Jonathan Blow about video games as media and education

It's a long presentation, but I think it worth to watch.

https://www.twitch.tv/videos/678729516

 

I wrote a lot, but I'll try to pare it down as much as possible.

In the end, his point seems to boil down to: games are good at teaching "the spirit" of a field, and thus we should use them and chuck books. I don't agree, ultimately because I don't think that's what education is about.

This is kind of a weird take for me: I did a master's in chemistry and thus I was a lab TA for almost every chemistry course we offered (intro courses for non-STEM and STEM, as well as the final chemistry course in the degree program). In addition, my research and thesis were about cellular automata, and something my research advisor has suggested (we're actually sitting on a paper about it) is using CA in a lab setting to help students better understand various (usually statistical) phenomena.

But ultimately that's not the overall purpose. A student can play with a random walk simulation and see the formation of a Gaussian distribution, or the principle that a single agent over time produces the same results as many agents over a short time (there's a specific name that I've forgotten), but that's not going to teach them the math they need to calculate a distribution, or the equation for a first order decay. They're not really learning anything they couldn't see in a book, or displayed in a video.

In a more general sense...I think it's worthwhile. I think labs are good. It allows students to get hands-on with things that are real, but seem so abstract in a classroom. It's one thing to read about chromatography, it's another to dissolve your M&M in HCl, dot it into a TLC plate, and then watch the different colors go up. Hands on experience is good, but it's not really learning the fundamentals. Learning the fundamentals is what education's about.

A couple things (that I wrote during the video, while the above was written after).

He claims that the fact that video games "react to you" is unique to them, and he's right, but he uses the example of Pac-man as an argument for learning in a way that's unique to games. I'm not convinced of this. If you watched a video with the same behavior, could you not learn the same information? The thing about games is that you spend many, many hours doing things, while we'd typically only watch videos for a fraction of that time. If you spent the same amount of time watching a video as you spent playing a game, would you learn the same things, or the same things to the same things to the same detail?

He claims that writing about Pac-man would require "many pages" but I'm not convinced. It could be described very simply; that's more about being able to write concisely. In my opinion.

He makes an excellent comparison to science and points out that in games one can explore and essentially form their own ideas and validate them. This is good* for something like science, where there are systems you can define and have the player run up against them and learn them. I'm not sure at this point what else he's positing using games-as-learning for, but I see two places where his argued strength doesn't help: 1) when learning non-systemic information like anatomy, or history/anthropology, and 2) when learning things without a strong consensus to their existence. Two seems particularly risky to me. Take psychology and sociology - these fields have competing theories on various phenomena. By presenting one theory in something like a game, you're implicitly suggesting to the player that that theory is predominant or correct (or even if you show all of them, it might make them less accepting of new ones).

He mentions randomness, but this is a weakness, not a strength. Learning the full "state space" can take far, far longer with randomness, while a textbook or video can present it immediately.

Everything he's saying about Braid is really standard to puzzles in general, his game isn't special.

He mentions the Witness. The fundamental thing he's mentioning, learning the rules from the game's response to your action, is good. But the incredible danger here (and with other things, see my asterisk in a previous paragraph) is the player learning the wrong things. If a player is developing the wrong rules, they can get halfway through a series of puzzles and smash face-first into a wall (this was kind of my experience with a couple of puzzles).

His description of someone explaining what they're trying to do in The Witness is really, really bad. Basically confirming my previous paragraph - people giving these very vague, literal explanations of their actions, rather than describing rules. We don't want that when educating people. He outright says it later - "competence without specific facts." That's literally the opposite of what we want with education.

I think it's significant that the games he presents are Zachtronics games, which are pretty niche (and for the record I own several). These aren't going to be enjoyable to the majority of students (in a world where they were incorporated into schooling), so we should be careful looking to them as justifications of games being fun while educational.

His example about achievements is bad. He implies that every variant of them is "the player was able to breathe," but that's hardly the case.

All of his language presenting the educational merit of games is very, very soft. "The spirit of engineering." "I optimized pseudo-assembly, so I can probably optimize real code." "This system feels weird just like that game system felt weird."

 

TLDR. Also you start on the wrong foot. Don't mind he states three times he doesn't think throwing away the current system would be good. So I don't know what did you watch, but it is not the same presentation, that's for sure.

 

He says numerous times that he'd implement games in the place of books very early in school programs.

The TL;DR is the first paragraph.

 

No, he says, it would be great to find out how to employ the games before the heavy books. It is a very important distinction.

Edit for clarity: At least that was my understanding all along and he stated multiple times and he specifically answered this in the Q&A section this way.

