since it has a good score from EdReports, it must be good.

• Ironically, I helped choose this curriculum. Teachers in my district were given two choices for middle school math. We looked at this program and one more. The other was worse, and it also got a perfect score from EdReports.

• I use the JUMP materials (which are not perfect, but I like a lot) to supplement my primary curriculum. They aren’t currently rated on EdReports.2 Does that mean I’m using Low Quality Instructional Materials to compensate for the shortcomings of High Quality Instructional Materials?

• If I were to generalize broadly, the most common issue with “High Quality Instructional Materials” in math is practice. Practice is important, but it’s not evaluated at all by EdReports. I’ve seen and worked with a number of highly-rated curricula, and a lack of practice or poorly-structured practice are common issues.

• I could name many more issues. Too many representations without enough sustained effort on each individual representation. All sorts of fluff about mindset and social-emotional learning that feels like it was thrown together at the last minute. A lack of spaced practice. Little attention paid to prior knowledge. Cluttered, hard-to-use handouts. Clunky digital interfaces. All in highly-rated curricula. My goal with this post isn’t to write an exhaustive critique, just to give one specific example of one way a highly-rated curriculum can fall short.

The moral of the story is that we have a desperate need in American education for someone to compete with EdReports, offer a serious definition of high-quality curriculum, and give decision-makers another source of information.

Raymond Callahan's 1962 book Education and the Cult of Efficiency remains a classic study of public education in the US, chronicling how in the early twentieth century schools' goals became business goals.

"The procedure for bringing about a more businesslike organization and operation of the schools was fairly well standardized from 1900 to 1925," Callahan argues in his opening chapter. "It consisted of making unfavorable comparisons between the schools and business enterprise, of applying business-industrial criteria (e.g., economy and efficiency) to education, and of suggesting that business and industrial practices be adopted by educators."

This move – the whole "factory model of education" thing – is often associated with the rise of scientific management, but as Callahan demonstrates, the transformation of schools began in the years before the 1909 publication of Frederick Winslow Taylor's book. No doubt, The Principles of Scientific Management would be wielded to reshape – to "engineer" – almost every aspect of society; and schools were a particular target for intervention, as they were decried by politicians, business leaders, and journalists alike as "crude, unscientific, and wasteful.” What was needed, they argued, was better management, both in and out of the classroom.

Management meant measurement, as Taylorism would dictate, and what emerged were all sorts of new, bureaucratic practices in schools: the recording of attendance, grades, test scores, and so on. Instruction, supplies, the layout of the classroom, drills – all this was to be standardized in order to control inputs and outputs, to reduce expenditure, to eliminate waste, and – ostensibly at least – “to increase quality of product, i.e., the pupil.” But teaching itself – the what and the how of it – really mattered only insofar as it was a line item in the school budget, something (someone) to be managed, work always to be done more cheaply.

What emerged in this Taylorist milieu was a new profession, the school administrator, whose interests were not pedagogical as much as financial. This is the “cult” Callahan refers to in his title – a group that believed unwaveringly in scientific management, despite there being very little “science” at all that supported its application to education (hell, even to the factory); a group whose values changed little, Callahan argued, over the course of the twentieth century and had in fact become more and more entrenched, shaping how education was even conceived. And there, efficiency remained – and remains -- the priority.

The whole development produced men who did not understand education or scholarship. Thus they could and did approach education in a businesslike, mechanical, organizational way. They saw nothing wrong with imposing impossible loads on high school teachers, because they were not students or scholars and did not understand the need for time for study and preparation. Their training had been superficial and they saw no need for depth or scholarship. These were men who in designing a college provided elaborate offices for the president and the dean and even elaborate student centers but who crammed six or eight professors in a single office and provided a library which would have been inadequate for a secondary school. ... They saw schools not as centers of learning but as enterprises which were functioning efficiently if the students went through without failing and received their diplomas on schedule and if the operations were handled economically.

