I was originally thinking about a tutorial where you actually build a game, but playing one is an interesting idea too. I threw this together over the last couple nights. You might like it if you're into badly written adventure retro games with a D&D fan fiction immersive fantasy world.
I was just having some fun. It needs a lot of polish, more in-depth tutorials and maybe even a genre change... but possibly, maybe, it will inspire some further ideas. ¯\_(ツ)_/¯
Try entering wrong answers and turning on your audio to get the full, uhm, effect.
You can type "cheat" to skip through most parts, or "fight" to go directly to the battle scene. Also, the audio works better on Firefox. There appears to be some random syncing issues occurring on Chrome.
Uh oh, I died... (I should probably be more careful, this project has a pretty low bus mummy factor.)
Very interesting start! This sets a standard and some styling for the tutorial--and it's the first we've seen...so I don't have any reservations about saying it is the very best we've seen!
The most important thing in my view is the kind of "wiki" mechanic, that things can get better by someone taking a few minutes to make a small improvement...and that these improvements add up. So if I could propose a "next step" it would be that we figure out how to make this tutorial pattern easier, to let people start working through their own narratives.
I thought that some kind of way that a Trello board becomes executable would be neat, and what you might be calling a "level" would be a column on the board. Trello is great for rearranging and thinking. It's not so good with versioning, so we'd need some kind of way of backing up the board on an at-least-daily basis and getting cards back. But I'd rather leverage a pre-existing UI of that type vs. be stuck trying to build one right now in Rebol!
Yea, for sure. I ran into some limitations with the print function where I couldn't get the HTML layout I wanted in some parts or highlight text in certain ways. There should also be some standard functions to do things like ask a question and verify the input matches a specific answer or is one of a series of correct answers. Also, some way to "do" certain types of input and be able to capture syntax errors and report back the issue in a beginner-friendly type of way.
I am not particularly stuck on this implementation, but it is a working prototype that we can use to experiment / test new design changes with and see if it makes creating a tutorial easier. I burned through writing the code, and so there is a lot of ugly hacks and un-optimized things going on. I think "next steps" is I'll start re-factoring things and through this process begin submitting PRs for standard functionality we could implement.
I also saw the comments you submitted on my branch. So, thank you for that. I haven't had a chance to look at those in-depth yet, but will do in the following days.
Going a little further with some previous comments...I wonder, @BrianOtto, if the first tutorial to cover should be PARSE-focused.
Despite there being some lingering questions, I was using PARSE for a task this week and feeling pretty pleased with it. I even finally understood why WHILE exists...once the ANY and SOME constructs were fixed to be what I think they were intended to do!
As I've mentioned before, if you are using PARSE it starts to motivate learning the quirkiness of the rest of the language. Whether you like the imperative portion of the language at first isn't that important if you like PARSE... it's what you have. It may grow on you to where you start seeing and wanting to imagine the potential (custom function generators, dialects, etc.) You have to understand it's dressed up as a "normal" language, but not one...and I could certainly see why many people would think it just weird if they don't get the trick.
And in terms of game goals, PARSE can have a clear one to start with. Get to the end of the input. Higher levels can go on from there.
Try entering wrong answers and turning on your audio to get the full, uhm, effect.
(I should probably be more careful, this project has a pretty low