Dec 312020

So how’s the whole remote learning thing working out for everybody? That’s a rhetorical question, people have no problem talking about how much they hate the whole setup. I get that parents want their kids to go back to school (or at least give them some peace and quiet while they’re trying to work from home), but not only have we been at this for a while, a lot of us are likely to still be at this, likely for the rest of the school year. It’s really frustrating, but now that we’ve moved from the “This is good enough to finish the semester and then next fall we’ll be back to normal” of spring 2020, to “Well, we actually did this for a whole year” (calendar year at least), it’s time to start having a little bit of a retrospective on virtual education, what does work, what doesn’t, how how it needs to change to be useful in the future.

So what about this is working?

What’s happening now isn’t a complete waste. If we’re going to talk about how virtual schooling could work, we should start by discussing what is working now. For starters, school here now starts at 9:00. Before everyone got sent to virtual school, kids had to be in the building by 7:50 AM. Even if you happen to live right next your kid’s school, that sucks. It’s 1 thing to have schools open for parents who have to be at work very first thing in the morning and need to be able to drop their kids off, it’s another to demand that every kid be at school, wide awake, alert and ready to focus long before a lot of us have made it to the office. Keep in mind that means getting your kids up early enough to get over the fact that they had to wake up that early in the morning, get dressed, eat, and make sure everything’s together and ready to go, all in time to leave the house early enough that they can be at school before 7:50 (which means including some extra time to account for traffic). I live in a reasonably small town, and to get the local high school on the other end of town is about 15 minutes assuming no traffic problems on the way, or that I don’t get stuck at a particularly long light. I have to drive past an elementary school as part of that process, so I don’t know how optimistic I should be about that, or if I should leave earlier because instead of driving through town the better route would be drive around it. Odds are anyone working an office job doesn’t need to be in before 9:00, so let’s not force kids to be in school starting class more than 1 hour before we have to be at work.

While we’re on the topic of virtual school having a better schedule, kids are getting an hour for lunch, which is better than the 30 minutes teachers got when kids were in school (and if teachers got 30 minutes for lunch, we can pretty safely assume that’s because it was the same block when their students were eating). Really some of the best benefits of doing everything online is taking advantage of the ability to decompress the entire day, which cuts back on a lot of stress (and not just for the kids).

To pull off the whole “all schools are going all-remote” thing, we had to guarantee everyone at least had a computer and some way of getting online, which led to universal Chromebook distribution with hot spot availability for everyone that needed it. I don’t really care if you like Chromebooks or not – the general infrastructure and preparatory work of getting laptops to all students, along with making sure they had access to wireless hot spots to get online can help motivated students follow through on any topic that interests even if they go back into classrooms. Universal infrastructure distribution like this is the type of work that gives you lots of options in the future beyond just the thing your original did it for, which makes it the type of practice you keep going after the whole covid pandemic passes.

1 such win you get with universal laptop distribution – an end to snow days. People who live in states where it snows regularly may not be aware of this, but in other parts of the country where it doesn’t snow quite as often, we don’t have the street-clearing infrastructure you guys do. Instead, we just shut things down for a day or 2 and let everything melt. Before covid, that included schools, which meant the kids had to make up the days they missed. Now we’ve gotten good enough at least doing temporary lessons virtually that missing days (and making them up later) can be a thing of the past. Instead of cancelling school, just announce that it’s moving online until the roads melt. Teachers can get by for a couple of days and we don’t have to screw with the school calendar to make up the missed time.

This last bit is pretty hyper-specific to my personal situation, but my oldest child’s schedule is “class” (the part where everyone’s in a Zoom meeting) is all morning, with 1 30-minute small group meeting either immediately after class, or shortly after lunch. Then they just have their specials class (PE/music/art/media and technology) at the very end of the day. As stressful as that morning block is, you push through that and it’s all smooth(er) sailing from there. Kids whose small group is right after class just stay on the call for a bit longer, but the whole class gets the chance to take a nice long lunch and plenty of time to work on their homework at a nice, leisurely pace. Obviously, you can’t do this for a whole school, but this is a recurring theme that’s going to come up again later.

