Discussing the problem

[Desrtopa]

As per the lesson of Hold Off on Proposing Solutions, this is the time to thoroughly discuss the project we're attempting to carry out.

What lessons do you think it's most important that the game convey? Is there anything important left out in the original post? What tradeoffs do you see us having to face between desired elements of the game? What sounds easy to implement, and what sounds difficult? Is there anything we ought to be particularly mindful of while working on this?

Post any considerations here.

[DaFranker]

I think it is very important that the foundational building blocks be established before the game even gets to teaching the Fundamental Question. Specifically, the player should get to a reasonable point where they accept the assumption that reality exists, the basic intuitive notion that The Simple Truth attempts to convey, and the principle that things can be learned and understood. Otherwise the userbase shrinks to those who already understand this.

I think it's also important that the game sets the expectation very early-on that the game will be fun, and preferably not disappoint. I think Kaj really understands this and already tries to consider it at every stage. And along the lines of what I said above, it should also prime the player's expectations that they will actually learn, and that they will actually understand what they do in the game. The effect of such expectations is well-researched in education.

It might take a lot of experimentation and prototype testing, but we need some way of figuring out what "lessons" to "teach" and in what order, and in what format the lessons should best be presented. I've mentioned Antichamber and The Sims before, and there's a lot of material on the web on making good game tutorials (or reviews of what makes a bad tutorial), but I don't know to what extent the skills this game should "teach" are different from teaching standard gameplay mechanics. It might also be that there's no relevant difference whatsoever, which would be nice and convenient.

The most amazing tutorials are those that last almost half of the game, and where the player barely even registers that there is a tutorial. If the player think "Oh now I have to go through this tutorial sequence before I can get back to the game", then the tutorial fails this test. It's a bit of a lesser version of "Oh, now I have to answer this quiz / solve this puzzle before I can get back to my awesome fun shooter segment".

I can come up with any number of examples and things to look at on any end of the spectrum if someone wants to play some reference material (be it for tutorial quality or some other aspect).

I don't know how much of this is obvious or obscure, but I'll err on the side of assuming that everything I know about game design is obscure and arcane to most until told otherwise ;)

[Kaj Sotala]

I'm currently reading through the Book of Lenses; after I'm done with my first read-through, I'll look at the list of lenses to find the ones that feel like the most important to me, and see if I can usefully analyze the project through them.

On Mon, Feb 18, 2013 at 10:54 PM, DaFranker <dafr...@gmail.com> wrote:

It might take a lot of experimentation and prototype testing, but we need some way of figuring out what "lessons" to "teach" and in what order, and in what format the lessons should best be presented. I've mentioned Antichamber and The Sims before, and there's a lot of material on the web on making good game tutorials (or reviews of what makes a bad tutorial)

Please feel free to link to the materials on this that you feel are the most useful. :-)

And indeed, things should be fun at every stage.

[DaFranker]

Re Lenses:

I just found this companion app (there's probably an iOS version somewhere), which I'll be taking a look at during my lunch break (i.e. in ten minutes) since I don't have access to that list of lenses from here. I'll see if I can find a convenient way to obtain the Book of Lenses. Judging from the description and reviews I probably already know most of the important points, but the gritty details and specifics and technical explanations and habits and so on all seem really interesting.

Thanks for the link, this one seems fun and useful!

[DaFranker]

paper-machine just made a topic looking for games that encourage imagining consequences of actions, and I immediately thought "is this something this game should have?"

I think we need more discussion of which specific instrumental rationality skills should be touched on in the game. It would be too much of a stretch IMO to say "all of them", especially since this game focuses on the Fundamental Question. But some basics like looking for unknown alternatives or choosing what probably leads to best outcomes, even something as obvious as just the fact that even under uncertainty you can still affect the outcome and it does still matter what you do, would, I think, be very useful skills for players to obtain if they can be weaved into the game without forcing it.

On Monday, February 18, 2013 9:19:05 AM UTC-5, Desrtopa wrote:

[Kaj Sotala]

On Tue, Feb 19, 2013 at 7:46 PM, DaFranker <dafr...@gmail.com> wrote:

I'll see if I can find a convenient way to obtain the Book of Lenses.

Not that I'd want to encourage anyone to do anything illegal, but I imagine that a site like *cough*libgen.info*cough* would have it... (Though I bought it myself, and then also ordered their set of physical lens-cards.)

Discussion about the rationality skills we want to focus on would be good, though I suspect that the best way of doing it might be to develop the mechanics and setting somewhat (while keeping in mind the goal of teaching rationality skills) and then keep regularly asking "what skills does the design currently teach, and could it be made to teach more of them?" That said, imagining the consequences of actions does sound like something that would be nice to encourage in this game, and I had a similar thought when I saw paper-machine's topic. Though just making it about the Fundamental Question by enough work by itself.

Risto_Saarelma's rule of thumb that "The scope of the project is about right when what you're planning to do starts to feel too small, simple and narrow" is something that rings true to me, and I've been trying to keep it in mind whenever my mind has gone on a too ambitious track for this game. "How can I make my ideas for this project feel more like they were 'too small, simple and narrow'?". :-)

[Max Hensley]

"
Discussion about the rationality skills we want to focus on would be good, though I suspect that the best way of doing it might be to develop the mechanics and setting somewhat (while keeping in mind the goal of teaching rationality skills) and then keep regularly asking "what skills does the design currently teach, and could it be made to teach more of them?""

This seems backwards to me. We could easily reach a point where we found that the mechanics we were working with were not conducive to adding on other lessons, or that a completely different design would have done a much better job at conveying the messages we want to focus on, and that we'd get better results by scrapping our prior work. We're starting with no product at all, so I think it's better to work out exactly where we're trying to go and plot out the best course to get there, rather than striking out and then trying to work out how to progress from where we've already gone.