One should view the history of education technology in the twentieth century alongside Callahan’s history of education, his history of school administration and scientific management. While there is a tendency to see ed-tech as a matter of instruction – as tools that reshape teaching (and learning) – these are often, more accurately, tools of management, or prescriptive technologies in Ursula Franklin’s framework. This is the learning management system, most obviously. But it is also the “productivity suite,” the software through which almost all school work (and thus all thinking) is assigned and accomplished, where students and teachers can be monitored and timed, surveilled and controlled.

The problem

I have dabbled with keeping journals in the past, but I've never managed to continue it for more than a year. The pattern is always the same: I struggle to write in the journal with any regularity, and sooner or later, I fall out of the journaling habit entirely. My problem is not that I run out of enthusiasm, but that I have trouble maintaining a reasonable scope for my entries.

Completionist impulses are my downfall. I'm prone to being too thorough – I want to record every event, document every detail, and capture every worthwhile thought from the day in service of compiling a meaningful chronicle of the life I've lived. But, as a logistical matter, this approach has proven untenable. Writing entries in that maximalist style takes hours. Eventually the task of recording a new day feels too burdensome to be enjoyable – to the point where I give up on writing new entries entirely.

I've long understood that I would benefit from putting some kind of constraint in place to guard against my burnout-inducing tendencies. And for a years, I've searched for one. Yet the minimalist alternatives I came across were unsatisfactory – and unsatisfying. The bullet journal method, for example, never resonated with me. It felt too much like a daily planner, or a scattershot lumping of to-do lists and goals that, to me, seemed symptomatic of an irritating hustle-and-grind mindset. Maybe it could help people with organization, but it didn't look like it produced a journal worth revisiting – and what's the point of keeping a journal if you're not going to read through it eventually?¹

What I needed was a method that would make journaling feasible while still producing a text I'd enjoy reading. In other words, I sought a journal format that would not only assist with remembrance, but would also make the act of remembering pleasurable. But what format could accomplish that?

The solution

I stumbled upon a solution in an unlikely place: Thousand Year Old Vampire, an excellent solo role-playing game by Tim Hutchings. In it, you chart the long life of a vampire in the first person, responding to random prompts as you follow the wax and wane of your vampire's fortunes. (You don't even need dice to play, since the rulebook includes random number tables you can rely on instead.) It's immensely enjoyable, and I've filled entire notebooks with my playthroughs.

What makes the game especially interesting, though, is how it treats memory.² You must respond to prompts based on the resources your vampire has to hand and the history that it can remember. But its capacity for memories is finite, and as the centuries pass, you must choose which aspects of your vampire's life to retain or forget. This leads to poignant moments where, late in its life, your vampire can no longer recall the time when it was human, or finds that it has lost all trace of friends and family who mattered dearly to them in earlier times.

The game uses an ingenious mechanic for tracking all this. Your vampire holds five "Memories," which each have room for three "Experiences." The rulebook defines them as follows:

Memories and Experiences are important moments that have shaped your vampire, crystallized in writing. They make up the core of the vampire’s self—the things they know and care about. An Experience is a particular event; a Memory is an arc of Experiences that are tied together by subject or theme.

Experiences cover a particular event, but the amount of time represented by that event might vary dramatically. An Experience might describe a few seconds of impactful events, or it might cover two hundred years of lurking in an old castle.

[...]

An Experience should be a single evocative sentence. An Experience is the distillation of an event—a single sentence that combines what happened and why it matters to your vampire. A good format for an Experience is “[description of the event]; [how I feel or what I did about it].” If necessary, you can add an em dash at the end to include more information.³

As for how an experience might be written, the rulebook offers the following example:

Stalking the deserts over lonely years, I watch generations of Christian knights waste themselves on the swords of the Saracen; it’s a certainty that Charles is among them—I dream of his touch as I sleep beneath the burning sand.⁴

Thinking about the game recently, an idea occurred to me: Why not try keeping a real-life journal using the same strict format?

A daily journal entry could be as simple as "[description of occurrence] + [its upshot]," adapted as necessary.

