Highlighting would be what to use for that, nucleotides are not UI elements, there is a list of available elements in the documentation.
I actually started a tutorial for defining the parts of an RNA, but I got busy with the work on GetSat and never got to finish.
Hi all,
First of all, thanks everyone for their feedback and contributions. I’ve now grouped learning goals together in this document. There’s a key at the top. I divided them into units and have them in a rough order from basic to advanced, while giving the player some variety along the way. There will also be educational interludes between groups of puzzles that will try to fill in some of the essential knowledge base. I hope it’s sensible – it’s not perfect.
I’d like to link these learning goals to specific puzzles. So over the next couple weeks I’ll be regrouping and editing things and adding links to puzzles that address specific goals. I’d love your feedback with that. If you have a puzzle that addresses a goal, add it to the document. If you have other feedback or comments about the progression, please let me know.
Here are some principles I’m working from:
Each puzzle should present something new. In other words, the player won’t be able to do exactly what she did in the last puzzle to solve the current puzzle.
Blend learning when possible (link RNA knowledge to puzzle tactics to effective use of UI).
Introduce UI elements when they are relevant. In other words, introduce pair swapping in a puzzle where pair swapping is useful (where, say, swapping the right pair(s) will solve the puzzle); zooming in and out might be introduced in a larger than average puzzle.
Use contrasting cases. The idea is to use puzzles that look similar, but differ in some critical respect so that players can notice what’s different. A number of player-created tutorial series already do this to great effect.
Show before telling. Some amount of text will be inevitable, but if we can demonstrate something elegantly, through a puzzle, that’s the better option.
The use of scaffolding. The idea here is to enable players to solve puzzles they otherwise wouldn’t by giving a little bit of help. Players have employed a variety of interesting methods of doing this: locked bases, hints in the description, limits in the number of mutations necessary to solve the problem (see the Fun with John series), starting players off with a bad design that they need to fix, etc.
I hope to have a working puzzle progression by the end of August to early September. Then we can iterate and beta-test.
Thanks!
-Ben
Also - as Nando and Rhiju reminded me - the immediate goal is to teach players everything they need to know to participate in the Eterna medicine challenges. So if you see material that doesn’t do that, please let me know.
Thanks,
Ben
Hi all,
This is the latest document I’m working from: https://docs.google.com/document/d/1hMiQcwQ3isGTzufex4Ir7JrKsrFbYtVZLPC0Pe8ChiU/edit?usp=sharing. Still very much a work in progress. The general outline is something like the following:
Basics - bases, bonds, folding, target/native modes,
GC - closing pairs, direction, structure energy, GC orientation, more loop varieties
AU - bulges, sliding, blocking, AU zipper
GU - using GU in stacks, as closing pairs, in bulges.
Non-canonical - advanced boosting/mismatches, catalyzing, energy profiles
Switches - ligand/aptamer, 2-state, 3-state switches
Labs - Eternabot constraint, FMN/MS2 model, RNA in/out model
I like the idea of moving from the strongest base pairs to the weakest (or, perhaps, least subtle to most subtle), which is how the first half of it goes. Very interested in hearing your thoughts.
Thanks,
Ben
Hi Everyone,
Thanks for your patience while we put together an early version of the puzzle progression. This is, basically, the first half of what we have planned (only single-state puzzles). Think of this as an incomplete alpha version.
You can start the walkthrough here: http://nova.eternadev.org/game/puzzle/3389880/?autoload=0
You can see the puzzles in the progression here (might not be in order): http://nova.eternadev.org/web/progression/
Right now, we have a handful of placeholder puzzles (spacebars currently, should be updated this week) and some empty message screens. We don’t have many of the accompanying graphics yet, so you will also see text like “You unlocked the swap tool!” and “(pic needed)”.
We would LOVE feedback from you guys about it. Technical glitches, puzzle suggestions, phrasing suggestions, concepts that you think should be added, suggestions about “flow”, etc.
I am demoing it with a few people who have never played Eterna as well, so we can get add new players’ perspectives to the mix.
Thanks, as always, for your help.
Best,
Ben
There are a lot pf puzzles in the puzzle progression. Are there any comments about the length from initial play testing? In addition, is it possible to make the transition from puzzle to puzzle a smoother and more seamless experience? Or were there no comments about that from play-testers?
We have done no play-testing yet. You guys are seeing what we’re working with right now. I have some people that will play-test over the course of the next few weeks.
The idea behind the progression is to give a player the skills and knowledge they need to participate meaningfully in the labs. So that’s going to require a fair number of puzzles. There will be “breaking points” in the progression, where you get a badge and links to outside information, plus unlocking of various side-quests. When a player returns to the site, the latest puzzle that they haven’t solved will be waiting for them to continue. All told, with switch puzzles and MS2 puzzles included we expect it to be 6-8 hours of play (but that’s just a guess at this point).
Can you talk further about the transition from puzzle to puzzle? Do you mean load times or the “mission accomplished” screens or is there something else that’s making it feel choppy?
What I mean by the transition from puzzle to puzzle is in three parts:
Fanfare, lights and bubbles animation when a puzzle is successfully solved.
