Ok I'm going to say it out loud. design puzzle play

Case number:699969-990190
Topic:General
Opened by:spmm
Status:Closed
Type:Suggestion
Opened on:Thursday, August 18, 2011 - 04:40
Last modified:Sunday, August 28, 2011 - 10:23

In the latest Design Puzzles (because they are so rigid) if one person in a very large group (with active coders) gets to a top ten rank then that solution is very quickly copied using a script, then a script is written which resets the mutables and is then shared to the team. The position can then readily be copied by loading the solution as a guide. This is a fairly recent development.

I don't really have a big problem with that happening in the evolver competition but in the soloist competition it means that you are really just getting a large number of copies of the same solution in the top places; and other teams and soloists who have high scoring solutions are forced down the ranks just by the sheer weight of numbers of players in the large group who have used the fairly effortlessly copied solution.

I'm not saying that all players in the group follow that path but with 150 points up for grabs it is a temptation. Other teams also share and copy each other's solutions but the groups are much smaller so do not have such a distorting effect on the results.

Obviously it is up to each player to get the best score they can, but this is not even a remotely level playing field.
Perhaps three or four solutions per group and one each for those without groups would be a bit fairer?
spmm

(Thu, 08/18/2011 - 04:40  |  56 comments)


Joined: 06/17/2010

Yes, this closes flu puzlle to qttn play at some point. But it is possible to get better score by other player/s than a creator of original solution.

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

I also have a big problem with that type of thing happening in Evolver play. Puzzle 445b is a perfect example.

The 2 point evolve requirement (which translates to anything > 1.0) was put into place to try to control that type of thing, but it is very obviously a total failure.

Joined: 04/19/2009

@spmm - I'm a little perplexed at this analysis...

In 425, there were 3 different teams in the top 5 - and one soloist. In 428, both foldeRNA & AD provided their AAs for global use (it turned out that position was everything on that one). In 439, in the top five were soloists #1, 4 & 5 - hardly surprising that they will occasionally show up in the top 5.

If you were put off by team AD dominating the top at the beginning of 448, that had nothing at all to do with sharing AA positions - we were trying a new technique to start a design puzzle, and each person started by using that with their own personal preferences. Twenty four hours later, the 9 of us in the top 21 atm have still not shared AAs.

I'm not sure which team or teams you have an issue with - but unfortunately & usually, when someone is complaining about large teams, AD is usually what they are complaining about. If this was about AD, please keep one other thing in mind - at the moment, we have only one player in top ten - at #10. The Contenders have 4 in the top ten - VC has 2 (and Madde has returned) - and foldeRNA has 2.

So please realize that our large team has been able to hold our #2 position by just that - being a large team. We work together without copying or banding to evos, but do share information. It can be any one of our members that end up having our top solution for any given puzzle - we do not depend on any one, two or three soloists. We are usually evoing more than one soloist solution at any given time.

@Brick - your team is very small, so it's understandable that you have problems with evolving.

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

I suspect Evolver really just needs to go away. It's redundant with the Team scores, and really, what is it measuring? It just allows situations to develop where a single good solo allows a team to totally take over the evolver scoreboard with very minimal contribution.

The top evolver on any puzzle didn't do much except add a few points to someone elses work, while the guy in 20th place probably did more to improve a solution.

Let's just have each team work toward their top team score.

Joined: 04/19/2009

@ Brick - if that is the way that you & your team approach evolving, then I can see why you don't really get the dynamics of evolving, and the potential value to the devs.

The best example that I can give you is to look at the results for puzzle 420 - although Team AD's highest soloist was only in 11th place, the team as a whole came in second, due to the work of its many evolvers. That solution was worked on by a lot of people to improve the solution dramatically beyond the solo starting point. It would seem that the devs are looking for that in evolving (this is what they have said), and I would hope that all teams would approach evolving that way - we certainly do.

Any team & the members of that team can choose to participate in evolving or not. If a team wishes to become one of the top teams, then evolving takes a lot of hard work and/or must have a decent number of soloists in the top ten.

Perhaps your team needs to become proactive to become better at evolving and placing higher. Recruit more and better folders. If you don't want to do that, then you can choose not to evolve. And if your team chooses not to participate in evolving, then you will no longer need to notice it.

In the meantime, I imagine that the top folders in the Contenders, VC and foldeRNA will continue to do their excellent solo work, and that their teams will evolve most of those - and AD will continue to look at every solo with promise in the hopes that one or more of our members will be able to significantly improve one or more of our solos.