I therefore decided I would approach journaling from this angle. I would aim to constrain each day's entry to one or two key things, and limit their expression to one or two sentences.

I allowed myself that latter tweak because the initial Thousand Year Old Vampire rule encourages improper semicolon usage and ugly run-on sentences, both of which strain language in hopes of cramming in as much as possible. The example sentence quoted above is rather clunky, after all – it clearly yearns to be two or three separate statements, and would be much more elegant if presented that way. I did not want such tortured, aesthetically unappealing text in my journal. I determined that capping entries at two sentences would leave space for well-composed writing without violating the spirit of concision behind the rule.

Benefits of the two-sentence approach

After conducting this experiment for a good while now, I'm pleased to have found it an unqualified success. Placing an ultra-strict ceiling on my journal entries has allowed me to maintain a continuous daily writing streak since starting. Not only has it been fun, it has brought about some other benefits, too. Here are some of the things I've found especially effective or rewarding about the two-sentence approach.

It requires almost no activation energy

Have you ever found that, the larger an undertaking, the less likely you are to start it? Maybe it's because you can't schedule the long block of uninterrupted time necessary for the task. Or it's because the prep work involved is an undertaking in and of itself. Or else the difficulty is that it takes too much effort to shift into the "flow" mindset you know you'll need.

I certainly have this problem – and it has halted my journaling efforts in the past.

Luckily, the two-sentence method sidesteps this issue by significantly reducing the activation energy necessary to sit down and write a journal entry. There's always time in the day to jot a sentence or two. And it's not much effort to write two sentences – it's the character equivalent of a text message or a microblog post!

Honestly, in my experience, the hardest part about keeping a two-sentence journal is remembering to do it.

It's reflective and meditative

The two-sentence limit forces you to think carefully about what you found most important, meaningful, or interesting about the day. I've found that fun in and of itself – it's like a puzzle, trying to home in on what I'd call the day's defining aspects.

But paring back the noise of the day's events doubles as a reflective exercise. Going over what happened during your day to pick out the part(s) that mattered most to you lets you process and appreciate all you've experienced in the past 24 hours. I have noticed that I'm more grateful for the things I choose to write about after I've sifted them from the rest of my day.

No day is wasted

With respect to the preceding point, to keep up with the journal is to engage in regular reflective/meditative exercise. It's good for you – you're always making the most of your day.

Yet there's also a sense of accomplishment that comes from making something. When you keep a regular journal, even your worst days add material to the book of your life. The two-sentence format makes it easy to take advantage of this principle, ensuring you can wring value out of every day.

This is both particularly apparent and particularly gratifying if you use a physical notebook for your journal. I fill out a page of the Moleskine I've dedicated to this project every 4-5 days, so I watch the literary output pile up at a steady rate. It feels great!

Days often turn into short stories

This is more of an emergent property that has nonetheless surprised me.

When I've looked back on the longer journal entries from my past efforts – or the entries in published journals by famous authors and thinkers – I've seldom found a narrative arc to them. It's not usually the intent behind such compositions, after all. They're typically collections of things that happened, and life rarely unfolds in accordance with the neat story structures of literature and film.

But an interesting thing happens when you present a reader with only two sentences: Somehow, a narrative logic appears.

I'm not entirely sure why this occurs, but I do have a working hypothesis. It's that noise dilutes story. When too many details pile up, or too many divergent threads appear in one place, it can be difficult to ascertain the overall point of the whole composition. This is basically what happens when somebody rambles in their writing or conversation – too much irrelevant material is brought in, such that the audience loses sight of the point.

Yet it's almost impossible to ramble when allotted only two sentences. This means that two-sentence journal entries tend to cut right to the point, with every inclusion being relevant. Thanks to that concise presentation, every entry seems to have a visible arc. And as a result, narratives often materialize from the short texts each enty comprises. (I've included some examples from my own journal toward the end of this post to help illustrate what I mean.)