Mission accomplished screen, which contains minor details and the ability to review or move on to the next puzzle.
The time spent loading the next puzzle
For the estimated 6-8 hours of play to finish (which sounds rather long as it is), perhaps we should try to minimize some of this?
Gotcha. Is this appreciably different from what normally happens when you complete a puzzle? I’m not opposed to minimizing this interim time, just not sure exactly what part to modify.
This is definitely something I’ll raise with the initial play testers, though.
There’s nothing different with what normally happens when a “normal” puzzle is completed. There is no time spent loading the next puzzle (Brourd’s 3rd bullet point), the corresponding puzzle definition is included in the response payload for the solution submission request, which the applet must wait for anyway. Or differently said, if one wants to speedup that process, it cannot be done at the client level (Flash applet), only server-side reaction time improvements are eventually possible.
In my opinion though, I don’t see that there’s that much time lost in those transitions, for one. Also, I don’t think that it is even remotely possible to bring an absolute beginner totally ignorant about nucleic acids up to speed for Eterna-medicine level puzzles (like the lab opened yesterday) in fewer hours than that. If anything, it’s not enough.
It’s not the individual time spent between puzzles that I consider to be the issue. Indeed, this transition time is typically minuscule for the individual puzzle. However, when you take into account that there are ~108 puzzles within the puzzle progression system, having this repetitious fanfare and completion ritual does add to the total time spent, and may not create the feeling of accomplishment we want for new players going through the system. While I would prefer to wait for the opinions of extensive play-testing before changing anything, it doesn’t feel like the pacing is on point right now.
As for the length, while 6-8 hours may not be enough time to learn about the basics of nucleic acid chemistry, it’s also the amount of time it takes to play many other decent games with far more compelling game-play or narratives. Not to mention, this is just 6-8 hours for the tutorial, and doesn’t represent the depth present in the rest of the game within the lab.
I do tend to agree with Brourd here, would be interesting to get the perspective from new users to see what they think of it.
Would doing some pre-loading of the following puzzle be of any use? So at the same time as you’re working on the current puzzle, there is a small amount of resources dedicated to pulling some of the basic data for the next puzzle. Also, perhaps the ‘puzzle submission’ could be done such that the user could continue doing other things while the submission is taking place, with a little indicator that says it’s still being submitted, then it has been completed.
I honestly wouldn’t know if either of these would actually wind up being beneficial at all, or even be realistic to implement, but figured I’d throw that out there.
Just speaking to the game ‘flow’ perspective, i’m not sure if perhaps having a right banner completion screen like NOVA might be beneficial? The new screen is better, but it might be nice to have that real-estate, and from a design perspective I feel like it might be feel a little less restricted (helping to direct the user more openly).
Many thanks for the feedback.
Re: the mission accomplished ritual. It could be the case that after a while, the bubbles, the mission accomplished screen, the loading, etc. would have less impact on the player (after they’ve seen it a few times), but, as Nando mentioned and I implied, what makes the context different here than with any of the other puzzle solving on the site? Do you guys want to have the power to turn off fanfare and mission accomplished screens with the puzzles you solve? You’ve seen it a lot of times - it’s always there.
Re: length. I still don’t quite understand what the issue is here (and perhaps we’re talking past each other a bit). We’d like new players to reach the depth present in the lab in an efficient (and enjoyable) way. If it takes 2 hours, it takes 2 hours. If it takes 20 hours, it takes 20 hours. It may be that in the 20 hour case you would get more attrition, but we would try to make those 20 hours as interesting and efficient as possible. It would be nice to have the ability to control length of time spent, which we could do if we were building a normal game by simply reducing the complexity of the game or changing how things worked. But here, the end goal - new players contributing to the labs in a meaningful way - cannot be changed.
Would it change your take on either of these issues if, after getting a badge (~6-14 puzzles or so) you are brought back out into the main website with, for example, a side-quest or video unlocked?
Interesting point. I can’t speak to the technical feasibility of doing that, but one thing that has come up in some early play-testing is testers solving the puzzle, but not realizing exactly what they did to solve it (which I’ve experienced as well). The side-view could enable players to see messages and the puzzle they just solved and relate the two.
Oops, of course I misread Brourd’s comment, but the puzzle submission thing can occasionally take a while. Not a huge issue though. Sorry about that.
As for the length, what I’m getting from Brourd isn’t necessarily that there’s an issue with the length, but for the time concentration doing lots of relatively small tasks, repeated ‘fanfare’ could get a bit old.
Thanks for clarifying. So if I’m understanding correctly, it’s more about proportion of puzzle-solving time to fanfare time? (e.g., if you’re doing lots of small tasks it becomes irritating to have delays to those tasks, rather than say, a longer task that you get fanfare for.)
That’s my understanding, but Brourd can confirm. 
Ok, I’ve been thinking and there’s a bunch of stuff I’m going to write up and put in a long-overdue forum topic. I’ve realized the sidebar is just a tiny aspect of what I’m seeing, and is a part of something I see as kind of a bad design. Look out for the new topic *hopefully* later tonight, and ideally all will make sense. 
Sounds good. : )