But I thought the whole virtual learning thing sucked all around?

Well, it’s certainly not living up to the hype and potential. That makes sense, when you first try to adapt something offline to online, the basic instinct is to just publish the original thing you were doing on the Internet. That means what was everyone sitting in a classroom is now just a Zoom meeting, because the most analogous thing to what everyone had been doing before. It seems like the frustrating part of all of this is that we not only haven’t progressed past this, but there doesn’t seem to be a lot of interest in progressing past this from the people in charge of figuring out how to make virtual education work (which in the public education system, isn’t the teachers and principals).

Generally speaking, the next step of this process would be people starting to ask the questions “What can I do now that I couldn’t do before?” and “What did I have to do when I was offline that I don’t have to do now?” This would be where you really start to see people coming up with new apps, services, and approaches to old problems. It’s likely something people want to do, but a lot of schools don’t have the autonomy to try things out and experiment, not to mention the fact that not every organization (or school) wants to be the one doing the experimenting.

What could virtual schools look like?

So let’s start speculating about potential solutions. To do that, we need to focus on the questions that get us from “just do things exactly the way we’ve always done them, but online” to whatever taking school online whole cloth should actually look like in order to be successful. So what are things we can do online that we can’t do with in-person school? Well, with in-person classes your pacing is generally limited by the average to weakest students in the room. Teachers can’t leave kids who haven’t grokked the material behind, but kids who have mastered it are going to get bored. In high school and middle school, we solve this by placing kids with different academic levels in classes together, and in elementary school that’s done by breaking classes up into small groups. But the reality is that these are all really just attempts to get the average of the group to cover as small and consistent a range as possible. 1 area where the Internet really shines is personalizing content for users at scale – you know, the things powering targeted advertising, and your social network feeds, and basically everything people say is evil about the Internet today. Well, those same technologies could give us the ability to finally offer students individualized lessons and learning programs.

Probably the biggest shift you’d likely see mirrors changes we see in software as companies moved from monoliths to microservices (no Maslow’s Law here). Specifically, with a shift to virtual learning, it’s entirely possible that we’ll see different aspects of teaching broken out. As Cringley pointed out in the blog post I linked above, “While Sal Khan of Khan Academy fame may be a perfect distance teacher, we have no idea how he’d do in an actual classroom.” There’s even a difference asynchronous lessons and synchronous lessons given online – namely synchronous lessons given online offer the ability to ask the presenter questions, similar to a classroom setting. There’s also making the course materials used by teachers when presenting these lessons, and the in-depth question and answer sessions, along with step-by-step walkthroughs, that help students understand material even after the lesson is over.

There’s a little bit of this breakout going on right now, but it’s mostly incidental. There are some sites like the aforementioned Khan Academy that explain concepts, and there are sites like Teachers Pay Teachers for getting lesson resources, and walking through problems and answering questions more in-depth are things we typically associated with tutors. The thing is we’re just not seeing these breakouts work together in tandem to create a powerful system, and I think it’s largely because nobody is trying to form the sorts of relationships between these pieces (and likely more) to actually form that value prospect. Because right now, lesson resources are being built entirely by teachers. Not by people who specialize in building this stuff working with teachers, but entirely by teachers. The problem with that is that while teachers absolutely understand learning and how people learn, they’re not necessarily good at building online tools to help people grasp concepts. That’s not a problem per se – the whole idea is that teachers go to school for years and years to learn how to teach, and to communicate information in a way that it’s retained, not build apps or online tools. In this scenario, teachers should be the product owners defining features, priorities, and in general defining how these online tools should work, working with people who specialize in building things in whatever medium the teachers think best helps reinforce what they’re trying to teach.