 

Make sure to ask him his opinions on Unity, Full-Stack developers and Adobe at some point.

 

Also on Javascript and on the state of the Internet in general.

 

I did not watch the whole video yet but having worked 7 years creating educational video games in collaboration with teachers, neuro scientists, kids, parents etc. I can for sure say that games are a great medium for education. Not every parents or teachers or countries share this point of view and things are moving really slowly, at least from where I come from, but it will eventually.

 

Bookmarking the video to watch later. A significant part of my masters so far has involved following the link between games, engagement and education, so I'm keen to see what he has to say.

Games excel at engagement. A good game is basically an endorphin generating loop. This keeps a player hooked. Games have to do this, because otherwise players stop playing. A game also has to teach players how to play the game. This means that in practice, many games are incredibly effective at teaching the player the principles required to play the game. Many games even manage to engage a player outside of the games, and have players doing independent research on their own to get better at the game.

Contrast this to the traditional education system. Students are frequently bored. They are generally reluctant to do anything other than the bare minimum. And very few students go out of their way to do independent study on topics not directly related to the assessment.

There is a lot of interest, both from practicing teachers and educational theorists, in bringing the two systems together. Figuring out ways to make students engage in chemistry or mathematics or English literacy in the same way they engage in Minecraft or World of Warcraft. So far attempts have been pretty clumsy, but they are getting better. I've been impressed with a lot of the newer pedagogies that implement games directly into the classroom.

We've still got a ways to go before you can get a bachelor of history from playing Civilization and Assassins' Creed. But there is good progress being made in that direction.

 

As someone who's done a bunch of both, there's a fundamental difference which applies before we get to what methods are used to engage and teach. In one case people are doing an activity because they want to, in the other they're doing it because they have to.

If you look at students studying things that they want to study then you often find that they do a lot of the stuff you describe with people playing games. They're not in a rush to leave classrooms, they do their own research and learning (and often know some of what's being taught before class), and so on. But when someone is doing something because other people tell them they have to it's different. They stop as soon as they can to do their own thing instead. This, I suspect, is where the whole "follow your passion" thing comes from... but it only works if your passion happens to generate value for other people.

Another consideration I suspect is closely related is the feedback loop between action and benefit. In a well designed game there are nicely balanced heirarchies of feedback, where you're getting small reinforcements constantly, with longer periods of anticipation building to larger rewards as you work your way up the heirarchy. Traditional education systems do a really poor job of this.

I remember in high school math class where the reward for doing well in one topic was getting assigned a harder math topic next year. From the curriculum designer's perspective that probably makes perfect sense - a) filtering kids by their ability gets the right kids in the right class, and b) the best kids need those topics to qualify for the hader university courses. Do you think they stopped to consider how it seemed from the kids' perspective? If I like math, no worries. If I don't, you're rewarding my effort by making next year harder for me.

 

Excellent points! What I think games can do, is provide tangible usecases for abstract knowledge that would otherwise seem very pointless for kids. For example I'd have been a lot more hyped for vector math in school if we had applied that knowledge to games-related 3D graphics stuff. Or if there had been a really cool game that is engaging enough on its own and offers tangible benefits to those that e.g. speak french, I might have been slightly more interested in french. The way it actually went, french was just a subject I had to pick because I wanted the other choice even less, so I expended minimal effort and was glad when I could ditch french again for good. English was the opposite, it always had super tangible benefits for me to learn it, from watching cartoons I couldn't otherwise understand on sattelite TV to opening up new gaming options to me etc..

 

Don't games work the same way? I don't have any specific quotes from developers to prove this, but it seems typical for the difficulty of an RPG, a hack n slash, a third person shooter, to increase over the course of the game, regardless of your previous performance (whether you die 0 or 50 times in a previous level). What are games doing differently here?

I think it's just what you say earlier - that most playing a game want to, and thus will put in the effort to improve. It might be worthwhile to compare the experience of someone who's not a fan of or good at a genre of game and the experience of a student who doesn't enjoy a subject.

 

That's definitely a major one. I think there are others. For starters, video games don't come with the same type or amount of time pressure. Nobody's telling you that you're spoiling your life if you don't pass Level 20 on the 11th of October, but that's exactly what many school systems tell us, sometimes explicitly, when it comes to assessments or exams. When I've had enough of playing my video games I can walk away. If I've had a gutful of studying but the exam is in two days and I'm not confident yet... keep on gringing, buddy!