It demonstrates the aesthetic merit of the two-sentence paragraph

Perhaps the curriculum has changed since I was an elementary schooler in the United States, but I remember having it drilled into me that a paragraph was always a minimum of three sentences. Those lessons instilled in me a pathological aversion to anything less. I carried that three-sentence prejudice for years, even shying away from the one-sentence paragraphs that have proven rhetorically effective in many works of prose. If you were schooled similarly to me, you might have the same hang-ups.

Yet my experiments in constrained journal formatting have taught me that the two-sentence paragraph is a remarkably powerful construction.⁵

Like Kuleshov's and Eisenstein's theories of cinematic montage – which observe that the juxtaposition of two images creates an impression or idea distinct from what either one produces in isolation – placing two sentences alone with one another in a single paragraph leads to intriguing effects.

Sometimes the two sentences assume a causal relation, with the first triggering the second. On other occasions, they invite a comparison, with each one throwing the other into relief. They can read like a reversal, where the second undermines the first, or a redirection wherein the second suddenly makes clear an unexpected dramatic trajectory. There are even times where you end up with an ambiguous but palpable connective logic, like what the kireji in a haiku provides, and contemplating the nature or meaning of that connection gives the paragraph its force.

The two-sentence journal provides ample space to explore these effects – and to observe their potency firsthand.

Sample entries

To give you a sense of how fun and evocative the two-sentence journal format can be, here are some entries from my own journal this past week.⁶ I'm amazed by the inner lives and wider worlds you can conjure despite (or because of) such tight constraints.

Limped my way to the work day's end, exposing how badly I need the week-long vacation I now begin. A surprise downpour cut short the evening walk with my wife and dog, but, drenched to the skin though we were, we laughed the entire trek home.

Torrential rains swept through our part of town all morning, washing away the road outside our neighborhood's entrance. Tomorrow the nearby construction sites will resume felling every tree in reach and introducing impermeable surfaces where grass once grew.

I spent this last day of my vacation polishing the living room table – a piece older than I am that once belonged to my parents. It absorbed the applied mineral oil like it thirsted, and as I ran the cloth over its sturdy surfaces, I felt as though it were a living being whose care I had neglected.

Parting thoughts

So there you have it – the complete rundown of the two-sentence journal method.

If my approach sounds appealing or interesting to you, I encourage you to try it for yourself. I have greatly enjoyed keeping a two-sentence journal, and I imagine you will, too.

And if you come up with any entries you're willing to share, please consider sending them my way! I'd be curious to see what people do with this journal format and where they take it.

Notes

This essay was originally published in The Atlantic on August 4th, 2025.

One common explanation for why children spend so much of their free time on screens goes like this: Smartphones and social-media platforms are addicting them. Kids stare at their devices and socialize online instead of in person because that’s what tech has trained them to want.

But this misses a key part of the story. The three of us collaborated with the Harris Poll to survey a group of Americans whose perspectives don’t often show up in national data: children. What they told us offers a comprehensive picture of how American childhood is changing—and, more important, how to make it better.

In March, the Harris Poll surveyed more than 500 children ages 8 to 12 across the United States, who were assured that their answers would remain private. They offered unmistakable evidence that the phone-based childhood is in full force. A majority reported having smartphones, and about half of the 10-to-12-year-olds said that most or all of their friends use social media.

This digital technology has given kids access to virtual worlds, where they’re allowed to roam far more freely than in the real one. About 75 percent of kids ages 9 to 12 regularly play the online game Roblox, where they can interact with friends and even strangers. But most of the children in our survey said that they aren’t allowed to be out in public at all without an adult. Fewer than half of the 8- and 9-year-olds have gone down a grocery-store aisle alone; more than a quarter aren’t allowed to play unsupervised even in their own front yard.

Subscribe now

Yet these are exactly the kinds of freedoms that kids told us they long for. We asked them to pick their favorite way to spend time with friends: unstructured play, such as shooting hoops and exploring their neighborhood; participating in activities organized by adults, such as playing Little League and doing ballet; or socializing online. There was a clear winner.