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

The end result is the same - the team gets a high score as a result of evolving. No need to make believe that every step along the way is so important that it needs to have points awarded. Why do you think that a 1 point evolve is better thatn someone who does a 50 or 100 pt evolve, but the resulting score is lower because of a lower starting point? Did that person do more work? No. They just happened to be on a team that works oh-so-hard to game the evolver scoreboard.

Maybe you're one of those soccer moms that feels every little kid on the filed is a winner, when in real life, there is only one winning team, and the others are losers. It's best to let them in on that early rather than lead them on to think every one is a winner all the time.

Joined: 04/19/2009

@ Brick - um, yes - "team gets a high score as the result of evolving". That is what evolving is - how long have you been playing?

I have never said that a one point evo is better than a 50 or a 100 point evo from a much lower score - I don't run this place & have nothing to do with the devs game rules - perhaps you need to have a heart to heart with the devs.

You seem to be in this just for the points - again, if that's what you want - better team placement - then go out & recruit more & better folders.

Or become a top ten folder yourself if it's a higher number you want next to your name.

*rummages in soccer mom's handbag, finds lollipop, hands to Brick for consolation prize*

:-)

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

Don't try to tell everyone you're just here for the science. There were a few of those in BOINC as well, but they all shut up real quick when I would challenge them to play without getting credit.

Of course we're in it for the score! Why do you think there is so much emphasis on it? Would you have a team if there were no scores, and everyone just folded "for the science"? Hardly.

Joined: 06/17/2010

Team score is only from best solution - team ranking.
Player evolver score is from his best evo score - player ranking.
Even if all of us made same very close solution and it will NOT be best one of players in team we will not get big credits as team. If our solution is best, why we (as players) should not get good evo score for that?
It is always team effort. Sometimes small pts bade by one user leads to huge jump made by other player. That how it works.
I see no point to change it, or talk about that is some kind ow evil or cheating. It is just how game is made.
Also if I contribute in team, why should I NOT use team knowedge to pump my solo score?

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

Conversely, if a team already has the high score for the puzzle, what good does it do anybody to have 15 or 20 team members all do the same evo to pack the top of the Evolver scoreboard? There is no new science there, it's merely repeating what has already been done.

The *only* possible reason for that is to push other people out of the top scoring range. That is the evil part. It's selfish, arrogant and unsportmanlike, and it shouldn't be allowed.

Joined: 06/17/2010

Wiki: "Teamwork is work performed by a team towards a common goal."
Well... because game rewarding players by points, score, and ranking our goal is make more points, get better score and rank. This is POINT if this GAME.
Very small number of players are real life scientist and can tell why this position is better than another. Rest of us have one gage: score.
I`ll be VERY happy if my worst scoring solution helps cure anything, but previous CASP results shows that in 99% better score = better solution -> more gain to SCIENCE.
We work on very SMALL objects, so even 1pt difference can be BIG difference.
IMO more better solutions (even close one to other) is better. But maybe s1 from foldit team will tell us :)

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

Rav3n_pl - I'm not saying TEAM scoring is bad, in fact I'm not even discussing TEAM scores.

This is about abuse of the Evolver scores, which really have NO USE and should be removed from the game.

Joined: 05/21/2010

Personally, I get great satisfaction from evolving my teammate's solutions. Some of them are so tight that it takes me a lot of effort to find those 2 points, and I learn a lot about structure from it. Evolution is taking a starting point (like wolf genetics) and branching off from that point (dogs in all their variety). If you're measuring success by something like compactness, you can start with a Great Dane, or you can start with a chihuahua. Maybe you can evolve something more compact from the Great Dane, but my bets are on working with the chihuahua. If the chihuahua turns out to be a dead end, and the Great Dane yields a more compact canine in the long run, it will show. The proof is in the poodle, I mean pudding.

I'm not in it for the science....I'm in it for the fun. . . .my teammates are FUN. Friendly, helpful, and ingenious. Sure, make it more difficult for us to score and we'll simply rally round and play by whatever rules. But for me it is NOT the score.

I now return you to your regularly scheduled protein.

Tyg

Joined: 04/19/2009

@ Brick - Team score IS highest evolver score, for goodness sakes... Is there something about that, that you cannot understand?