Also note that my complaint about the balancing of math class wasn't the harder math in and of itself, it was that you're "making next year harder for me". If the math is harder that requires more study time, which (for me) had to be balanced against 4 or 5 other topics, and for plenty of other people also had to be juggled with challenging home lives and/or part time jobs.

 

I definitely agree that they can do that in many cases. As for whether they're the best way to do it in all of the different ways that matter... I'm open, but not convinced. To clarify, I think they're a great idea in some use cases. I'm not at all convinced that there should be any deliberate push to specifically increase their uptake.

This is in part informed by having made a bunch of educational and/or training games/sims. The barriers are certainly getting lower, but those things aren't cheap to make (or maintain), you need to be really careful with the design, there's a whole bunch of IT support infrastructure which needs to be in place, and so on. It doesn't necessarily make a teacher's workload easier, and quite possibly even increases it.

I do think that electronic games, like any other form of media, are definitely a great tool for teachers and curriculum designers to have in their toolbelts. I don't think they're a panacea, even though I used to make and sell them for a living.

(To be clear, I haven't watched the video, so I'm not arguing for or against anything in it in particular.)

Hypothetically speaking, then, how would you design that into a video game in order to teach it?

Then the follow up question is: is that solution the best way you can think of to teach it? Because if we're shaking things up we shouldn't stop at "functional" or even "better than what we already have". We should look at all of the alternatives and identify what's holistically "best". (Note that "best" != "perfect".)

 

I haven't watched the video either but I agree with most of what you said. I don't think schools should push for games directly, that probably wouldn't work. But maybe there should be some incentive for indie devs to make games can boost engagement with class topics. Niche games that require real-world-knowledge are financially risky without funding, and with the ludicrous sums that governments waste on complete bullshit, you could easily fund dozens of great games (in their own right) that incentivize learning (with books etc. - outside of the game) in clever ways.

For vector math I think the obvious application is like I said, working on 3D graphics problems, ideally interlinked with an actual programming course. I remember our vector math homework to often be very abstract "calculate the point at which vector A hits plane B and the reflection vector C" etc.. At that time I already had interest in simple application programming and 3D graphics software, but didn't realize that this vector stuff would be really useful for games-programming. I don't think turning vector math learning into a literal game is the way to go, but if I was forced to come up with a game idea for it, it'd be something like the the zachtronic games where you're solving "puzzles" that are embedded into a narrative and contextualize the work and progress with something less abstract.

Also I was really frustrated after school that they had forced us through so much stuff that most of us would never need again, but at the same time explained nothing about our overly complicated tax system (in Germany). I think teaching that stuff earlier and teaching the relavant math aspects in context of a real world application that almost all students will face later in live would already be a huge improvement imho, even if it doesn't make it more "fun" necessarily.

Or a much simpler example: those "free" smartphones people get with their mobile plans... I have a strong feeling that more than 50% of the broader population are having trouble calculating whether that monthly payment is more expensive or not compared to buying the phone separately and picking a mobile plan that isn't bundled with a phone.

 

But that's not a game. It also requires a whole bunch of background knowledge in a chicken-or-egg style cycle.

Yeah... I was thinking about this before. A way to make it immediately less abstract could be to start the topic by getting students in a computer lab playing with something like Unity where the numbers are on display. This is a direct practical application, and now the math can be taught in that context, and easily transferred to other contexts because it's nice and concrete in their heads.

Mileage will vary from student to student, but that's true with any approach.

 

Worth pointing out - Blow directly addresses "educational" games and says that they're the wrong approach, in that they "use book methods [of teaching] in a non-book medium." He's talking about more standard "systemic" games (like Zachtronics stuff). As I said before I disagree with him overall in that I don't think what he's arguing for is really education and more motivation, but he does talk about those "games."

This is less about games and more about something being useful outside of the classroom.

At the school I went to there was a guy who hated math. Struggled with Calculus, and he figured he wouldn't ever need it because he wanted to go into construction.

The school administrator took him out to talk to some folks who do that, and asked them about it. They said they used math a lot. Following that, the guy in school didn't have any other motivation problems. He'd seen the real-world application.

Good point about the time pressure.

The school I attended used a pseudo-homeschooling method, in that you had these little booklets you had to complete on your own. You could ask a teacher for help explaining something, but for the most part you were doing the work yourself. In some ways this was probably a drawback - a teacher has way more flexibility to tailor a lesson to the class. In other ways it was a benefit: you were generally "expected" to complete some amount of work in a time period, but there's really nothing preventing you from going faster or slower than everyone else. Some days you did 1 page, other days you did 5 pages.

Our system had other seemingly-unrelated issues that make any direct comparisons with the typical system unwise, but having that freedom of pacing would definitely change things.

