I am really tired of doing 3sheets and 2 helices just for the points.
I fail to see how this is helping the scientific effort.
Must have done that 20 times.
Maybe 3 sheets and 2 helics is the best way and technically helping science progress
What sort of things are you interested in? (It's cool we know you don't like those, but it's also important to keep feedback constructive and help us with what you do like.)
I agree: there are a lot of disincentives to trying complex designs. They're both harder to make than structures in the "hot dog" style (helices in the middle bracketed by sheets take a lot of work to get right) and they tend not to score as well.
The scientists keep encouraging us to try new things but look at the structures posted in the two recent blog posts: at a casual glance they're mostly very similar. Maybe they're interesting to the scientists in ways we don't appreciate or perhaps they'd rather have good-scoring "hot dogs" than bad-scoring exotic structures. In either case it would be nice to know.
So what would encourage folks to try complex designs? :) Is it a matter of score?
You could give out extra global points (BONUS points) to players for interesting but poorly-scoring designs. You could award these after analyzing each puzzle's results. BONUS points would be subjective and awarded at the discretion of the Foldit Team. You could leave unchanged the original RANK, SCORE/BEST SCORE, and POINTS columns for each puzzle (as listed at http://fold.it/portal/user/me and http://fold.it/portal/node/2003150/show_players for example), but add an extra column called BONUS listing any extra global points the player got for a poorly-scoring but interesting design.
When you give a player BONUS points, you should also send that player a Foldit e-mail including a 3D image of the interesting design and listing the design's Foldit score (to 3 digits after the decimal point), sequence, and other key info. Then the player can share this info on the puzzle's wiki page if desired.
Recent design puzzles and filters favor three sheets and two helices (3s-2h) as the highest scoring designs. The designs selected for the wet lab were also 3s-2h. Is there a point in making other designs if the only designs going to the wet lab are also the highest scoring designs? Other than the fun of coming up with a complex design. It appears high scores are a major factor to whether a design will go to the wet lab. Other designs just don't score as high as 3s-2h recently.
New filters basically encourage me to try new designs. With experience with the same filters, I end with same kind of design because of score "only". As a player, the score is my main motivation. Hopefully the score meets Scientific goals.
For example, in the past, we had a resulting 3 helices top scoring (and scientific) solution.
With the new filter, we had to adapt, and now we tend to find a 3s-2h, but still in various alternatives: SSSHH, HHSSS, HSSSH
With longer 100 residues, this might change to 4s-3h, with many alternatives. Moreover, the 3 helices might be on the same face, or alternate 2 on one side, 1 on the other.
I suspect that when we all will converge to always the same best design type, the Scientists will change the filters (only sheets, etc).
It's ok for me. I like the time of exploring new designs for new filters.
We know that the science of things may require working a format beyond what is interesting to citizens. Positive feedback about designs helps to stay engaged. Thanks for doing so lately. We did even have some hot dogs surfing both sides of the board, which I think spvincent was the first to try
Most of us also know that points and good science are not equal, so if we see a puzzle to work some more (like with one of Stomjoh's designs once) that can add interest
Beyond that, I'd say mix it up. And give us some training puzzles for ED. small de novos that would bore the heck out of the 15 or 20 people who already know now to do this. Bruno suggested a string of puzzles that start small with good resolution and increase in complexity. Then everyone could play ED. Reloading a puzzle one just did (e.g. cow eye) into ED is useless as a teaching tool for someone like me who can't see anything in the cloud, especially as I didn't have a good fold to start. It would be fine to exclude points as far as I'm concerned, or kick out anyone with a high solo ED rank
I don't know if there's any one solution as individual motivations vary. A couple of possibilities though:
A 'Complexity' filter could be added to design puzzles (after all, one can never have enough filters). This might be based on the number of secondary structure elements, or perhaps by summing differences between torsion values of successive residues. Or something.
Better feedback as to what designs are useful: Jeff and Skippy make some good points here. And it needn't all be positive: if there's a design type that's not useful let's here about it so we can stop wasting our time.
Player motivation is super difficult to pin down, but threads like this absolutely help. I'm glad to see everyone chiming in.