I think that your perception is colored by a former team member of ours who was on your team for a while - the concept of incremental jumping on an evo was largely his invention - and one of the main reasons he left - because the rest of us did not like it!! The argument was that we wanted all promising solos evoed by NOT simply walking the heck out of the first one at the top in the first 24 hours. Funny - we seem to agree on that, but you have a mindset here that no amount of reason can penetrate.

This is simply offensive & obnoxious:
"The *only* possible reason for that is to push other people out of the top scoring range. That is the evil part. It's selfish, arrogant and unsportmanlike, and it shouldn't be allowed."

It would seem that you'd like all the rest of us to get on board somehow to push up your rank, while you are being as nasty as possible.

Get a grip - it isn't going to happen, and no amount of whining is going to help you. Become a better folder - that's the only way here.

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

"Team score IS highest evolver score, for goodness sakes... "

This actually makes my point - the evolver scoreboard is redundant and not needed, and is subject to great abuse.

infjamc's picture
User offline. Last seen 13 hours 47 min ago. Offline
Joined: 02/20/2009
Groups: Contenders

Re: B_2

While I don't entirely agree with your argument, I can see your point that large teams can have an advantage simply because they take up more places on the scoreboard. With that in mind, would you be in favor of the following alternative scoring system?

* Instead of awarding global evolver/soloist points based on relative ranking, calculate them based on the relative position of the game score instead. For example:

Current system:
Achieving 11th place out of 101 players in a 100-point puzzle gives you 100 * ( 1-(10/100) )^7 = 47.8 ==> 48 points.
.
.
New system:
The top-ranked player gets the maximum number of points as usual; this player's game score will be called s_max. Those below the 50th percentile will get a single point; the median score for this puzzle will be called s_min. For everyone else in between, the number of global points they will get will be calculated as follows:

(player_score - s_min) / (s_max - s_min) * points_for_this_puzzle

For example, suppose that s_max = 10000 and s_min = 8000, while the same 11th-placed player scored 9900 (say that he/she only got pushed back to 11th place because #1-10 are players from the same large team who easily reverse-engineered the team solution). Well, this player will now receive (9900-8000)/(10000-8000)*100 = 95 points as opposed to 48.
.
.

==> So why is the new system an improvement? It gives people an incentive to keep playing even if they cannot improve their ranking significantly.

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

Maybe if I say it a different way -

Team scores are meaningful and should be rewarded

Soloist scores are meaningful and should be rewarded

Evolver scores have no meaning and are NOT a measure of success, only a measure of a tiny incremental improvement to someone elses work. Maybe the solo originator of the evolved solution should get the credit!

infjamc's picture
User offline. Last seen 13 hours 47 min ago. Offline
Joined: 02/20/2009
Groups: Contenders

A possible fix would be awarding evolver points based on the amount of score gain, but that could still be abused (just imagine someone uploading the starting configuration and having a teammate evolve that). I see a way to get around this, but it will complicate the scoring system even further:

1. Only count score increases above an arbitrary cutoff (say, the median score of the puzzle) into the equation.

2. Alternatively, the score gains could be weighted based on when it occurred. Improving a starting configuration that scores 8000 to 9000 is probably something that everyone could do via a shake followed by a wiggle, while improving 10400 to 10500 near the endgame is huge.

==> Possible implementation example:
Maximum score = 11000; Median score = 10000; Score of starting configuration = 8000
a) Evolving a 8000 solution to 9000: No credit
b) Evolving a 8000 solution to 10100: Credit is given for the last 100 points above 10000
c) Evolving a 10000 solution to 10500: full credit

As for how to convert these to evolver points:
(score_gain_above_cutoff) / (max_evolver_score - cutoff_score) * points_for_this_puzzle
a) 0 points
b) 100/1000 * 100 = 10 points
c) 500/1000 * 100 = 50 points

Joined: 01/19/2011

I don't think you've picked it up the entire conversation, but we're not evolving in tiny incremental improvements. Sometimes when we evolve, we acquire huge gains (50, 60 points) before sharing it with the rest of the group. Going back to aunty's example, on puzzle 420, Anthropic Dreams managed to get second place overall, even though our top soloist was only in 10th place. This is not, by any means, the result of "tiny incremental improvement"- no amount of LWSing could ever get that kind of result. Evolver is an important part of Foldit because it shows the overall skill of the team. The power of the team's soloists, and the knowledge and skill of the people who improve the solos.