 

The training stuff I was involved with almost always had to slot into an existing system. That has a huge impact on the flexibility and available approaches in your own design.

 

The games before the heavy books is a natural part of learning. Children need to learn things you can't find in programs, including how books work, reading. writing, being intolerant to other kids, learning what that means and so on.

No disrespect to Blow, but this idea of interactive entertainment, it's called a teacher and that teacher reacts.

Replacing that with software, sure if you want - Tesla will probably have a robot teacher in a few years.

You want to get kids away from screens for the formative years. All big tech agree and keep their own kids far from devices until they're old enough to manage them effectively.

That's a real psychological danger, relying on a screen when your brain is still forming.

 

get kids outdoors and away from screens as much as possible. Development isn't about learning facts and abstract things. Let them develop as normal human beings first before entering the matrix.

But... I think games can be a great introduction to a topic simply because they are familiar and fun. Not to toot epic's horn too much here in enemy territory but they are using an educational version of fortnite to introduce topics like math and physics to children. If that had existed when I was in school I might have seen the value in math. As it was for me, I didn't have any need to do more than add and subtract before I was 30 years old.

i think we have to keep in mind, schooling isn't really about education. It's just day care so the parents can serve our corporate overlords. If you want kids to learn, all you got to do is entice them a little then leave them alone. They gonna go a lot further on their own than anybody could drag them kicking and screaming.

 

School barely teaches how to find stuff. You learn this quite late in your life, which is around university time. All before then is just memorising, in often most boring way.
Technology should be used to enhance learning skills and ability to search for information in a fun way when feasible. This way young students gets more focused and engaged. Too many clever students, to happen be successful later in their life, is neglected by generic school teaching methods. Generic methods also does not work for people with disabilities. Methods are ancient and need be modernised. These ignores completely risk of modern world. That including digital world. I don't say everything need be on digital screen, but none student should be techno-phobic.

 

The physical interactions with other kids, adults and more are the key parts of school, not so much what you learn at that point. So much of human brain development is physical, not mental, which leads to better reasoning and understanding. There's a lot of studies on this stuff if anyone's interested in going for it.

But can't assume you only benefit from textbooks at school. Can it be improved? I would say definitely, including changing what schools even are - bursts of mental and physical challenges. I guess for some it would be hell, but it's what they need.

Totally agree there are improvements to be made.

 

IMO, its generally not learned that well at all by then, Im afraid. And over the past 20 years, I'd say that's been getting worse. The first-resort response to a problem is asking someone else to solve it, and the notion of checking the documentation or searching online first doesn't even sink in when you demonstrate it.

 

I recently heard an interesting video about how the currently popular games like COD or Fortnite lack the "freeplay" component of more open ended sandbox games like Runescape, where kids needed to engage at least in parts of the behaviours that the traditional reallife freeplay is supposed to teach them.
I'm really lucky I'm old enough to still have experienced this "go play outside" stuff, and I worry for those that didn't.

With the pandemic lockdowns I've heard things like "My daughter has been playing Minecraft with her friends for 8 hours today, and I've truely run out of arguments to say 'no' to that."

So they're like boomers in that regard? My parents can't get used to googling things either.

I think I got lucky that during dial up internet days when I was still in school I had a sidejob where I was "broken download administrator" and had to google replacements for broken offsite download links for a gaming website. So I spent hours just googling stuff, and that was good practice for a skill that is still relevant today.

 

Actually, that's a double edged sword.

Heavily relying on technology makes a human into a cyborg. If you're relying on external tech for memory storage (reminders, etc), then your own memorization facilities are not engaged and will not develop. Likewise, if you are really skilled at googling things, then your own facilities you'd use to go through a library catalog do not develop.

Basically, when was the last time you met someone who effortlessly could precisely recite multiple books for fun? IIRC that was not something uncommon, say.... 50 years ago.

There were also several alarms about "clip thinking" which surfaced in last two decades and is result on tech advancement.

The problem mentioned by @sxa where the person asks the hivemind for advice instead of looking it up on their own also exists.

 

not necessarily...

"I think we will always be remembered as the first generation to experience the internet. I was born in 2000 and I always find it so amazing – like, in 1998, what did they do?"

<eyeroll> and <cough>

 

Funny saying about all leaterature, personally I know kid, which was able recognise most of letters of alphabet before it could properly speak. It was playing leapfrog device.

He was learning letters and writing them with pen, as it was made as a game. Had follow line of letters, to complete task, backed with voice input. Along with reading traditional kids book and other more fun excersise, there is place for everything combined.