And just so we’re clear, this isn’t a knock on teachers, teaching yourself the basics of a whole other career in your spare time to help your other, full-time job is quite frankly insane. My first job was for a small medical records company that focused on selling software to register patients and bill insurance to smaller practices precisely because we want doctors to spend years of schooling to practice medicine, not operate their small business. Teachers doing this stuff so far are putting in a heroic effort, but a lot of what I’ve seen has been an exercise in every dark art that can be performed on Google Slides. It’s an impressive feat for slapping something together to try to keep education going. But teachers should have software developers (or software companies) they can rely on, either for online education or to supplement in-person education, so they can focus on teaching. More importantly, online and in-person education is different, and we need to be baking that into every aspect of the education process.

Embracing asynchronous

1 of the biggest differences between virtual education and physical education is that virtual education has the ability to be asynchronous. That’s a major selling point for every online school, and there’s a powerful reason for that. People who work full time jobs need to be able to learn not just outside of normal working hours, but at times and paces that fit into when they’re able to focus on what they’re trying to learn. Maybe that’s an hour or 2 weeknights, maybe it’s weekends, maybe it’s a random pocket of time here and there. And don’t try to dismiss this by telling me kids don’t have full time jobs – their parents (hopefully) still do, and that means if you want any meaningful parental involvement in kids’ education, you better take that into account. Not everybody can leave their kids with a tutor while they’re at work, or be a homeschool teaching assistant on somebody else’s schedule.

This need to be asynchronous with virtual learners creates an emphasis on independent, self-contained lessons. It may make sense to group these into bigger topics, but the idea should be that you can sit down, do an individual lesson, and be done with that concept when the lesson is over. In other words, a lesson should be like a well-scoped user story – short, self-contained, and easily digestible.

Mix and match interoperable tools

Here’s another thing we can do to make online learning better – interoperability between tools and resources. Instead of making everyone do the exact same sets of activities, people should be able to plug-and-play any resource they want that reinforces the information covered in class. Some kids may need resources that are touch-optimized (like a parent’s tablet), some are on devices like Chromebooks, some are on full powered laptops, and some may be just using a parent’s phone because that’s just what’s available to hand their child. The catch here is that these resources aren’t just different variations of the same activities. Different students learn differently, and not only should learning resources change based on the medium they’re being presented in, but should also change based on the student (people like me that tend to have things “stick” after doing them a few times would get activities that have you do something repeatedly, visual learners would get videos, you get the idea). Having a class’s teacher put together all this content in all of its different forms (task-based learning, extra videos, and so on) is too much to ask, which is another benefit to breaking out education into smaller microservices (for lack of a better term). Content gets presented, and then everyone gets some form of follow-up assignment.

So if kids are using all these different resources, that aren’t being handed to them by their teachers, then we need to answer the question of how we can evaluate students when they’re doing different work? Well, for some subjects it’s easier than others. For instance, I had a chemistry teacher in high school that had 4 versions of a test when he gave it in class (which eliminated cheating concerns, because if you copied a nearby student’s answers you got a zero regardless of whether he saw you or not), but also offered unlimited re-tests. When a few of us mentioned that was an insane amount of work, he told us how easy it was. It turns out science tests are largely mathematics-based (with some chemical reaction balancing thrown in). So for the overwhelming majority of his tests, he just changed question order, which variables he provided from the formula being tested, and the numbers themselves. With that system, he had basically an infinite set of questions he could provide students that all provided the same level of gauging mastery. More importantly, that’s a system that can be scripted.

Not every subject is this lucky (there’s not a lot of calculations being done in music class last I checked), but it’s a start. You can adapt this system for vocabulary – take a set of words mapped to definitions, and give the student either a word or definition, and ask them to match it to its associated pair. With more reliable language processing, we can probably even make those questions open-ended. The point is that we can start focusing on helping take teachers’ visions for great online tools that they just don’t know how to build, and build them. We can work with them to write tools that eliminate busywork like grading, so they can focus on going over results and figuring out how best to teach this so it resonates more strongly with students.