If you look at the top 30 evolvers, you'll notice that a lot of these people are also high-ranked or high-scoring soloists. This is because they are excellent folders in general. Those high-ranked evolvers that AREN'T high-ranking soloists (myself included) are people that are using evolver as an area to learn- we learn to work with proteins that are close to being optimized and eventually apply that to our solos.

And in response to what you said earlier about people being in Foldit just for the points, and not for the science, speak for youself- a lot of us ARE either in Foldit for the science (I eventually intend to go to college for molecular biology and biochem) or just for fun. The competitive aspect of Foldit is the reason that Foldit is presented as a game and not an experiment. So for the people that are in it for the points (I think it's fair to assume that you are), you've done just what University of Washington wanted you to. (:

Joined: 05/21/2010

Oh, we're back talking about points and scores....okay

BUT

I have learned more about folding recently from taking each and every evo I can and doing as much as I can with it than I have in muddling about with my laughable solo efforts. And my team is to thank for it because they have taught me and TRIED to show me (I can be a bit slow to 'get it') and encouraged me. When a high percentage of our members are active we ROCK!

Tyg - AD and proud of it!

Joined: 06/17/2010

Concept on change calculating personal evo score.

I`m not very good in making this percentage maths but my idea is:
player evo score should be calculated based on team score and evo place in team.

- base evo score is team score
- top team player gets same evo points as team
- rest of evo players get smaller amount of evo points basing on place and/or score
- last team player gets same evo points as top player in next team

It would be more "fair"? S1 can calculate it?

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

That's nonsense. What if the 2nd place evolver score belonged to a different team? All the other evolver from the first team get more points?

Joined: 06/17/2010

I we agree that best evo is team effort then yes.
But you have right, minimum should be calculated in different way...

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

And the pile-on practice continues in 447 (but they don't do that, do they?)

spmm's picture
User offline. Last seen 6 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

Whilst the competition is critical (IMHO) to drive us towards better solutions I find the group thing a bit high school some times.

IMHO an evo is a group solution, so the group should get the points, no individual player would get points (well maybe a few like getting one point for doing dev puzzles), as you can't evo without being in a group there is no disadvantage to solo players.

Everything else remains the same, except that you would only have one personal soloist ranking.

The group scores would reflect the work done on evolving by the groups concerned, there would also be a group soloist score.

So:
In puzzle - no change, evo comp and scores the same as now. All the benefits of possibly beter solutions, sharing and learning remain.

On the active puzzle page same as now:

Top group (based on top evolver score)

Top player (evolver)
Top player (soloist)

But once it has closed no individual points for the evo comp (perhaps one point for the effort and 1 point each for top 20?)

At the moment there is a group table, no idea how that is calculated, but there could be two, one top solo group, one top evolver group. This would perhaps increase the effort players put into group evos.

I would also like the top 5 or ten solutions to be available for everyone after the puzzle closes for a few days so that everyone can get the beneifts of interacting with top solutions. We may be missing good possible solutions from smaller teams because they have less compute power, time, or scripts. A final all hands evo at that point would be great, so the whole foldit group can work together.

spmm

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

I think is the best proposal so far to change the Evolver scoring.

The idea that the evolved solution is a group solution is key, the group should get the points.

* Score the best solo solutions as now for the Soloist rankings

* Score the best over-all team score - which may be a solo solution or an evolved solution - for the Team Rankings.

* Score the best team evolved solutions for Team Evolver standings - each team gets one entry in the table

This may have just been a re-statement of spmm's post, but it really makes sense.

I'm not sure about the the top five solutions after the puzzle closes point. Are you proposing that we get to work on them as "All Hands" solutions? Perhaps those chosen would not be the top scoring solutions, bus chosen by the project as the most promising. But - isn't that what happens now when puzzles are re-released with a number of starting points based on previous versions of the puzzle?

spmm's picture
User offline. Last seen 6 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

Just for the record 447 3 days before close:
447: De-novo Freestyle 12
Status: Active

Evolvers
Rank Player Group Score Points
1 barbaranne 2 116 Anthropic Dreams 10,869
2 tyler0911 11 51 Anthropic Dreams 10,858
3 auntdeen 8 10 Anthropic Dreams 10,857
4 Rav3n_pl 28 14 Anthropic Dreams 10,844
5 harvardman 31 123 Anthropic Dreams 10,774
6 phi16 6 88 Anthropic Dreams 10,765
7 Tyggy Too 62 99 Anthropic Dreams 10,758
8 gdnskye 4 13 Anthropic Dreams 10,716