Devices with giroscope and accelerometers can help with motor coordinations. Specially for both for mobile and consoles. Obviously always better is to do physically stuff. But consider things, these may not be possible for everyone for one or other reasons.

Talking about Google, or other search engines, how this differ from looking in home base encyclopedia? I got 7 havy books alphabetically ordered, of wording explanations. But I don't use it anymore past 2 decades, since I can type digotally for an answer anywhere and within few moments. Sure let teach kids how to use dictionary, or class book encyclopedia, but no need wasting time on it, when can spend time on stuff more productive.

Physics, geography, biology, history, literature and more, has more engaging content than ever, with digital approach.
If school can afford physical props, that is grat and first hand thing to go. But let face it, we can not expect that in every school. Or what about showing 3 min interactive examples, of use cases for given subject. Much better than just tell arbitrary school words, to explain what thing can be applied in real life.

Show kids interactively, where pitogoras rules can apply. Let them play with Tales theorm. Then easier to understand, how things works, scoring better on exams. Show visualisations of geological changes. Let them see interactive maps of historical events.

Regarding literature, at least in England I am aware, in schools students can choose their own books to read, than be forced to read certain titles. I don't know if that is for all schools however. Sure some are mandatory. But that is much better, than when I was forced to read some books, which I never found interest with. Had no option for school time. But was able read Lord of The Ring physical books until 3 am every day, where I had wake up before 7am to shool. That how I was engaged in reading. And mind, I could spend that time at computer instead.

Hence in my experience, it is about giving engaging tools.

Paper is good, but also there wasn't other alternatives 30 + years ago. If showing young people how to search for interesting subjects, maybe will spend less on TickTocks. It is all about interesting engagement, not making things boring. Hence better understanding of the topics.

 

If they unironically say stuff like that, this is shocking.

Being able to "get through" boring things, is an underrated skill. The "making topics more engaging" needs to be balanced against that, or else you'll end up with students that are good at the things they care about, but unable to read and understand something excruciatingly boring like tax laws, because they will fall asleep or otherwise zone out when trying to absorb information that isn't available in a "fun" way.
One of my main gripes with school always was that the boring S*** they force you through often has so little realworld application (I don't think I needed integrals even once in my life after school), while totally leaving out things that so many of us will need later (like doing your taxes etc.).

 

the issue with booger picking kids jumping straight to "ask the hive" isn't necessarily an issue of motivation or laziness. It's the most efficient thing to do. All animals survive by finding the path of least resistance.

If somebody else already figured out a problem, it does team no good to have somebody else solve it again. I guess the presumption is, "I had to do some work so why should you piggy back off me?" and the further presumption is, "if I help you, you won't learn to help yourself."

In my experience this usually isn't true. Everybody wants to help and be useful. If you just offer as much help as you can whenever you can and keep judgement to yourself, most people find the best way to be useful in their own time. Team gets stronger faster this way, versus everybody assuming the other is lazy and keeping tabs on who did more work, etc.

So when it's time to remind the knuckleheads to use google, the reason isn't because "we all use it and you should too," it is because it is simply faster way to find more thorough answer than asking a single individual.

As for memorization instead of knowing how to research, I can't see the practical benefit memorization has over knowing how to access the internets memory faster. Younger people probably are more adept at figuring out how to play the game they've been presented with.

 

I'm of opinion that gadgets are strictly not for very young children, because they royally screw up ability to socialize. Plenty of information about that too.

Google is opposite of encyclopedia, because information it returns is unordered, unstructured and (the worst thing) is tailored just for you. Basically, if you can google that does not mean you can work with manuals. If you can work through an encyclopedia, then you can work with manuals. In essence google sort of has devolved into an "answering machine", which is not very different from asking for community help instead of trying to solve a problem on your own.

On other hand there's another trend where the forums are significantly less popular, and preferred means of communication do not leave any digital footprint at all.

Making engaging content sounds like a sure fire way to raise a human that can't focus on mundane work because it isn't engaging and interesting enough. The purpose of education is to educate and not entertain. Trying to entertain at the same time can dilute and lower quality of the material. If a new generation requires engagement while previous one could do without, I'd say there's a problem at hand.

The point of forcing people to read specific things is to make them step out of their comfort zone and get acquainted with something they wouldn't try on their own. That widens one's horizons. If they're only allowed to read things they like, then they will never learn something new, and that something new includes other things they would've liked.

 

They want the world to forget the irc and icq-times...

That's why it's not about not using them, but being selective what's going in your own memory and what's remain in external search. Knowing the thing doesn't always have comparative advantage, knowing where to find often has.