In-person educators trying to do online learning, getting poor results, and then saying it's the fault of online education is a bit like a YouTube educator teaching in person classes, doing poorly, and then declaring that in-person education is inferior.

Also worth noting that students who have spent years being conditioned with in-person learning might have an adjustment period when starting online education, one that might not go smoothly if it's happening during a crisis and guided by people inexperienced with and predisposed against this form of education.

I imagine a huge factor is also parental involvement, when it comes to encouraging students to set aside several hours each night for study and homework. In the absence of parental involvement or in cases of parental neglect and indifference (woefully common in many situations), the role of the teacher becomes far more important in encouraging the student to develop good study habits. This might be more difficult in the online situation.

One of Malcolm Gladwell's books made a convincing argument that reducing class size quickly runs into the issue of diminishing returns. A cursory Googling suggests there's plenty of research to back it up[0][1]

The effectiveness of in-person learning may have more to do with social interaction and peer motivation. A teacher with about 20 students who is able to create a general excitement for learning in the class seems to be the most effective. A classroom where the "spirit" of learning is alive is ideal for a lot of kids.

[0]https://edcentral.uk/reading-list/431-what-we-have-learned-a... [1]http://hanushek.stanford.edu/publications/evidence-class-siz...

it works out really well for the student. instead of working a regular job during the year you get to do something that benefits your career, you build a relationship with your professors and it pays more than if you were working at the local grocery store.

https://cse.ucsd.edu/undergraduate/readers-and-tutors

https://biosciences.uchicago.edu/content/teaching-assistant-...

> All graduate students are required to serve as a Teaching Assistant in at least one course for academic credit before the Ph.D. degree is awarded. Appropriate courses may be undergraduate, graduate, or medical, but must be in the Biological Sciences Division (exceptions may be made for students in the Biophysical Sciences program).

Personally I was desperate in March 2020 (as were my colleagues) for any tips about what would work online. I didn't want my classes to suck, nor did my colleagues. There was very little information available at the beginning about what worked.

But what's worse (and what makes me disagree with your comment) is that even after a year of being online it still wasn't clear how to make it not suck! Basically no one had discovered anything which made students like it OR perform well.

I polled my students every midterm and final exam (and gave them actual points for their answers!) on whatever they had found to work in any of their online classes. Collectively they were exposed to several hundred other professors at my university. While they did have some suggestions (which I did implement), nothing really worked.

On top of this, schools tend to be constrained with trying to fit everything into the particular confines of what's considered a class. I don't know about your particular school, but all of the ones I've seen would never try to educate students with something like The Odin Project, even though many here think it's a great example of online education.

Also, even with a great program and experienced teacher, changes in education styles entail a transition period for the students. We would expect this to be much slower in such a rocky transition with the professors themselves trying to figure out what to do. But even with poor circumstances I'd expect to see at least some amount of progress as students adjusted to things. Did you not see any?

You mention them being exposed to the classes of several hundred professors at your university. Surely there was some variability in the success the professors had. What difference did the school see between the more successful classes and the less successful classes?

I work in sales and direct face-to-face sales ALWAYS has better close rates, happier customers, etc. than indirect sales through phone or internet.

I sold private whisky casks to individuals or groups. Before the pandemic, I would sit down at the table with them and guide them through a tasting, several hours long. My close rate was close to 50%.

With the pandemic, I switched to shipping samples and moderating the tasting online via Zoom or Teams. My close rate dropped to noise level.

I don't think I fundamentally changed my "tactics", and since it involved drinking more than a few sips of alcohol, I never required the customer to sign the order form in my presence.

In my experience, physical presence was waaaay more conducive to sales.

And sales where you're giving a sample like whisky? The social obligation to reciprocate for receiving something "free" is very strong. The idea behind giving a sample in a sales pitch isn't to get them to like the whisky (though it helps eliminate/reduce resistance), it's to incur that social obligation.

Were these Zoom classes? I can't think of a single synchronous (like Zoom) class that didn't suck.

What I liked as a student was prerecorded lectures that could be replayed at high speed (or even skipped) on my own schedule. Personally I much prefer these over live lectures.

This resource is available on most college campuses for the teachers who have the humility to ask. Those folks were intimately involved in MOOCs and knew a lot about how to make online teaching work. They might have been able to help, but there was a bit of a sigma around talking to them.