9 CharlieFortsCon... 7 18 Contenders 10,664
10 Bletchley Park 1 15 Contenders 10,663
11 svenmh 18 81 foldeRNA 10,618
12 bertro 34 59 foldeRNA 10,591
13 hansvandenhof 18 23 foldeRNA 10,590
14 B_2 15 39 Go Science 10,587
15 Roseanna 47 91 Go Science 10,569
16 infjamc 36 9 Contenders 10,566
17 justjustin 49 73 Anthropic Dreams 10,559
18 itskimo 10 58 foldeRNA 10,545
19 firejuggler 70 258 Void Crushers 10,542
20 cbwest 29 25 foldeRNA 10,540
21 micheldeweerd 37 122 foldeRNA 10,470

Joined: 04/19/2009

For the record - at this time, all 8 of the evo players listed above (with the exception of one at #5) - have done independent evos from Pletsch's solo uploaded at 10732.

There are 2 uploaded evos from that score that no one has touched.

Each of the seven have arrived at their evo scores using different hand work, scripts & techniques.

What some are just not getting is that we enjoy evos - finding something that a great soloist may have overlooked or not yet done - whether it be hand work (which many of us do in evos) or scripts.

So if it's the will of the devs to discourage this in some way - then fine. They have access to the logs, so whether Brick thinks it's nonsense or not, they can confirm this for themselves - and quite frankly, their opinions are really the only ones that count.

If the devs wish to limit the size of teams, how evos are scored or worked on - whatever they want is fine by me. But I am more than a little tired of foldit's least favorite concern troll curmudgeon taking potshots at our team, and stirring up storms. Maybe if he turned his efforts to becoming a better folder, we'd all have some peace.

I appreciate the thoughts of inflamc, rav & spmm for coming up with reasonable thoughts on evo scoring. But quite frankly, it would seem that the only one who is so very unhappy is Brick, who either wants a better number after his name, or failing that, wants no one else to have it.

When I took a good look at the top evolvers list today, I saw a healthy mix of folders from all the top teams. That isn't surprising - the top teams either have the top soloists and/or the energy & dedication to work very hard on their evos.

I am so done with this conversation. I am very done with trying to reason with a brick wall.

spmm's picture
User offline. Last seen 6 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

Making this discussion an emotional personal argument between two people and 'being upset' doesn't make it go away and should not be an excuse for others to stop the conversation.

Others are also on record as saying they can't see the point of the evo points system for individuals, it is a group solution.

You haven't commented on my suggestions, which is your option of course.

I certainly 'get' that some people in AD enjoy evos, on puzzle #446 there were five AD evolvers, on #447 there are nine; and they don't share their much vaunted evo techniques with the rest of foldit, and as I said earlier, to me the group competition seems quite childish, and I would like everyone that folds to have an opportunity to improve their game.

I agree that it is up to fold central to make any decisions but we can discuss things as rational adults if we choose, hopefully informing their decisions.

If we humans are entered into CASP again; is it going to be a divisive battle between individuals and groups, (I missed the last one) or is there going to be an opportunity for everyone to work together? If the latter then perhaps some practice may be of benefit.

Joined: 04/19/2009

@spmm - I have no problem conversing with rational people, but am very tired of some false perceptions that are repeated over & over again, in particular when a certain person deteriorates into slinging garbage along with them.

We do not have any one technique - or even a series of them - or even super secret special recipes. We all have different methods of approaching evo. Mine, for instance, is to look at whose solo it is. Pletsch & keypad, for instance, use few scripts - Rav3n does most of his hand work only in the beginning (and then allows his excellent scripts to work for him). When I see a solo from any of these - I do the opposite of what I know they have focused on. Other people on the team are excellent at hand banding techniques to knock the protein loose and pliable again. Barb is incredible at selecting the perfect script at the perfect time to get the most out of a solution. But our goal is always to try to knock a protein loose enough to be able to make significant improvements - even at the end of every game, we always have at least one person working on a "hail mary".

This would appear to be what the devs are looking for with evolver play.

The other thing we do is clearly label our uploads - so that when someone is looking for a shared solo or evo to download, they are aware of how many hands have already worked on any given solution, and what's been done to it, or whether it's a new high scoring solo.