Children want to meet up in person, no screens or supervision. But because so many parents restrict their ability to socialize in the real world on their own, kids resort to the one thing that allows them to hang out with no adults hovering: their phones.

Share

Since the 1980s, parents have grown more and more afraid that unsupervised time will expose their kids to physical or emotional harm. In another recent Harris Poll, we asked parents what they thought would happen if two 10-year-olds played in a local park without adults around. Sixty percent thought the children would likely get injured. Half thought they would likely get abducted.

These intuitions don’t even begin to resemble reality. According to Warwick Cairns, the author of How to Live Dangerously, kidnapping in the United States is so rare that a child would have to be outside unsupervised for, on average, 750,000 years before being snatched by a stranger. Parents know their neighborhoods best, of course, and should assess them carefully. But the tendency to overestimate risk comes with its own danger. Without real-world freedom, children don’t get the chance to develop competence, confidence, and the ability to solve everyday problems. Indeed, independence and unsupervised play are associated with positive mental-health outcomes.

Still, parents spend more time supervising their kids than parents did in the 1960s, even though they now work more and have fewer children. Across all income levels, families have come to believe that organized activities are the key to kids’ safety and success. So sandlot games gave way to travel baseball. Cartwheels at the park gave way to competitive cheer teams. Kids have been strapped into the back seat of their lives—dropped off, picked up, and overhelped. As their independence has dwindled, their anxiety and depression have spiked. And they aren’t the only ones suffering. In 2023, the surgeon general cited intensive caregiving as one reason today’s parents are more stressed than ever.

Kids will always have more spare hours than adults can supervise—a gap that devices now fill. “Go outside” has been quietly replaced with “Go online.” The internet is one of the only escape hatches from childhoods grown anxious, small, and sad. We certainly don’t blame parents for this. The social norms, communities, infrastructure, and institutions that once facilitated free play have eroded. Telling children to go outside doesn’t work so well when no one else’s kids are there.

Leave a comment

That’s why we’re so glad that groups around the country are experimenting with ways to rebuild American childhood, rooting it in freedom, responsibility, and friendship. In Piedmont, California, a network of parents started dropping their kids off at the park every Friday to play unsupervised. Sometimes the kids argue or get bored—which is good. Learning to handle boredom and conflict is an essential part of child development. Elsewhere, churches, libraries, and schools are creating screen-free “play clubs.” To ease the transition away from screens and supervision, the Outside Play Lab at the University of British Columbia developed a free online tool that helps parents figure out how to give their kids more outdoor time, and why they should.

More than a thousand schools nationwide have begun using a free program from Let Grow, a nonprofit that two of us—Lenore and Jon—helped found to foster children’s independence. K–12 students in the program get a monthly homework assignment: Do something new on your own, with your parents’ permission but without their help. Kids use the prompt to run errands, climb trees, cook meals. Some finally learn how to tie their own shoes. Here’s what one fourth grader with intellectual disabilities wrote—in her own words and spelling:

This is my fist let it gow project. I went shoping by myself. I handle it wheel but the ceckout was a lit hard but it was fun to do. I leand that I am brave and can go shop by myself. I loved my porject.

Other hopeful signs are emerging. The New Jersey–based Balance Project is helping 50 communities reduce screen time and restore free play for kids, employing the “four new norms” that Jon lays out in The Anxious Generation. This summer, Newburyport, Massachusetts, is handing out prizes each week to kids who try something new on their own. (Let Grow has a tool kit for other communities that want to do the same.) The Boy Scouts—now rebranded as Scouting America, and open to all young people—is finally growing again. We could go on.

What we see in the data and from the stories parents send us is both simple and poignant: Kids being raised on screens long for real freedom. It’s like they’re homesick for a world they’ve never known.

Granting them more freedom may feel uncomfortable at first. But if parents want their kids to put down their phones, they need to open the front door. Nearly three-quarters of the children in our survey agreed with the statement “I would spend less time online if there were more friends in my neighborhood to play with in person.”