Take my 18-yo and 22-yo students. They have responded to my online teaching in a very different way. The 18 year olds responded much better to physical teaching - most vocally preferred it - while older students generally preferred online. I could write a lot more about the details here but I'm on my phone

My daughter teaches math in high school and what she sees can be summed up as: kids are social and need supervision, and a different approach depending on age.

12-year olds are basically still kids but as soon as puberty kicks in kids are overwhelmed by a series of physical, psychological and social changes. This makes life hard to navigate for them and for this reason they need guidance.

So in my opinion remote learning (without local adult supervision) is the worst option for high school.

I wrote briefly about the topic in a blog post, titled "Professors as Creators". It explores the idea briefly, and how treating teachers as "creators" could add value not just to remote learning, but to in-person lessons as well. https://medium.com/age-of-awareness/professors-as-creators-h...

But it's not a choice anymore in India where the online-school-education started out as a necessity during pandemic lock-down and now kids are enrolled in it due to peer-pressure; Parents in India already spent >40% of their income on their kids education before pandemic.

What's worse is that these online school education startups copied Khan Academy's coursework and some even became an unicorn during pandemic with fraudulent practices(False advertising: E.g. 'Kid who took our course got employed by Google', Untrained/Unqualified teachers etc.); So I wouldn't think twice about enrolling my kid into an online-school run by Mr.Khan if I had one.

This 1000x. My son is at UCSD. Very few of the online lectures were better, majority were worse.

I think online can be better, but it requires expertise that doesn't exists in most teaching professionals.

I teach at a university in Japan. I have been teaching my classes exclusively online—live using Zoom—for more than two years, and overall the experience has been better than with the in-person classes I taught for many years before. The class discussions have been meatier and more focused than in person, and the students have been turning in better papers.

I recognize that the results would be different in other situations. I am fortunate to teach small classes of motivated students with good study skills. And as one of the “very online crowd,” I might have been able to adapt to online teaching better than some others.

But the big revolution of online learning is the opportunity it gives for people to take part in interactive classes regardless of their location. Yesterday I taught a graduate seminar with fourteen students, eight in Japan and six in China, including two in lockdown in Shanghai. Everyone was able to take part actively. Starting next Monday, my other graduate class will shift from afternoon to morning Japan time so that a student who is in Mexico and unable to return to Japan can take part in real time. This past Monday, one of the other students in that class was in COVID quarantine near Narita Airport but was able to participate fully in class.

Until recently, it was assumed that the only way to conduct interactive classes in real time was for the teacher and students to all be in the same physical location. If students couldn’t get to campus for whatever reason, they were excluded. Online education opens up educational opportunities for many people who couldn’t participate before.

This in particular:

> The class discussions have been meatier...

is a literal miracle.

I think it might be this:

> I am fortunate to teach small classes of motivated students with good study skills.

I teach an honors and a regular section of one of my classes. The honors students (who are more serious and motivated in general) definitely are affected less.

If you are teaching grad students that is also not going to be representative of the overall population. Those guys are definitely in the right tail of motivation.

I call on students more systematically rather than just waiting for volunteers to raise their hand. When I taught in-person, it was usually the same few students who spoke up a lot, while others never said anything. Now everyone contributes, making for an overall better discussion.

With my larger online classes, I will sometimes throw out a discussion question and give the students five or ten minutes to write up their responses, which they submit through a Google Form. I then display those responses on screen and respond to them. Having the time to write up their responses, and knowing that their responses might be shared with the entire class, seems to make students respond more thoughtfully than if they were just making an ephemeral spoken comment.

I sometimes ask all students to submit questions to me through an online form, too. In in-person classes, many students seem embarrassed to ask questions in front of their peers. That doesn’t apply to online forms. (When I display the students’ questions on the screen, I don’t show the students’ names.)

Some of these ideas come from workshops I attended years ago on “active learning” and could be implemented in the classroom as well. But I began using them only after I started teaching online.

> If you are teaching grad students that is also not going to be representative of the overall population. Those guys are definitely in the right tail of motivation.

You’re absolutely correct.

I can understand why this gives a better learning experience and results (for most students). Hmm, also depends on the subject -- they're saying and discussing things? Eg social sciences? Whilst in maths, maybe there is less to discuss and talk about, and the positive effects you're seeing wouldn't be there?

Seems to me all this can be done in real life too -- but requires more discipline, since the "default", that the more talkative students speak out loud when they want, is what comes naturally? Rather than writing questions and thoughts on paper and handing in, and you read them.

Would be interesting to try IRL with pencils and paper :-) I'm not a teacher though.