That's it - other than the energy that we bring to it. I've been working for 2 days now in one client on Pletsch's original upload, and have managed to improve it by almost 130 points by using both hand & script.

Because we do so much with evo, I would not be happy with a change in evo scoring for individuals for a couple of reasons - we use it to train newer or newly returned folders, and it does give a decent sense of accomplishment to them while teaching/refreshing folding skills.

Most experienced folders agree that it is silly to have the default players page open with evolvers - I would like to see the page open with soloists! But I see no harm when looking at the top evolvers list where all top teams are represented. I can't imagine that it "hurts" anyone to have an evolver number - but do see that it can help a tenuous or inexperienced or returning folder get a feel for working on a good solution, and a sense of accomplishment that can give them the self-confidence to turn their skills to solo.

Thunk was a top evo person for many months before he finally started feeling confident enough to really work on his solos - we actually had to kick him through his first high scoring solo. He is placing often in top ten now, and recently got his Master Soloist. Without the competitive feedback that came by being the #1 evolver for a time, it is very possible that foldit would have lost him.

Most competent soloists know that the number that counts is solo - certainly we have excellent soloists who do not play on a team so they obviously don't care. I see no real reason to change evolver scoring, and some harm that could come from doing that, from my perspective.

Your last question does make me smile, about CASP. Our team was started by Pletsch & Renton as a big, open team for all folders as an opportunity for everyone to work together for CASP 9... Please read our page for the details.

Considering all the flack we've had about being open and being large, I wonder if anyone would want to attempt the same thing again...

Joined: 06/17/2010

During next CASP we should have all-hands puzzle ability, cross-group sharing. We all be visible as one team - Foldit team.

infjamc's picture
User offline. Last seen 13 hours 47 min ago. Offline
Joined: 02/20/2009
Groups: Contenders

1. While I'm not against the intention behind all-hands puzzles, let's not forget the important observation from the Foldit developers on this type of puzzle: In practice, people tend to converge toward evolving the top-scoring solution rather than exploring.

==> So, as a compromise, I would be in favor of a two-round system where the top 5-10 unique solutions from Round 1 (where "unique" is defined as two solutions having an RMSD above an arbitrary cutoff) are automatically provided as templates for Round 2.
.

2. Regarding the idea of individuals not getting credit for evolving: my concern is that this could decrease the incentive for evolver play. But the following system might work:

a) Record the score gain achieved by each player above a certain threshold (say, the media score of the puzzle).
b) The player on each team that contributed the most to evolver player will receive the same number of individual evolver points as the team global points for this puzzle. Everyone else on the team would receive a reduced amount that is scaled down proportionally.

For example: A team is placed 2nd in a puzzle and receives 75 points. The median score of the puzzle is 10000 points, and the team's top evolver solution scores 11000. Suppose that one player contributed to 400 points of the 1000-point score gain, while three other players contributed to 300, 200, and 100 each. The four players will receive 75, 57 (56.25 rounded up), 38 (37.5 rounded up), and 19 (18.75 rounded up) points, respectively, in evolver global points.

Now, this system does have the downside of giving too much credit for the lower score ranges, where evolution is easier. A possible fix might be reweighting the score gains by the range where the score gain has occurred. For a more complicated example:

(Starting configuration = 8000, median score = 10000, team evolver score = 11000)

Player A: 8000 -> 10400 (note that only the last 400 points would count)
Player B: 10400 -> 10450
Player C: 10450 -> 10500
Player D: 10500 -> 10600
Player B: 10600 -> 10850
Player C: 10850 -> 11000

Now suppose that relative contribution is reweighted based on the function (credit) = 2 * (location_of_score_gain - median_score) / (team_evolver_score - median_score). Integrating this function yields x^2/1000, which is then applied for each segment of score gain

Player A: x = 0 to x = 400, relative contribution = 160
Player B: x = 400 to x = 450 and x = 600 to x = 850: relative contribution = 405
Player C: x = 450 to x = 500 and x = 850 to x = 1000: relative contribution = 325
Player D: x = 500 to x = 600: relative contribution = 110

So Player B actually gets the most credit despite contributing to a smaller number of points from an absolute standpoint... because most of the points occurred in a higher range, which is more difficult. Now Player B gets the full 75 global evolver points for his/her evolver play, while Players A, C, and D receive 30, 61, and 21 points, respectively. (As you can see, the main downside of this system is sheer complexity, as the scoring function would have to be integrated for each team.)