What’s important to keep in mind is the various high-dollar, big-name online app-a-palooza schools are using right now don’t work. I have apps for messages from my sons teachers, Google classrooms for school work (multiple, 1 for his regular classroom and 1 for every other subject), and there’s apparently a whole other app to see grades that to this day I’ve never seen instructions on how to sign into (luckily, my wife teaches at that school and already has a login). More importantly, there’s no interoperability between them. I can get a message in 1 app about an assignment being posted on Google Classroom, but I still need to log into Google Classroom to get the specifics. Same thing with grades, I can get a message in 1 app that his report card is posted, but I still have to log find this site and log in directly. Oh, and that’s just the final grades for his report card, for individual assignments, I have to go back to Google Classroom to review those. So far, the closest thing I’ve seen schools come to app integration is custom Zoom links to the different classes.

Hate Facebook all you want, but the News Feed at least put all the updates Facebook is going to show you in 1 place (that doesn’t excuse the fact that their algorithms automatically reduces content rather than let me manage how many posts from whom I want in my feed). Even when they made Messenger a stand-alone app, I can still see message notifications on their main site, get a summary of my most recent messages, and click 1 link to open Messenger to the exact message I want to read. The same thing can be said about people who still check RSS feeds and read blogs regularly, No matter what topic I’m interested in, if there’s a blog on it I can get all my updates in one place. There’s not a lot of interaction between blogs unless the writers happen to write posts responding to one another, or posting comments on individual blogs, but there’s not a lot of room for much more than that anyways.

With education, on the other hand, there’s a bit more potential. The system that hosts and lets students submit assignments (and thus where they’re manually graded) can post those grades to a feed that the report card grade app can read from so that I can see all my son’s grades in 1 place. Whatever app I’m supposed to use as a “home page” for school related resources can just listen for updates to grades and post a notice that grades have been updated (you can even let parents configure this to get every assignment, or when mid-semester and final grades are posted), along with a link to view all the details. Announcements posted in the messaging app can show a notification on other sites to tell me to go back and read the latest update from school. In fact, we need to get these kinds of integrations just so parents can even have a reasonable “home page” for their kids schoolwork.

Just like there are tradeoffs to everything in software, there are trade-offs to virtual vs. in-person learning, and making an informed decision involves acknowledging them. There are some things that can’t be taught virtually. You don’t learn cutting, gluing or handwriting online (can you tell my most recent exposure to learning is kindergarten?). In fact, any physical skill (like drawing, playing an instrument, or proper workout techniques) isn’t going to lend itself well to asynchronous, online lessons. So no matter what, you’re never going to completely replace synchronous, often in-person, instruction. That’s fine – we’re not trying to completely replace education as we know it, just make virtual education the best it can be. Some skills need to be taught in-person, and we shouldn’t be trying to shoehorn those skills into virtual lessons if we can possibly avoid it.

Getting virtual education right has benefits beyond children. Online guides, tutorials, videos, and articles are how most adults avoid stagnating professionally. Getting “how to do virtual education right” benefits them too, whether it’s working professionals going back for additional degrees (or looking to change careers), people looking to stay current in their field, or just someone who’s curious and wants to learn something new. Let’s ask ourselves what happens if we grouped people by skill and mastery level, instead of age? Maybe it works, maybe it doesn’t, but now we’re in a position to try and maybe discover something new that’s really effective in some situations.

Virtual education is new to a lot of us, and we’re at the very beginning of trying to implement this at a wide scale. We’ve done the instinctive implementation, namely copying everything over to the closest 1-1 analogous “online” thing we could think of. Now is the time to start thinking about how we can completely change the way things are done because education is happening in a new medium. We should be encouraging, supporting, and looking for experimentation. It could be deconstructing education into separate components is the way to go, with different people focusing on the parts of the “education stack” where they fit best. That could be completely off. The only way we’re going to know for sure is if we take advantage of the fact that “just doing it on Zoom” sucks and start trying other stuff.

 Posted by at 8:00 AM