Edit: what do you teach? / What do the students study

You’re right that the same method wouldn’t work with some other subjects and some other types of students.

I did take a mathematics class as an undergraduate, though, in which our teacher—Paul Halmos—had us work together in small groups on problems throughout the semester, with guidance from him only when we got stuck. That could presumably be done online, too.

> I learn as much from them as they do from me.

Sounds like a nice job :-)

Paul Halmos, cool to have had such a person as one's teacher. (Was that in the US? I wonder how he was like, as a teacher and person)

https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...

He seemed a bit formal and intimidating at first, but he turned out to be a warm, considerate person. He and his wife had me and several other students over to their house for dinner a couple of times, and he enjoyed talking with us about whatever youthful nerdish topics we were interested in.

I was about twenty years old then. I just realized that he was in his early sixties, a couple of years younger than I am now.

Also, interesting to read about the Moore method (I found vid the Algolia link). If I'll ever do some teaching stuff, I'll try Halmos' flavor of that method :-)

Online, everything needs to be done differently from in-person:

- Lecture is a waste of time. Pre-made stuff will be higher-quality than anything you can deliver, and save a lot of time

- Grading should be automated as much as possible. Immediate feedback is powerful for students, as is being able to move at their own pace, have targeted remediation, etc.

- Automating stuff leaves waaaaay more time for 1:1 interaction, reviewing student work, etc.

- Audio isn't the only way to interact. Students can use chat, embedded surveys, forums, etc. There are many ways for having interactive engagement impossible in person.

- You should make heavy use of peer teaching: Breakout rooms, structured peer review, etc.

Online during COVID19 crashed-and-burned since people took in-person and tried to run it over Zoom. Good online can be better than in-person.

Oh, and details matter. Everyone should have a headset. Mute is generally bad. You should have good whiteboarding tools. Everyone should have a pen tablet ($40) or copy stand ($100). Etc.

NONE of this happened at most schools during COVID.

That said, my highschool did teach, more forced, me to read vast volumes very quickly. That skill really helped at various levels later. But that can be taught in other ways than sitting in a classroom slowly memorizing Shakespeare.

My first guess is "there are some pushy parents who will want their kid to be in a higher class even when their kid isn't qualified for it". The answer to this is, have a rigorous test and a high standard for skipping a class.

My second guess is "teachers have a general stance of not wanting to make any 'concessions' to parents, otherwise that will attract more pushy parents". (I did once have a math teacher who said "If I let you ... then I'd have 100 parents wanting the same for their kids.") So... keeping a strong negotiating position for the teachers is more important than doing right by individual students.

I think the answer is that, ultimately, the people making decisions about children's education don't have strong incentives to make it go well. If the kid sits in class bored but not causing trouble, that doesn't create a problem for the teacher or the principal; if the kid is enthusiastically learning in the next-level class, that might be nice for that teacher, but on average likely won't make a huge difference—and actually the teacher whose class the kid came from will get a replacement student who might need help, so that teacher may be genuinely disincentivized to recommend the kid's advancement, even aside from the "negotiating position" aspect.

Maybe there's also something of a resentment about them not being needed when they encounter an autodidact? I had problems with an AP chemistry teacher who was angry that I was doing the homework in class instead of listening to the lecture. When I said that I thought textbook had explained the topic quite well, her reaction was extremely negative. The existence of the autodidact sort of threatens the proposition of school itself.

Perhaps my junior and senior high schools, in the south west England in the late '60s, early '70s, were better than yours, or perhaps they just suited me better than yours suited you, but I learned a lot. We also did Shakespeare but I don't remember anyone having to memorize it unless they were actually putting on a play.

I think there's a huge under-served group who are specifically not the majority. Smart kids are generally held back by being shoehorned in with other kids.

Personally, I think we need to figure out how to use online resources best and that the future will most certainly be a mix. But also please entertain that maybe the optimal audience for this program isn't the majority at all, actually. And that doesn't make it less valuable.

When I discovered WGU 1.5 years ago, I did much better simply by being allowed to go at whatever pace I felt like, and taking time off when I felt like it, and I managed to get through school in a fairly short amount of time.

I'm doing online graduate school now, and fortunately my supervisors are somewhat amenable to this style; they simply give me a bunch of recordings of their lectures and all the assignments that I'm expected to do all at once. Some days I don't do anything, other days I'll spend six hours straight watching lectures and doing homework.