Joined: 05/21/2010

Okay...last time for my input...talked to death and I really don't care...

BUT

AD has EVOLVED into a powerful group that folds particularly well. Think of us as a really good multicellular organism in a sea of colonial organisms. We use our knowledge across puzzles, and are constantly coming up with new processes that benefit foldit, the team, and the solutions. And IF a good percentage of the team gallops merrily down the paths of high scoring towards a single pinnacle, doesn't that leave every other pinnacle for other individuals and teams to climb?

And if you want to whine about my evo-ing....here is my secret technique...three hours solid of hand work using a technique Dat showed me a year ago: Move something, wiggle, shake, repeat.

Submitted for your approval by a top predator,
Tyg

Joined: 05/09/2008
Groups: None
Status: Open » Open

Seth and I will be in chat for 1 hour tomorrow starting at 13:00 PDT (20:00 GMT) to discuss potential solutions to the design problem spmm original brought up with this feedback.
We will be happy to discuss Evolvers as well.

We will post a transcript of the chat in the forum.

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

How about having these chats when the majority of the US population is NOT at work?

Joined: 05/09/2008
Groups: None
Status: Open » Open

1pm tomorrow is the soonest Seth and I are free to chat.
We do not have a large staff with Foldit customer service available 24/7, that is why every single developer chat has been during UW work hours Seattle time.

If you are not able to take a lunch break at that time, please find a teammate that IS free to represent your views.

We hope that this chat will be as useful as the previous chat on cloned accounts, which resulted in the addition of Tracks.

Please leave the status of this Feedback as "Noted" since we have noted the problem that has been brought up.
As you pointed out in this feedback: http://fold.it/portal/node/987997#comment-12371 you can easily find this feedback by selecting "All". Thank you.

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

So in this case, only the unemployed, self-employed, and kiddies get to participate.

IRC is not allowed on most real-life enterprise networks for very good reasons, so lunch-break is not an option, and I would not want to impose on the other team members to try to get my ideas across.

And yes, I was very artfully blocked from the last chat discussion for the same reason.

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None
Status: Open » Open

Except your team is NOT open, and you do NOT share outside your team. Lots of grand words, but the actions do not hold up.

Having 15 AD at the top of the evolver list is a very strong dis-incentive for others to evolve. There is simply no point to it, and is truly a waste of time.

gramps's picture
User offline. Last seen 6 years 26 weeks ago. Offline
Joined: 12/16/2010
Status: Open » Open

**ahem** neither is this thread ;-)

Joined: 05/21/2010

Oh, this is sweet. Too sweet to give up. I haven't watched someone have a conniption fit over a game in a long time.

Top Predator,
Tyg

spmm's picture
User offline. Last seen 6 weeks 5 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

As a footnote, thanks to Beta_helix and Seth for taking the time to speak to the folders.

This from the AD page:
''Anthropic Dreams was formed in Spring of 2010. Pletsch and Renton, two top folders, decided to create a team open to all Foldit players whether they were in teams or not, to make the best possible folding team for CASP. They hoped that by pooling all of Foldit talent, Foldit would have the best chance of winning CASP.''

So what happened? This is a closed group, and you cannot be in another group and be a member of this group. It is not all hands. So there is now one group three times larger than all of the others.

Group sizes based on active players August 2011, top ten only

GROUP #players Top solo rank
Contenders 19 1
AD 60 12
foldeRNA 19 4
VC 12 2
GoScience 10 6
Russian 8 20
Tornado 3 36
Seti 12 55
LÁlliance 12 60
Purdue 1 99

Joined: 04/19/2009

@spmm - might as well finish what you lifted from our page: "When CASP was over, the active team members decided to stay together." Because we discovered that we had something special going - we found out that the synergy of many went way beyond the dependence on any one person/soloist. The team is moderated, not closed, just as yours is. And by the way - our true active is more around 35 - our team members feel free to come & go, dependent on real life concerns, because they know that we are not dependent on one or two people.

I'll tell you what - let's make a rule for all foldit teams that they are capped at a certain number - whether it be active players, or gross number, or number in the top ten, twenty, fifty or elebenty seven. If that will cut the garbage around here - then fine.

However, if that appeals to you - let's go a bit farther to level the playing field, since this is where it seems to be heading... And it will be SO much more even, that those of us forced to leave our team when you cap the size will feel SO much better...

