We could/should make better estimated shapes.

  • 2
  • Idea
  • Updated 8 years ago
Looking at the estimated shapes in the lab results, for the switch lab, the unswitched estimates look very grey and as a result the shapes seem malformed, perhaps a player might try to guess the grey values and could assign them shape notation - trying to guess the actual result (which is presumably a centroid of some sort of the full base sequence) and from the shape data we have the knowledge that a certain set of bases actually paired, while a certain set of other bases actually didn't and a certain set are unknown but could perhaps be worked out, giving a better estimated shape?? It almost makes a second game out of the lab results.
Photo of Edward Lane

Edward Lane

  • 139 Posts
  • 8 Reply Likes

Posted 8 years ago

  • 2
Photo of Adrien Treuille

Adrien Treuille, Alum

  • 243 Posts
  • 33 Reply Likes
Hmm... That's a really cool idea. The key problem (with all these awesome ideas) is that it's so very hard to create a whole interface and then wrap the whole thing in a game and then make sure that players who contribute get sufficient social capital, etc. If anyone case solve that one, then we could do all these cool ideas really easily.
Photo of merryskies

merryskies

  • 40 Posts
  • 3 Reply Likes
Hi Edward and Adrien,

Regarding your comment “then wrap the whole thing in a game and then make sure that players who contribute get sufficient social capital”, it turns out that not every aspect of Eterna has to be in the form of a game, because players are already enthusiastically participating for no points. Examples would be designing Player Puzzles and participating in Player Projects .

Regards,
Merryskies
Photo of merryskies

merryskies

  • 40 Posts
  • 3 Reply Likes
Hi Edward, Adrien, and Jee,

It would be awesome to someday allow players to postulate alternative shapes from interpreting the SHAPE or other lab data.

It does not have to start as a whole new game, it could just be an additional tool. It could use the current Player Puzzles model/interface. Players create the shape that they postulate was formed in the lab, and then fill in the sequence. Then players post it. No filters, energy calculations, or points are needed.

BTW, players have already been doing this in player puzzles.

Thanks for Eterna, it is terrific,
Merryskies
Photo of Edward Lane

Edward Lane

  • 139 Posts
  • 8 Reply Likes
Short term it doesn't need point rewards, the social capital can be earned by playing the proposed game - not directly but by gaining a better understanding of the lab results and thus a better chance of getting a good solution in future rounds/labs. Long term there might be a way of giving 'points' or whatever but I (and presumably a proportion of players) don't really have an interest in those points as such. Or just stick contributors names in the credits for a paper relating to 'better shape estimation' or something :)