Have recently watched my daughter respond quite differently in online learning in group environments. One is Wingchun( a martial art) taught by enthusiastic and outright funny instructors. My little girl loves every minute of class. The other is oddly-enough Montessori class where she feels held back by other kids and their chatter. But she LOVES the IRL Montessori classes.

I feel that this form of online high school wont be for everyone but there is a segment of those for whom this medium suits them best due to combination of circumstance, motivation, and personality. Education is not 'one mode suits all'. I struggled to stay engaged at high school. Tertiary wasnt much better until i discovered extra-mural (distance education), and loved it.

I think this is exactly it.

In person never worked for me. I was bored - and being bored I ended up with “discipline issues.” I would never do my homework, never pay attention in class, was always late because I was talking in between sessions, etc. My teachers weren’t a fan, the administrators weren’t fans, parents were upset.

Every year, starting in 2nd grade, at the end of the school year I was given “token” exercises as a way of getting passing grades so I wouldn’t be held back. I’d knock those out and continue on. At the time I was thankful to the instructors for giving me that opportunity. In retrospect I’m fairly confident I know why that happened.

Every year we were administered state achievement exams. The school was evaluated based on the students performance on these tests. While I was failing every subject, I carried the class (scoring at least 10% higher on the exam than the next highest grade) every year. Holding me back would have raised quite a few questions about their curriculum.

Fast forward to 12th grade. I drop out of in-person school and switch to online. My in-person school refuses to release my records (we owed them money) so I was starting 12th grade with no credits. The online program was self paced. I knocked out 3 years of eduction in 3 months. Then turned 18, walked downtown and took the GED instead of finishing the last year.

Self paced education is huge for certain people. Every once in a while I think of all the kids in the world who could achieve so much with their youth, but instead they are strapped to a chair being tortured 7+ hours a day by well intentioned adults.

Have you ever considered that the digital teaching methods you employed are subpar compared to those a dedicated online only teaching platform might have.

I don't mean to suggest you did not try.

Most kids had no interest in working without supervision forcing them to work. Now that they're back in school, they are continuing to not work and it's a disaster, with teachers and staff quitting or retiring en masse. Again, this is in a school district that people had been trying to get into for years.

It's really not the coursework delivery style, it's about forcing kids to do what they don't want to do. This force just isn't a component of online learning -- it has little to do with content delivery, imo.

The upside is that those kids that actually want to learn are freed up more than ever before and can actually thrive.

Moreover (as I replied to a different comment above) even after a year of being online the students were unable to identify things that made any of their online classes work. At that point they had resigned to their classes sucking. (Not just my class - I asked my students to tell me what worked in any of their online classes. Collectively that's asking about the teaching methods of a few hundred professors.)

If you have information about some methods which work, I would have loved to have them. I would still love to have them! (If you can provide some evidence about whether and how they are known to work that would be better still.)

Were learning materials which cover the required material made available to students?

Were those materials made available on an approachable platform?

Were those materials made available on a platform which encouraged habit formation for students to work on the material?

Are students able to track their progress?

Are students able to confirm their progress and check with where they are supposed to be if the exam dates are fixed?

Is the progress of student embedded into a (meta) narrative which ties together the lecture and provides topical humor (like a lecturer would when cleaning the blackboards)?

Are means in place to ensure passive diffusion of important information if students are stuck?

Are means in place to encourage or enforce formation of student learning groups and cooperation between students?

The traditional lecture hall model has a lot of mechanism which need to be replicated in the digital sphere.

Can you provide links to any?

It may not work for everyone but the important thing is that it may and will work for someone for whom other options are not available or subpar.

I grew up in a small backwater-ish city and school was useless, outside of socializing. By far, the most important thing I did for my education was enrolling into a distance learning physics and technology school managed by one of the nation's top technical universities. You would receive learning materials and exercises several times during the school year by snail mail (yes, I am that old) and send your answers back for grading.

Whatever I am and the life I have started with this.

At least that last factor is directly correlated with online learning.

Also I know from experience that most students in online classes aren’t even actively listening. A good chunk of them end up on YouTube or Instagram.

It might be worth considering whether synchronously sitting in a lecture with 30 other kids isn't, in fact, a good model for learning in any environment, online or otherwise.

Think of a meeting of 30 people in your own office environment and wonder how many people are spacing out at any given moment.

Unless they live in rural settings, it’s likely a very loose correlation.

When I was a kid, I grew up in a neighborhood in a walkable town and my friends and I saw each other everyday, though we went to different schools or were in different grades at the same school.