Let's only allow teams to have only one or two soloists in the top ten. Hmmm - that will be a problem for the Contenders - they have 4 in top ten (and number eleven). But it's just SO not fair that they have that much talent, right? Just like it's SO not fair that AD has so many folders.

In fact - let's be really egalitarian, let's level the playing field entirely and restrict to just one top ten soloist in each team - oops, there goes VC with 2, folderna with 2 (Go Science will really be thrilled since they only have one). That little rule will help us atm, since our highest soloist is only at #12. Could be a problem for us in the future if more than one of us makes top ten (but if we don't try as hard as we do now, then no problems - maybe we can take turns).

In fact - let's force the teams to break up and restart under this new system. It's only fair, right? And if someone else on a team manages to improve their rank to top ten - well, they'll just have to leave - or maybe the rules should be that the old one goes.

If you want a level playing field, then I guess that this would be the fairest system possible. Who cares if people want to be on a team together? Let's even have foldit assign you a team by rank - that would be best.

Ridiculous? Very. Few of us would stay if any of this was implemented.

Any team can choose to be large - actively recruit and then train. Any team on foldit right now can "compete" with AD simply by increasing their size. Don't want to be part of a team that size? The answer is simple - don't be. But for goodness sakes, stop the sniping at the team that is.

vixx's picture
User offline. Last seen 9 years 7 weeks ago. Offline
Joined: 08/27/2011
Groups: None

Perhaps if a few of the players of foldit had behaved as adults, instead of jerks..AD would still be open.

Spmm and brick..the two of you seem to spend all of your energy having a go at one group...are you both peeved that they havent sent you an invitation to join?..why do you both continue with your garbage...focus on achieving your best instead of looking like plonkers all of the time.
The two of you read like pathetic little boys because mummy took your ice cream..grow up and spend the effort on yourselves and your game..you may one day become better people

Joined: 05/21/2010

And it DID put together an awesome open team for CASP. So what's your point? And your little chart...um, what on earth was it supposed to demonstrate?
How our team is managed or comes together is frankly none of any other teams' concern, now is it?
Our team is active, comprised of many happy and competetive players, also many curious and intelligent people who think outside the box and make things happen. Can't deal with it? Awwwww. As I said previously, AD ROCKS. You can try to 'level the playing field', but we'll always have a leg up on you because of our players. Always.

Grinning and bearing,
Tyg

B_2's picture
User offline. Last seen 5 years 30 weeks ago. Offline
Joined: 11/29/2008
Groups: None

We just get tired of hearing the queen bee harp on how open the AD team is, when is is NOT open, not anyone can join, they have to get approved by said queen bee, and if she doesn't like someone, they get kicked off, as a number of people did this last few months.

Joined: 04/19/2009

rofl - just rofl...

Joined: 04/19/2009

Whew, had to catch my breath there. Sometimes you are just one of the most amusing people on foldit - nasty, ill informed, pigheaded, no sense of decent manners whatsoever - but always good for a belly laugh at the end.

Joined: 05/21/2010

Once again, if you don't want to be on team AD, that's okay. But how we run our team is really none of your concern. Your envy is palpable.

With love and kisses,
Tyg

Joined: 05/21/2010

Just realized that this isn't a populated forum. . .packing up and taking all my brilliant one-liners and going back to my jungle.

"Queen BEE"
Tyg

ps, no particular interest in designer puzzles yet. Hope any tools that my team cares to share to teach me aren't yanked out from beneath my paws.

Joined: 05/09/2008
Groups: None

I just posted a news item about the developer chat tomorrow.

For all of you that are very passionate about the original topic of this feedback:

"In the latest Design Puzzles (because they are so rigid) if one person in a very large group (with active coders) gets to a top ten rank then that solution is very quickly copied using a script, then a script is written which resets the mutables and is then shared to the team. The position can then readily be copied by loading the solution as a guide."

...it would be great if you could try thinking of solutions to THAT specific problem. You can bring your ideas up in the developer chat tomorrow or (if you can't make it to chat) feel free to post them here AS LONG AS IT RELATES SPECIFICALLY TO THE ABOVE concern.

If you have other concerns that you feel more passionate about (such as the last 40 comments) then I encourage you to open a feedback about THAT topic.
Or if you rather just start a discussion with other players, then opening a new Forum post would be useful.

If future comments in this feedback post could stay on this feedback's topic of "design puzzle play" that would be helpful to the developer team.

Thank you.

Sitemap

Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, RosettaCommons