Share

If nothing changes, Silicon Valley will keep supplying kids with ever more sophisticated AI “friends” that are always available and will cater to a child’s every whim. But AI will never fulfill children’s deepest desires. Even this generation of digital natives still longs for what most of their parents had: time with friends, in person, without adults.

Today’s kids want to spend their childhood in the real world. Let’s give it back to them.

Subscribe now

As a teacher, I am continually thinking about when I should talk and when I should listen. Early in my teaching career I believed that the type of textbook dictated who was talking in my classroom. I started my career teaching with a problem-based curriculum, which meant students solved problems together by talking to each other and my job was to walk around the classroom listening to the conversations and then use what I heard to lead a class discussion.

My struggle while teaching with this approach was deciding when and what to talk about. I didn’t want to tell students too much and take away their thinking. At the same time, I felt the need to talk and show examples since that is what my teachers did when I was in school.

Later in my career, my school adopted a more traditional textbook and my role as a teacher changed. The book was full of examples so I would talk for most of the hour as I worked out examples in front the class as my students listened and vigorously copied the examples onto their notes.

My struggle with this approach was finding ways to help students talk about their ideas and explain their thinking. I knew if I talked too much, my students would stop listening.

Subscribe now

Recently, I have found some answers by digging into different types of instruction. Munster, Stein, and Smith (2015b)1 define two types of instruction: dialogic and direct.

• In the direct instruction model, when students have the prerequisite conceptual and procedural knowledge, they will learn from (a) watching clear, complete demonstrations of how to solve problems, with accompanying explanations and accurate definitions; (b) practicing similar problems sequenced according to difficulty; and (c) receiving immediate, corrective feedback.

• Whereas in the dialogic model, students must (a) actively engage in new mathematics, persevering to solve novel problems; (b) participate in a discourse of conjecture, explanation, and argumentation; (c) engage in generalization and abstraction, developing efficient problem solving strategies and relating their ideas to conventional procedures; and to achieve fluency with these skills, (d) engage in some amount of practice. (p6)

For simplicity, let's say that these two models relate to who is doing the talking. In direct instruction, the teacher is doing the majority of the talking while in dialogic instruction the talking is done by the students.

In the book Visible learning for Mathematics, Hattie, Fisher, and Frey (p 24)2 discuss the effect of each of these types of instruction. Stating that direct instruction has an effect size of 0.59 and dialogic instruction has an effect size of 0.82. These authors encourage teachers not to choose one method over the other but to choose which method is best for the situation.

I saw positive results with dialogic instruction early in my teaching career when teaching a problem-based curriculum and wanted to incorporate this model again. So three years ago, I started implementing the practices of Building Thinking Classrooms (BTC). What immediately drew me to these practices was the ability to change traditional “show and tell” examples into thinking tasks. I write about how I did this in the article Building a Thinking Classroom with a Traditional Textbook and students now do the thinking and the majority of the talking in my class.

While dialogic instruction yields a high effect size, it does not always reach all students. As much as I enjoy hearing students discussing math, at some point I need to talk to the class; either to verify that their thinking is correct, to fill in any missing pieces, or take care of misconceptions. Which is why I incorporate both types of instruction in my lessons.

Dialogic Instruction

All lessons start with dialogic instruction. Students can’t explore and engage in new mathematics if they have first been shown how to solve the problems. In BTC, this happens when students are working on thinking problems at the whiteboards. Here, students discuss the problems together and develop conceptual understanding of how to solve them. During this time, I am listening to student discussions to help me offer the proper hints and extensions to each group. I’m also listening for different approaches to solving problems and common misconceptions.

Direct Instruction

Direct instruction comes after students have time to make their own generalizations and evaluate their conjectures. Students need to leave class being sure that they know what they are supposed to know. To do this, I choose a problem that encompasses the learning goal, and we solve it together. Students tell me what to write down and they include this example in their notes. This is a form of direct instruction, but in this case, the students are instructing me, and by writing down their thoughts, I am confirming their thinking. I also use this time to highlight student work by pointing out different strategies and addressing any misconceptions. I write about this in more depth in the article The Power of Consolidating a Lesson.