I later moved to a rural area for high school and only saw people at school due to how far away we all lived from each other.

In the first situation, I suspect online learning would have been great for me since I could do it at home and I’d still have plenty of social interaction without the downside of hauling books between classes, sitting in an uncomfortable classroom environment, and largely not paying attention since there wasn’t any way for the teacher to track my participation in a large class.

I’m not sure how physical education would have worked at that time, but maybe fitness trackers help with that nowadays.

I do suspect that online learning isn’t great for lower income households since they may not have dedicate space set up for learning, and it would be a shame to create yet another class difference in education.

Living in South Africa, we have a huge difference between the best private schools and the worst public schools. It would be nice to know that every child in the world has at least a base level of education available that reduces “quality of your teacher” as a variable.

Obviously this assumes they’re able to learn remotely. I realise that’s not everyone. That doesn’t mean this has no value. I had extra lessons that were part recorded (so I could rewind etc) and tests with a teacher who could help guide through aspects I didn’t understand. Improved three symbols in one year with about a solid week of study over each holiday. Self study worked very well for me, as a supplement.

I’d also argue that online school during the pandemic, like remote work during the pandemic, is not representative of online school during non-pandemic times.

Now this is speculation on my part but the students who may be suffering from bullying (leading to that anxiety and depression) now have more control of their environment and connections. Its easy to ignore someone typing in some chat room (and maybe even mute them entirely) that you can't ignore when they are physically in the same location as you.

And I know being able to control my environment, and fidget or move around, or whatever I want to do allows me to work better than being stuck in some classroom bored and daydreaming. So for certain people there are definitely positive outcomes. Even if they are a minority. And a service like this might be a great asset to those people.

[1] https://sphr.nihr.ac.uk/wp-content/uploads/2020/08/Young-Peo...

The child's age is also significant... I certainly would not have a child under 12 working under the the same expectations as a teenager. Additionally, every child develops at a different rate, so finding that sweet spot is an important part of home schooling. These are all things that the public school mentality finds incredulous.

So I agree with what you are saying, but so many people still think the model for home school should be the same as a public/private school situation, which is where things fall apart. It's obvious to me that children always need to be well socialized and have a variety of teachers and experiences, but that can be achieved if the parent is proactive and willing to spend the required time and money to foster learning over optimizing for succeeding in the public school system.

No disagreement that grades are lower, but is it possible we are in some ways measuring the wrong things?

In a modern age, “remembering information” or “doing homework” don’t seem so applicable to the future.

Being in person can be important, but it’s for mostly social reasons, not the ones you stated.

No disagreement, but we were measuring the same things pre- and post-pandemic. The negative effect on those measured outcomes is very real.

It is possible that if we were measuring the "right things" (whatever those are) we might see no effect. A priori that's a little implausible IMO but I'm willing to entertain the argument!

> In a modern age, “remembering information” or “doing homework” don’t seem so applicable to the future.

Strong disagree on "remembering information is not applicable to the future." I partially concede on "doing homework", but honestly there is no other way to make the students attempt to apply the material they learn, and no way for them to understand what they don't understand except by attempting to apply it!

Homework might suck, but there is a good reason everyone uses it!

I did quite well in the program and on the SAT I took to get into college in lieu of a high school diploma. I did okay in college too, graduating with honors. I really can't see what I lost by basically doing high school alone in my bedroom reading books and taking self-graded exams.

I think it can work with the major caveat that it has to be both voluntary and the student has to bring a genuine desire to learn to the table. Not all students do. I think Khan World School will work very well but for maybe 10% of students leaving 90% worse off. Conversely I think in-person traditional schooling will work well for 90% but make 10% worse off due to stuff like bullying.

It would be unconscionable to withhold something like Khan World School from the people it could benefit. It's okay that it doesn't work for everyone. Neither does the traditional system.

How did everyone else mess it up so bad? I'm just reading this and thinking back to my resolve to use this time as positively as possible, like study and doing my best with savings was the most obvious thing at the time to me

Might be something to do with me living in the third world and adaptation culture being different

Kids. A different set of mental-health challenges. Being worried about losing their job. Being more dependent on their non-coliving family than you are. Being more worried about their family than you are. Having a job that they're not used to doing online. Having a job that's hard to do online. Having a job that couldn't be done online. Being deprived of life-long hobbies and interests that can't be done online or in the home. etc etc.

In my experience from observing my own children who universally despise online classes, Khan Academy has been excellent (with the exception of their Physics curriculum which is a complete mess.) I have no doubt that if I enrolled them in the 'classes' that they would absolutely hate it, especially based on that Daily Seminar image that looks like a Zoom meeting.

With the shift to enrolled classes it seems like they are pivoting towards the Educational Industrial Complex, which is unfortunate. That discourages me from donating to them again. Although I am always happy to pay for quality learning material and workbooks I'm not interested in supporting online classrooms.

In person learning is about learning from your peers as much as it is about learning from your instructors. People are quick to point out that the commonly occurring bad in-person instructor can be compared to online-learning. But, they leave out the fact that online-learning completely sidelines any prospects for peer-learning. I am not even bringing up the role schools play in socialization, physical health and as day-care. No amount of online-anything is going to replace that.

I would also chalk up the students generally being less organized to the stress of the situation, rather than a degeneration of study skills.

In fact, my homeschool friends who did remote video courses (as close as you could get to online at the time) were much better at many subjects than I was.

I don't think its online vs in person. It's online vs nothing or online vs shitty teaching.

The "it's never going to be" part does not necessarily follow from any of the data presented so far. Perhaps we simply need a better understanding of how people learn online, and get better at using technological tools to facilitate learning?

I mean, is there any particular reason to think we've reached the pinnacle of what learning software and online educational platforms can do?

The futuristic dream: kids use fancy tech like AR goggles and haptic tech to manipulate shapes in a collaborative learning game and ultimately learn math in an intuitive way. Think of the best Jypiter notebook you've seen and then take that off the screen and into the real world. Then add an AI assistant that constantly guides you through common questions and make it so that the whole program is continually being refined to make it better and better for each successive generation of kids. Every lesson has tens of millions of dollars poured into it since it will be reused potentially billions of times.

The practical dream: okay we don't have money for any of that, but at least use the internet to break geographic constraints. Have actual math teachers teaching math to various classes around the country/world, have actual music teachers teaching music, immediately direct gifted kids to accelerated classes, have more flexibility with special needs kids (e.g. take classes in a different time zone so that their parents can help), offer a wide selection of second languages by connecting kids to foreign teachers, do virtual exchanges, etc.

The reality: most kids don't even have computers. Actually I was shocked to learn that many kids don't even have chairs and desks at home. This is in Canada by the way. My wife was teaching K-8 online classes to kids lying on their beds, propping up their mom's borrowed iphone on their bellies, trying to not fall asleep as the front camera streams a dimly-lit view of their chin.

So to summarize, we start with the same in-person learning, from the same teachers to the same class, remove all of the blackboards/manipulatives/etc., reduce the child's field of view to a 6" screen streaming a laggy 480p video with horrible sound, delete all friendships by enforcing quarantine both during and after school, and finally conclude that online learning doesn't work!

Many commenters have mentioned that socialisation and its associated obligations are beneficial to learning, and this is what we are reminded of everyday among our students.

Kids learn best when they see other kids learn. And kids learn best when they are able to directly apply what they learn into a project that they own. We enable our own students to publish their own books and to perform in musicals that they produce.

Many online education companies fail because they think that all online education ("edTech") is simply digitising the school. It's not. To succeed, we need to think deeply about what online and physical education does best.

Hit me up if you want to learn more.

[1]: https://storychopsticks.com

Khan Academy has very strong brand recognition.

Many parents pay to have their child tutored. Partially because the quality of math teachers in high school is random (some are very good, others are not). Partially because one on one help is always very useful.

Trying to replicate the in person experience online is going to be a worse experience. The benefits, reduce costs and chances for a better education are there.

While this is obviously a little more difficult in a classroom format, I have seen it done in the 80s completely offline for math where the whole grade of students take a bunch of pretests at the beginning of the year (including the tests for the prior year material) and then are assigned to two week groups or classes for the topics they need. After two weeks, they get a post test. If they haven't passed at 90%, they stay in that topic until they complete it. If they have, they move on to the next topic. It got a little complicated as a few people were in groups that went down to one person after a while, and they were just giving me worksheets and then I would ask if I had a question, but I was able to proceed at more than 2x the normal pace for a few glorious years.

So, maybe doing the Khan Academy thing in the building, where you have a supervisor there to keep kids on task and prevent cheating, as well as ample people to whom one can ask questions, could be a net improvement.

For those who couldn't stand it should stay behind by attending classes physically