Direct instruction at the end of the lesson reinforces the concepts that I want my students to understand. I’ve noticed that my students will have wonderful discussions showing me that they understand the lesson goal, but don't feel confident that they have learned the concept until I work through an example in front of them.

It is important to note that since students have already worked through and solved multiple types of problems in groups, they only need me to work through one or two examples with them. Therefore, I have shortened the amount of time needed for direct instruction to the last few minutes of class.

As teachers we know that students need time and practice to let a concept sink in. To do this, I start class with a warm-up. This routine is one of the most important routines I have established and another chance for direct instruction. As students walk into my classroom, I have one or two questions displayed for them to start working on. I usually choose problems from previous lessons giving students the opportunity to check their understanding. After a few minutes of work time, students tell me how to solve the problems. This process fills any gaps that students might have in their understanding and catches up students who were absent. Even though this routine happens at the beginning of the class period, I wrote about it last because it really is a wrap up of previous lessons.

Warm-up time only includes problems that students have solved before and is never a time for “pre-teaching”. I mention this since it's easy to fall into the trap of pre-teaching during warm-ups. In my experience, the temptation to pre-teach is strong and often fueled by the desire to try to make the lesson run smoothly by eliminating the amount students struggle with solving the problems. But productive struggle is part of dialogic instruction and one of the reasons dialogic instruction yields a high effect size. If I pre-teach a concept, then I eliminate the dialogic part of the lesson and thinking problems turn into practice problems.

I also want to mention that warm-ups can be a time sucker, so it is important to keep them short and watch the time. I plan to write more about how I run warm-ups in a future post.

Lastly, there are times when I take the reins and do more of the talking during the direct instruction time. This is done mainly to save time. While I prefer to have students do the talking, I believe it is ok for me to explain problems if I know my students have first solved or tried to solve the problems on their own.

Subscribe now

ChatGPT 5 was released today.

ChatGPT-maker OpenAI has unveiled the long-awaited latest version of its artificial intelligence (AI) chatbot, GPT-5, saying it can provide PhD-level expertise. Billed as “smarter, faster, and more useful,” OpenAI co-founder and chief executive Sam Altman lauded the company’s new model as ushering in a new era of ChatGPT. “I think having something like GPT-5 would be pretty much unimaginable at any previous time in human history,” he said ahead of Thursday’s launch. GPT-5’s release and claims of its “PhD-level” abilities in areas such as coding and writing come as tech firms continue to compete to have the most advanced AI chatbot.

I had a chat with it. Yes I know the questions surrounding AI are tricky. But I do not pretend to address those here; I merely report the following conversation.

1. The thesis outlined.

2. The thesis patiently explained, defended in the face of objections, and further elaborated.

3. The collegial chat and a final effort at rebuttal.

In fairness to GPT5, in my career I have indeed encountered PhDs with this level of commitment to their particular blueberry. And many have also had that blithe confidence — the use of “Ah”, the “Let’s slow it down” (to your two-B level), the “Exactly” (Now you see my genius), the confidently colloquial “Yep” and “Nope” … actually I retract my earlier skepticism; the lad has the makings of a fine philosopher.

Very quick update: Posting this on social media led to some interesting responses. While certainly well-meant, I’m afraid I can’t say I am especially impressed by helpful suggestions that I spend time to do a little prompt engineering in order to get GPT5 to emit the right answer to the question “How many times does the letter b appear in blueberry”. Meanwhile, my excellent student Andrés got a different, and correct, response from his attempt. So score one for replicability, I guess.

Update 2: Look, these tools are amazing in lots of ways. But if you’re gonna pitch and sell—or foist—them on the world at large, or your employees, by saying “This is your PhD level genius expert buddy you can ask things in natural language”, don’t get all pissy when it comically bombs tasks people reasonably think of as trivial.

Update 3, now in meme form: