34 replies [Last post]
Joined: 12/06/2008
Groups: Contenders

Everyone has their own tips and tricks that help determine how successful they are at this game. With the advent of recipe and script libraries, we can share what we've done with an audience the size of our own choosing. We can keep things to ourselves, share them with teammates, or make them available to the entire global membership.

The main question is: How much information are we expected to share with the rest of the Folding community?

The corollary questions:

* Does keeping information to oneself prevent someone else from using that information to make a greater gain for the good of the whole?

* Does sharing information with all dilute the friendly competitiveness of the game and negate the reason for scores and rankings?

I see pros and cons for all responses from "none" through "some" to "everything"; and I would like to know your thoughts on this, fellow folderers.

Joined: 03/20/2010
Information Sharing - Self, Team, and Community

Hi Boots,

This very same question has occurred to me from the very beginning of my folding experience. On the one hand, the whole purpose of Fold-it being for the good of the scientific community, and by extension, conceivably, for the good of the health of the whole human race - would seem to elevate knowledge sharing to the highest possible priority, over-riding all other concerns.

On the other hand - I myself - find myself rather jealously guarded and protective of my very best techniques and discoveries, while sharing mainly only what I might call my "second-tier" insights, techniques and scripts, and then, only with my group - holding back the "tweaked" versions, to squeeze out a tiny edge; there is, I think, definitely an intrinsic human urge to gain advantage in each of us, and I must "fess-up" to having perhaps more than just a touch of that myself, before continuing. So I cannot fault others for having this very human, understandable view or stance. Some would even say there is great virtue in that natural self-interest

On further reflection, however, I know that the people I respect and admire the most in this game are not always those at the tops of the scoring heaps (although I do definitely admire their skills); rather, they are those few who excel not only in scoring, but in willingness to teach, and to share, and to assist the less accomplished members of their team, and to foster newbies' growth, and to elevate the level of skill of all other players they come into contact with, whether it be by sharing knowledge and answering questions in chat, pm, or email, or by publishing and sharing hints, tips, tricks, scripts and code.

These few who have overcome their need to excel individually and have channeled their energies into elevating everyone around them are at the heart and soul and essence of what makes Fold-it such a great thing to do, and this site such a great place to be.

Because of this, I have found myself slowly overcoming my "hoarding" instincts about knowledge and code, and have begun, if modestly, to contribute back, at least to my team (so far). I envision, that as time passes, if I should become as accomplished in this folding endeavor as I seek to be, I will become more and more a "contributing citizen" of the folding community at large, and less & less a solely competitive player.

That being said, competition is FUN! Being part of a team is FUN! ...and winning is FUN, FUN, FUN!!!

But losing really Sucks - except insofar as it motivates us to increase our knowledge and skills. But c'est la vie; such is life in a zero-sum game.

I think perhaps the main argument in favor of knowledge and tool sharing (aside from the altruistic) is that, if everyone has exactly the same tools and knowledge, then those that rise to the top in score would become, more and more (over time), those with the skills and ability to use those tools to best advantage, and less & less those that have acquired the most outstanding tools, but are perhaps not as individually skillful. Personal motivation, talent and ability would become even more the separating factors than they already are. In my opinion this would probably be a very good thing. I think it would probably eventually result in the scores reflecting in much greater degree, a player's inner abilities and gifts that cannot be shared - this would be a much truer distinction.

On the other hand, many of those currently at the top of the scoreboards may understandably be loathe to give up or share their knowledge and tools that continue to give them their current distinction.

I may even be among them, even though I'm certainly not at the top by any means; I can at least understand that very human urge to guard one's position and whatever satisfaction and esteem may be derived from it. Not even to mention team loyalties, where peer pressure to guard the team's position may well over-ride one's instincts and desires to share knowledge and elevate others' games

Overall, I think the democratization of knowledge and tools, tricks & scripts could result in a new level of accomplishment for Fold-it as a whole, albeit unfortunately paid for by ranking decreases for some, perhaps myself as well.

The crux of the matter may lie in each individual player achieving a quantum leap in personal identification, i.e, from:

1) "I am in this to succeed on a personal level and derive great pleasure from that, to...

2) "I am in this to succeed with my team, and to help my team succeed as they help me to succeed so that we all may enjoy mutual success, and derive my personal pleasure from that" to...

3) " I am in this to help the whole Fold-it Community to succeed, to grow and improve, even if it may mean that my individual rankings or even my team rankings may take a step or two backwards, and derive my personal satisfaction from that" and, possibly even to...

4) "I am in this to truly do whatever I can to help everyone else around me to possibly help the whole human race, and derive my own sense of personal satisfaction from that."

What would this game be like if we were all to take a step up from whichever of the above best describes us now?

Anyway, I am delighted that you raised this topic for discussion, Boots, and delighted to be able to air my views on the matter. Although ostensibly just about the game, I find these questions you've raised apply equally if not more so, to our very core values, and to our individual characters.

I really hope many others find this thread and reply as well; I'd tremendously love to hear others' views on this matter.

-Best Regards to all,

dimension9

Brick's picture
User offline. Last seen 12 weeks 4 days ago. Offline
Joined: 07/15/2008
Groups: Beta Folders
I think this quote from

I think this quote from dimension9's post says it all. It represents to me what Fold.it was supposed to be about - adding the human element to the Rosetta algorithms, using the power of the mind to visualize the problems in a protein fold, and use intuition to modify the result for the best outcome.
I think perhaps the main argument in favor of knowledge and tool sharing (aside from the altruistic) is that, if everyone has exactly the same tools and knowledge, then those that rise to the top in score would become, more and more (over time), those with the skills and ability to use those tools to best advantage, and less & less those that have acquired the most outstanding tools, but are perhaps not as individually skillful. Personal motivation, talent and ability would become even more the separating factors than they already are. In my opinion this would probably be a very good thing. I think it would probably eventually result in the scores reflecting in much greater degree, a player's inner abilities and gifts that cannot be shared - this would be a much truer distinction.

As it stands now, it's become a boring ritual of merely applying the shared recipes, dumbing down the game, making it a competition among only a few with the skills and knowledge to create the scripts.

Without a common shared set of tools, including the scripts and recipes, Fold.it will become the playground for an ever-decreasing group of those-that-have against all the rest.

Joined: 09/18/2009
Groups: SETI.Germany
interesting topic

Well, most is already said before.
In reference to the recipe topic, I've also the opinion that sharing them for all useres makes sense, because it creates fair conditions, and the result diversity will mostly depend on the human abilites than on technical.

Plus, playing a non-profit-game helping science for the benefit of ALL people but using recipes available for only some people playing it looks ironic to me.
There are enough companies which try to gain or keep their monopolic status quo (oil companies for example, which try to prevent or handicap the invention of alternate power-sources).

One thing which makes me a little bit unsure is the changing of recipes created by other users before,
when it comes to the question: Upload it or not?
I'll never upload an edited recipe created by another user if he didn't give me the permission.
I can change a recipe, remove some flaws and add some features.
But what if the person doesn't want me to upload it, or simply doesn't give me some answer.

Tlaloc's picture
User offline. Last seen 10 weeks 4 days ago. Offline
Joined: 08/04/2008
Groups: Mojo Risin'
This has been on my mind

I am in the "sharing knowledge" camp. Many of you use my scripts and they find you points. Every time I publish a script, it makes my job as a soloist more difficult because I have given away a tool that I find points with to "the competition." Although my score doesn't change, my rank suffers. What kind of irks me a bit is that many of the people using my scripts are taking my knowledge and using it to rank up, but don't share back. They have virtually all of my techniques, but I have none of theirs. There are people who consistently score in the top 20 who know things about folding that they aren't sharing.

What softens this a bit is that rank is a game...the techniques I publish aren't. If a script I write helps fold one protein that creates one medicine that helps one person, I will be feeling good about it for the rest of my life.

Joined: 09/18/2009
Groups: SETI.Germany
Right, Tlaloc.

Yes, looking at some script sites, you see that there are many "children" of it, but they are not available for download.

On the other hand I guess, some of them are very close to the original, but only use different parameters.

Joined: 05/19/2009
Groups: Contenders
"There is no spoon"

I recognize dimension9's view and appreciate his comments.

Some members of Contenders are masters in hand-folding. Their use of scripts was limited until the pace went up and some tasks just had to be automated to keep up. They do not use secret superscripts. They use the tools that people like Pletsch, Grom and others have generously provided and for which they deserve credit.

Anyone can make advances with these basic scripts provided. A change in banding can cause a different quake to yield better results on a given protein. These variations are available for anyone to make and to explore. In their own time and by their own efforts they can so advance their scores. Those who write LUA scripts had to learn LUA, this is their personal effort, and they bear fruits of it. The definitions of LUA are available to everyone, so everyone has the ability to apply it.

Achieving high scores consistently does not require superscripts, although they help. It first and foremost requires dedication, many hours of observation, trial and error, hand-tweaking and last but not least, expertise which you will only get by playing often.

When I was a newbie I was amazed and suspicious about those incredible high scores and thought "they must be using some secret script". Along the way I learned what works and what doesnt.

Because my time is so limited I make extensive use of scripts, even if they are less efficient than hand-tweaking. To run these I purchased parts and built special dedicated hardware, available exclusively 24/7 for FoldIt. This required finances, build time, knowledge and travel. This machine is portable, I take it with me on my travels so I can keep up with the pace of CASP9. All this involves effort, dedication and determination too.

I do not use superscripts, although one might call Quake, Overnight and Acid tweek superscripts. I use *many* variants of these scripts in a variety of situations, in various orders, based on visual judgement. I use scripts because I simply lack the time to manually tweak proteins. My life is very busy.

To make a script public is up to the maker, and for derivative works from a non-published script also on the original author. It is highly appreciated what people like Grom, Pletsch and others have done. It has most certainly advanced the pace of FoldIt. It does not mean though that every script *must* be public. A script is like a diary, if the author writes in it, it does not mean you can claim to read it nor that you should scorn the author for keeping it from you.

You (in general) may not realize it, but the playing field is more level than you think.

Joined: 09/18/2009
Groups: SETI.Germany
Hm...

Yes, hardware and time is a big factor.
I've seen recipes, which took hours on my PC on a mate's overclocked Quad, taking only minutes.
So, running a brute-force recipe (random trying or something like this) on a fast PC for one night can give better results than running it on a slow machine for the same time.

Brick's picture
User offline. Last seen 12 weeks 4 days ago. Offline
Joined: 07/15/2008
Groups: Beta Folders
No "superscripts" ?

If this is so, why are certain groups and individuals so concerned about the scripts being shared?

The playing field really isn't level. There are those who can write scripts, those who can only use shared scripts becasue they have no scripting or programming skills, and those who have the time or inclination to hand fold everything. I don't know how anyone could do that during CASP and ever make any headway.

I suspect that the groups that have those few talented script writers will always exceed, and the rest will never be able to compete. That is very demoralizing.

xiando's picture
User offline. Last seen 4 years 36 weeks ago. Offline
Joined: 10/19/2008
Groups: Oma Gawd
So what you're saying?

So what you're saying Brick, is that given two players of similar actual skill, one which is new to fold.it or simply independent, and the other who is involved with a group, the use of a script or scripts could theoretically result in a situation in which the group player could create a solo score far in excess of the other player's score, just due to to someone else's actual folding skills and scripting skills?

That would be terrible. I could even imagine situations in which a very poor player could exceed the score of a moderate or good player, just by using some "superscript" provided by his/her group.

Considering the hours required to fold a protein, (presuming one does not simply clone someone else's view point of how a fold should be performed by pushing the button on a shared or otherwise superscript) what then would be the point of a moderate or even good solo player to bother participating to the project if his or her results would be obviated by an essentially cloned result (even if the scripts were tweaked so that the scores didn't show as "too" similar)?

Once again, the questions come up, what is Solo, and what is Competition? And what is the _stated_goal_ of this "game"

Sorry, I'm trying to get my head around the idea that running a script someone else wrote to achieve 8 or 10K is either a "human" or "solo" achievement by a solo competitor.

Afaik, this is a game. it may be part of a larger picture, but like MMoprhs and similar, one does not just create a script to get the +5 sword of Valkerie so one can tell everyone at the fruit juice bar that they're really cool...one actually achieves the sword on their own merits, or one is simply a fraud.

SteveL's picture
User offline. Last seen 3 years 6 weeks ago. Offline
Joined: 12/29/2009
Groups: None
Sharing Scripts

I'm pretty much a newbie at this, so keep that in mind, but I think sharing scripts is a great idea. I'm not as interested in getting into the top 10 as I am in simply improving my score--and in learning about how proteins fold. (My final ranks still seems to vary between 30 and 50.) I study the scripts I download and try to figure out why they work well or not, and in return, if I ever get good enough to create a script that really smokes, I'll definitely upload it so that anyone can use it. That's the least I can do after making use of others' work.

When I find a script that helps my score significantly, I usually write to the author and thank him or her. Some players are incredibly helpful: Madde in particular suggested numerous scripts that are useful in various situations. My real goal is to learn about proteins, and I've already started looking into some of the textbooks. If I rank up over some of the stars (Boots, Bletchley, Marie Suchard, Tlaloc, etc., and many others), it generally lasts for only 2 or 3 minutes, but it's a thrill nevertheless, and that's all I need to keep playing. The way I see it, the only important competition is with myself.

Joined: 12/06/2008
Groups: Contenders
On the other side of the debate....

Just for grins, I'm going to play Devil's Advocate here, and take the position that it's okay to share information; and that it's equally acceptable not to share it.

It seems we might have forgotten what the purpose of the FoldIt project actually is. It is about trying to discover methods to help program computers to more effectively predict protein tertiary structure. The game itself is ancillary to that purpose, and serves only as the vehicle through which these discoveries might be made.

I would think the Baker Lab doesn't give a rat's patoot who did what, or whether they used "superscripts", hand folding, or The Force to do it. All that likely matters is results, and how they were achieved.

Please allow me some license to use a fast-food analogy. "McDonalds" is analogous to the top scoring teams and individuals. "Burger joints" are the lesser scoring teams and individuals. "Hamburger" is analogous to folding predictive methods; the "consumer" are the folks at the Baker Lab. "Restaurants" are the entire community of folderers; and "sales" is analogous to their scores or ranks.

Is McDonalds obligated to disclose their methods of success to the mom-and-pop burger joints, because "the playing field really isn't level"?

No. And why does it matter, except to the jealous? In the consumer's quest for the best hamburger, all restaurants, no matter how they prepare, cook, or serve, have something of value to offer. It doesn't matter to the consumer which one has the largest sales, as long as he eventually gets what he seeks. He will review each and every restaurant, and then decide for himself what is best for him.

So you don't know or have what McDonalds has? So what! Go invent it, yourself! If you don't have the knowledge, go learn it. If you don't have the time, go make it. If you can't or won't do that, make the best of what you do have. No matter what your circumstances are: high; low; or in between, you have the innate ability to create. You might just come up with the formula that trumps all.

It's all about the consumer, fellow folderers, not about how well we restaurants are doing. Wisdom suggests we instead concentrate on creating the hamburger the consumer will want the most.

xiando's picture
User offline. Last seen 4 years 36 weeks ago. Offline
Joined: 10/19/2008
Groups: Oma Gawd
and how they were achieved.

exactly. From the perspective of whether or not a protein is folded, it doesn't matter how it is done, as long as it is done well.

The issue really boils down to whether or not Baker is presenting the results as examples and/or affirmation of the "human element", as specified in the faq. If so, then the apparent prevalence of largely computer crunched proteins calls any such claim into question.

As to it's use in solo competition, the use of any such "superscripts" is not analogous to McDonalds. It is analogous to copying someone else's homework and calling it your own. I believe there's a word for that, but it escapes me right now.

xiando's picture
User offline. Last seen 4 years 36 weeks ago. Offline
Joined: 10/19/2008
Groups: Oma Gawd
*correction

that would be by anyone other than the author of said superscript

Brick's picture
User offline. Last seen 12 weeks 4 days ago. Offline
Joined: 07/15/2008
Groups: Beta Folders
Look at the recent puzzles

There were only 12 evolvers on the puzzle just closed, 185 solo participants.

That seems to indicate that everyone is simply running up their soloist scores using the recipes du jour, or the superscript developed for use by specific teams.

Also take a look at the scripts that have been shared - there are only a handful of people who are really writing original scripts, everything else is derivative.

So, each puzzle is overwhelming the result of people all running the same group of scripts. The scripts have become so good at making points, there is no incentive to spend the time exploring on your own and experimenting to see what works. You end up left in the dust at the bottom of the score heap. And remember, this is billed as a game, a competition, so it's the competitive nature of people to do what is allowed to win, in this case running scripts.

Joined: 09/18/2009
Groups: SETI.Germany
Hardware...

I believe what Bletchley Park wrote, he convinced me.
It's the hardware, which can make a regular script look like a superscript.

But anyway, it would be interesting to save the recipes' names which were used for a solution and to show them, too, if player's name and score with the best solution is published.
This would also be informative for improving ROSETTA.

marie_s's picture
User offline. Last seen 1 year 50 weeks ago. Offline
Joined: 05/18/2008
Groups: None
I am not a star, SteveL

I start playing 2 years ago.
See the top soloists of the week, most play for more than 1 year, frequently more than 2. Many play very often.
I think scoring rewards personnel experience with assistance of tips learned in chat, wiki and scripts.

I think the equilbrium between competition and cooperation is good but reward of innovation is difficult, here I see no other reward than the assessments by peers. (reward of innnovation is always difficult : licence, copyright, ... are not perfect systems).

The barriers to entry for new players are high. They are not knowlegde in biology or big hardware (CFC have a not so good PC) but the sum of knowledge to be of the best and some knowledge in computer language to understand scripts.

So, new players, learn for yourself (we play for many months because we discover new trics every weeks), with an eye on the chat, ask a question for time to time, we will be happy to answer.

Joined: 09/18/2009
Groups: SETI.Germany
More...

I know players, who have only very rudimental script knowledge as they told me, so they can't write superscripts, but do good folding jobs, way better than me.
They have even won some contests by using scripts just as they are here available for download.

Brick's picture
User offline. Last seen 12 weeks 4 days ago. Offline
Joined: 07/15/2008
Groups: Beta Folders
Lucky tries

There are always going to be the lucky folds that outperform the scripts every now and then, but the overwhelming truth is that the jealously guarded group scripts have an advantage over those that have to make do without these tools.

It's very much like the early days when there were people folding on contest puzzles using beta tools.

The "haves" are not sharing with the "have nots" and it is creating a very unfair playing field.

Brick's picture
User offline. Last seen 12 weeks 4 days ago. Offline
Joined: 07/15/2008
Groups: Beta Folders
Challenge

If your group scripts are not an advantage, then go ahead and make all scripts public. Then we'll find out.

xiando's picture
User offline. Last seen 4 years 36 weeks ago. Offline
Joined: 10/19/2008
Groups: Oma Gawd
clearly the stance is showing in participation numbers

Considering the steady decline in public participation, it should be apparent to all but the most obtuse that the present paradigm of anything goes FTW! (as long as one is a member of the fold-it-at-home camp) is failing to achieve or retain new players. But then, perhaps that's your goal?

Joined: 05/03/2009
Groups: Contenders
My 2 cents, for what it's worth

Forgive the history lessons, but take a step back, and reconsider FoldClub afresh. Rosetta@home, is a crowd-sourcing project, whereby you give up your CPU, and automated prediction algorithms endlessly crunch numbers. The genius of FoldClub, is that you get your hands on a protein and see what you can do with it. This is the whole essence of the 'Komps v Hooman' debate that crops up in chat. Komps are told what to do, Hoomans can always choose. But in order to get you addicted in the first place, there is a competitive element, in that scores get you positions on the board, repeated positions give you global rank. And so, we keep coming back to 'play'. A further development of competition is team play, a casserole of individual approaches and talents.

But, the human element is not as straight-forward, as simply being able to say, 'That bit, doesnt look right'. The way we play, is typified by knowing WHEN to do things, when to rebuild, when to wiggle, when to stop shaking. That only comes with experience.

The development of FoldClub, the new tools that arrive, is an intereting learning curve. In the days before freestyles, we would usually be given Rosetta predictions, to refine. Then as those refinements edged towards a resolution, we'd get a Native guide, to Quest Towards. Then we were given raw Freestyles to work with. As an insight to the way FoldCentral and Rosetta work, these freestyles are the raw material that would be fed into alignment comparisons, then passed to Rosetta, to generate the prediction puzzles, that we were used to. So we had the opportunity to add our human touch, earlier in the process. And we found, that freestyles had their own problems, it would take 2 or 3 days to get even a decent starting structure, and you might get lucky. Or not.

Alignment tools, are just the next part of the Rosetta process that we have been given the opportunity to adjust with our human eye. They are the gift-wrapped head-start that saves us those 2 days of processing freestyles, and they are now far more malleable.

So, does scripting help or hinder the human element?. You might argue from a Purist point of view, that it’s encroached too far into the human part of the game, that our only creative contribution now, is setting up the start-game, before letting loose the uber-meta-super-script to do its magic. You might argue, that we’re simply fulfilling the job-description of Foldit, by striving to replicate a humanised prediction algorithm from scratch. It’s the SkyNet paradox - will we one day fold ourselves out of a job?

The answer, lies somewhere in between. Scripting was supposed to augment Recipes, by giving more range to things you could do. It can automate certain repetitive tasks, it can refine rebuild into a valuable tool of exploration, it cant make the perfect of tea. The danger is, it can turn folders into users. If you choose to play the game, as a point chase, then be a user. Download the highest scoring scripts, build yourself an i-10, and good luck. If you want to be a folder, try to understand by hand, firstly, why you’re doing what you’re doing. The secret to scripting, is yet again the human element - knowing WHEN to run a script. AS BP said earlier, the playing field is a lot more level than you may believe it is.

Speaking personally, I spent 3 or 4 months finding out how things work, to understand why it works. When I felt I had something to add, I joined a team. I’m not a programmer, I use recipes and scripts to automate the dull stuff. I’ve no desire to spark life into SkyNet, I’m a firm believer in the humanist approach. If you care to look Brick, my team has 5 Perpetual Motion Machines. Our achievements are based on hard work, and the experience that it brings. It’s a little petty and disingenuous to suggest that it’s merely down to a few lines of code. If you believe in the magic passport of scripting, Brick, try ‘editing’ your downloaded scripts, and ‘save as’ under a different name. You can then edit the contents to your hearts content. Try building a quake variant library, by varying where you place your bands...

Personally, I think we’re already in a position where useful scripts and recipes, as distilled packets of experience and information, are probably used by the vast majority of players, so the notional advantage is already diluted beyond a competitive edge. As FoldClub continues to evolve, the most important thing to remember is the original message. The biggest weapon you bring to Foldit, is you. Why choose to dilute that advantage, by downloading scripts that other people have written? All you’re doing is playing your game in their way. Take the time, to explore and find your own way.

Charlie

Joined: 03/20/2010
Ideas: Manual Folding Master Class Training & Game Addition

Hello Charlie,

Let me start by saying that, in my opinion, your "2-cents" ranks among the best of the responses to this thread so far, (right up there along with BP's). So much so that I am hesitant to add any further comment, as, in many ways, you've tied up the whole thread of responses to Boots' original post into a neat little package that seems almost complete.

...almost.

The issue I have with most of the responses (my own original response possibly included) is that while Boots' original posting posited the question about "information," somehow, most of what followed subtly became subverted to the question of one specific form of information: ...that being "scripts."

Now, while that is a valid interpretation, about which many have strong opinions, and while it is certainly an interpretation deserving of all the attention and response it has engendered in this thread, I believe it is now very important to refocus whatever may remain to be said in this extended conversation to the original focus laid out by Boots in the beginning: the broader topic of "information." In particular, I'd like to broaden "information" to emphasize an almost opposite aspect of Folding from Scripting; Manual Folding Techniques.

Being a player/participant who came to the game only lately (started only on March 19th, 2010), I have quite naturally been "funneled," as have many others, (by timing and necessity as well as inclination) into the "script" game. However, from quite early on, I also began to realize that there were levels of knowledge regarding the game that were, through natural evolution, no longer being taught or emphasized, and that were rarely even discussed, at least publicly, in the chats. After the initial tutorials, "manual" folding seems to have all but been forgotten. I have come to be of the opinion that script-folding, with its easy-to-achieve leaps in scoring performance - which is very seductive to newcomers, (myself included) has overshadowed, and even replaced many crucial lessons and fundamental understandings that are necessary to succeed at a higher level later on.

In a nutshell, I fear that "manual" folding skills are sadly on their way to becoming a lost art. I also perceive that this art is critical to a player eventually maturing into a really great folder, and not just a script runner, and I am increasingly aware with each new puzzle, how much this lack is preventing me from progressing to a higher level of Fold-it success.

While it is a tribute to the game designers that they were able to design the game in such a way that newcomers can begin to do quite well quite soon (by discovering and putting to use all the years of experience and expertise of others, which is encapsulated in the scripts in the public script library), it is also amazing to me that so many of us - myself included - can get such a feeling of accomplishment; of competence; of success; and such a "thrill of victory" - by simply utilizing all this work that was done not by ourselves, but by others.

Eventually, however, the realization must come... "I did not really do this on my own merit." I am merely re-employing through script, the accomplishments and knowledge; the "blood, sweat, and tears" of others' efforts.

And those of us to whom this realization comes, and those of us to whom it is also important to achieve success on our own merits - then face a choice: to either learn enough to begin to be able to succeed on our own merits, or to accept that all of our "achievements" in scoring within Fold-it are largely illusory; not really our own achievements at all.

I choose to learn - as I suspect most if not all of the other new players who catch the bug and stick around - also do.

If I am jealous of anything in this game, it is not that others may have better scripts than I do; it is that I missed out on the beginning of Fold-it, when the True Art of Manual Folding was everything, and the experiential foundations were being laid for today's folding elite.

And so, for me at least, it is not the scripts of the best players which I covet nearly so much as simply their knowledge of and skill - at manual techniques - forged through long hours of study and practice - back when there was no other way.

It is unfortunate for the newer players, that some of the "old guard" are reluctant to teach newcomers more than just the basics of such things. Many are incredibly helpful and magnanimous about helping newbies with low-level, basic Fold-it function questions, but some, when asked a question like "How did you do so well on 'such-and-such' type of puzzle - (where I could get absolutely nowhere)," ... become vague if they answer at all. I have even been told in chat that "I'm sorry, I cannot help you - someone on my team is watching." In one respect, this is very sad, that even within a competitive game, an experienced player cannot even answer an earnest question posed by a newcomer with a sincere desire to learn, for fear of incurring the wrath or disapproval of a team-mate. It has the effect of making one stop asking questions at all, at least of that person. However, on the other hand, I must add, I have come to realize, it is also quite understandable in the competitive team environment, to guard one's team's advantage, and I hold no judgment toward that person for doing so. For all I know, no verbal explanation would have been adequate anyway, to convey the knowledge I sought.

So what are we newcomer's - we "have-nots" of old manual-folder experience - to do? Simply resign ourselves to our fate of not having arrived soon enough to garner the same education within the game as the older players? Continue to ask questions of those whom we know must rebuff our inquiries due to team loyalties? Attempt to achieve Fold-it Zen Master competence through scripts alone? (that latter being a contradiction in terms)

What I long for is a Manual Folding Master Class, where the Master Manual Folders may pass on the benefits of their knowledge to the "new guard," in apprenticeship fashion. Now that the game has advanced and progressed so far down the scripting path, these manual skills may not be transmissible/preservable by any other means.

But, I must admit, perhaps this role is inappropriate for ANY player in a competitive arena. Perhaps even attempting to answer a question such as the one I mentioned above really would be inappropriate for any team-member with loyalties and allegiances, oaths and pledges to his or her team. Perhaps this "Manual Folding Master Class" I envision and long for can truly only appropriately be put on by the Developers who are the only ones capable - who are also outside the competition. And by "Master Class," I mean something that goes FAR, FAR beyond the depth and scope of the the current tutorials.

Developers... are you reading this? You may be the only appropriate avenue for the dissemination of this type of information and skill enhancing training. I really believe at this stage of Fold-it's development, it is a cogent question on an important subject, and I would be surprised if there is not a lot of agreement on this out there amongst the newer folks in the Fold-Club membership. With all the new developments, has there perhaps been some omission of valuable training that only those fortunate enough to have gotten in on the ground floor will ever be able to receive?

Let's start with the assumption that the answer to Boots' original question here - ("How much information are we expected to share with the rest of the Folding community?") - is "None." For the sake of this conversation, let's posit it as if it were LAW: "No player shall be expected to divulge to any other player any piece of knowledge or experience regarding the game, especially where such information may profit the recipient to the detriment of the dis-closer."

How then, can the newer players come up to speed with the techniques that so enhance the older players games, but which are now held as closely guarded secrets (and perhaps rightly so) within the individual teams, where sharing of these "secrets" is (perhaps also rightly) considered tantamount to treason? ...techniques which are also not being as fostered within the game and it's training as perhaps they once were, due to the evolution of the game, the availability of scripts, and the easy successes that scripts enable.

An Idea: Perhaps there should be a new class of game for which scripts are not even available in the interface; puzzles which can only be affected by manual interaction. Perhaps there should also be a separate scoring division; a separate competition, where only non-script puzzles are scored completely separately; a whole separate arena where players can learn and develop manual skills without being trampled by high powered script folding. This, in addition to a the "Master-Class" folding tutorial would provide both the education and the "training ground" for players desirous of achieving the deeper understanding possessed by the older players; It may even attract the interest of older players as well, who may be desirous of testing their own manual skills against others in an arena that is made level by the exclusion of scripting. In fact, to go a bit further, I'd add that the inclusion within the game of a script-less, manual-folding only competition, would be the key element to the success of any master-class manual folding training that may be implemented - an arena to practice and hone these skills without being blown off the map by scripts.

In my humble opinion, this would be a very good thing which would attract a lot of interest and support from all those in the Fold-it Community who are hungry to increase their folding knowledge and skills, and those who are hungry to re-learn and preserve the manual folding "secrets" of the "old guard" or "elite" folders.

Am I wrong in this thinking? Have I overlooked and/or misunderstood something? Players and developers alike, I'd love to hear your opinions & feedback. Is another level of Fold-it training doable and practical? Am I the only one who perceives a need and a desire for this among the other newer players ... and even the old ones? Would it even be Possible to add a "scriptless" game arena to Fold-it that disabled script usage?

Please forgive my loquaciousness... this subject has struck a chord in me that has gotten me quite carried away - that, and perhaps way too much caffeine.

Best Regards, and Good Folding to all

-dimension9

infjamc's picture
User offline. Last seen 2 years 45 weeks ago. Offline
Joined: 02/20/2009
Groups: Contenders
Re: dimension9

As yet another Contenders member who has been playing Foldit for 1.5 years, here's my perspective:

* I've always been playing mostly by manual methods-- after all, neither recipes nor LUA scripts were available back in February 2009. In fact, even now, the only recipes I use are the simpler ones-- a few for backbone walking, banding/compressing, changing band strengths, and side chain tweaking (e.g. Blue Fuse). This might be part of the reason that I've only accumulated 500k moves despite having started playing earlier than some of my teammates.
.
.
* Why do I choose not to use complex scripts? To put it bluntly, because it doesn't suit my playing style. Plus, my impression is that I probably wouldn't benefit from them anyway-- I've heard more than once from teammates who have ran scripts overnight only to gain one point (which is not cost-effective considering the extra carbon output, in my opinion). Furthermore, some of my tactics would require coding an AI if one wants to automate it completely-- what I call the "side chain scramble" is one of them. (In a nutshell, it's similar to blue fuse in concept in the sense that it involves wiggling/shaking at different clashing importance settings. An important difference, however, is that it also requires manual side chain matching, usually with the recent best loaded as a guide and with coloring relative to guide turned on.)

==> Does my not using long, complex scripts put me at a disadvantage? From my experience so far, no. In fact, there have been more than one case where I've shared tips with others (mostly newbies, but occasionally with veteran players as well if I'm so far ahead in a puzzle that people are jokingly complaining about it), and even then the net effect on my rank is minimal. (The idea is that, if there's a good chance that I'll be passed sooner or later, might as well help the eventual #1 player so that the top solution will have an even higher score and, hopefully, be closer to the native.)
.
.
The tricky part, though, is that tips from manual playing seem to be more difficult to teach-- with a script, all one has to do is to share it and have someone else re-download it. A good example of this is Puzzle 165 ( http://fold.it/portal/node/986452 ):
.
.
* The description of the puzzle specifically said that the Rosetta decoy (trial structure) "has one major error" and suggested that people experiment with the then-just-released behavior tab. When I loaded the puzzle and found that it had no secondary structures labeled, I immediately turned on loop hydrogen bonds and made my best guesses of where the helices and sheets were. One thing that stuck out to me is that one relatively straight section, which I assumed to be a beta strand, was not forming hydrogen bonds... so I tried swapping two beta strands and ended up getting a top score. The tip was shared with others later, but to my dismay, all it did was reducing my lead to 30 points (I was hoping that other better folders would surpass my score given the hint .

==> Well, I was later contacted by the Foldit staff about this because my structure was apparently way too close to the native. Clearly, I was very lucky, because there are many possible permutations for swapping two beta strands; and even if I guessed correctly, it is not guaranteed that I would be able to put the protein back together the right way. The ironic part, though, is that the same tactic didn't work quite as well in latter puzzles where I feel that a sheet doesn't look right. In fact, when the same structure was re-released as a quest to the native in puzzle 176 ( http://fold.it/portal/node/986597 ), I only got 8th the second time despite that the native was in the game for copying at will.

Moral of the story:
1. While you can't depend on luck, sometimes good instincts and manual folding techniques can still prove pivotal when given the right situation; and
2. Considering the results from 176, one could conclude that cheating (in the sense of copying from a native that one finds elsewhere if it's not given in the game) probably wouldn't help in most cases.
3. Corollary to #2: However, extra tools, scripts, information, etc. MIGHT help in the hands of a more skilled player (why else would aap and several others end up with an even higher score the second time around when the native became available?)
.
.
==> With all of this in mind, I think the underlying question isn't whether scripts should be shared, but HOW the knowledge related to folding in general that complement the existing tools could be shared. After all, even as someone who doesn't mind about sharing knowledge in foldit even at the expense of his own rank, my experience is that some of these things can be be difficult to teach (sometimes even to Top 10 players).

Joined: 06/17/2010
Little addon from me :)

I found Fold.it by accident, in Rosetta project on Boinc I see strange button :)
I like puzzles, so after download Fold.it eats majority of time I spend on my screen now.
Because I`m home-made programmer scripting pulls me in. I see many ways to simplify/quicken some moves we do in game - thats are scripts for. So I share my ideas, and read as more as I can form others.
ATM I only missing "basics" of manual folding - what type of aa takes place as sheet or helix. I tried to analyze two natives we have atm, but results are not good (see my shared script... or better not :P).
So now in game I use one of template given and tune it - shorter/longer sheet, helix instead of loop etc. If looks good then scripting it.
I think what is set at start, the "original"/manual fold b4 scripting determining final result. Faster PC helps only to try more starting points.

Back to topic: sharing knowledge and scripts is very important. Competition makes it harder :D
I think we need few videos (on youtube for example) witch good comments where we (newbs) see how to start fold.
Also sharing best scoring and/or looking results for closed puzzles would be nice -as target practice ("Oh my! How I missed that!").
More "human language" articles would be nice too.

Susanne's picture
User offline. Last seen 4 years 20 weeks ago. Offline
Joined: 11/09/2008
Groups: SETI.Germany
haves and have nots

Interesting arguments given above and I would like to ask those who can write recipes to please share them with the folding community as they improve a protein beyond what would have been possible only a short time ago and that is good for R@h. Otherwise there'd be 2 tier folding: --> the haves and the have nots.

My personal preference is to fold the proteins manually from the start which is a method I adopted when I started out just over 18 months ago. My method is freely available in the German Wiki) but I need the recipes to work on the protein once a certain stage is reached just to see where it ends up and will then work on it manually for a further few hours etc.

I can't write recipes so am dependent on those shared with others. There are no special recipes in my group and only sheer determination, some intuition, a faster computer and lots and lots of my time go into each protein.

So to those who want to guard their script(s) I would like to say: - keep it, I don't need it
and to those who freely share with all of us, I would like to say a big THANK YOU :D

Joined: 09/18/2009
Groups: SETI.Germany
One Point...

If I write a recipe, I always upload it with default parameters.
I think, there are many similar recipes, but with just small changes.
"Publishing" all, also the derivates, would fill the recipe download section, making it hard to find the originals again.
I guess it would be good to have not only a regular (general) recipe section, but also one per puzzle.

Tlaloc's picture
User offline. Last seen 10 weeks 4 days ago. Offline
Joined: 08/04/2008
Groups: Mojo Risin'
Recipe parameters

Let's take requests for features to a separate posting or feedback. This discussion is far too interesting and important to get derailed by feature requests.

When I publish a script, it is a way to package the hours that I've spent learning about folding. Some people pass on their knowledge through chat, or wiki postings, either in private or public. I pass my knowledge in scripts. I can't do everything I know in scripts, because the lua features available are too limited, but I can do a lot. My knowledge is enough that it can get someone to about 15 to 30 in rank right now almost exclusively through running my scripts in the right order. (BTW, I can't take sole credit as my scripts also include code such as vertex's bluefuze and co lapse's settle, which I include because there is no way to call their code.)

If someone treats my script as a black box, just running it with no understanding, she will get the rank. However, if you look at what the script does, it points out techniques that acquire points. Most of my scripts use banding in ways that I developed in the early days of folding before scripts existed. Some are techniques that I wanted to do in those days, but were too tedious to do each time. In the description of the script and on the web page, I try to give a general idea of how the script works. If you want to get the exact technique, it is right there in the code.

I am most appreciative when someone else publishes a good script (thanks vertex, co lapse, spvincent, pletsch, madde, rav3n_pl). I define a good script as one that finds me points after I've exhausted everything I know. It gives me a new technique. Why didn't my scripts find those points? I look at the code and see what those scripts do, and I learn something. Sometimes I can take what I learn and form new scripts that use those techniques in new ways.

Do I hurt your rank when I publish a script? I hope so! I am spreading knowledge to your competition on how to fold better. If what is in my script was one of your 'secret techniques', then your competition just became able to use that technique. However, if that technique wasn't in your set, then both you and the competition will get better scores. Can you apply the script and the technique behind it better than she did?

Scripts raise scores. I am happy about that. When the top 5 people are nailing the native structure every time, I think you can complain about a level playing field, and this game will be finished having accomplished its goal. Until then, there are many things to learn about folding, and I hope to take each one of them and put them into a script.

Joined: 12/14/2008
very intersting discussion

How unfortunate I haven't been here for so long.
This post is mostly answers to dimension9's first one:

Winning is fun, but it is more fun to share and especially to see someone grow with your teching who becomes even better than you.

But for the question of sharing:
What would happen if all is shared? There would be tons of scripts and tips. Nobody would be able to assess them all or to use them exactly or better then yourself (when you would guard them). So keeping them is now very effective in this sence ;)
But of course, an evolution would start and go on to advance the game. Why am I so sure about this?

I always said that foldit is like the game of Go, so I will explain it with Go ;)
1) In Go it is normal to speak about a game after you have played it with the other player (=the enemy). This is a crucial part of becoming better in the game. Nobody can teach you more than your enemy. And of course, the saying goes: the teacher learns as much as the pupil.
People are even replaying games from people that are dead for a century or longer.

2) In the history, there were different schools of Go, finding new ways of play (=scripts) for certain situation. They kept them for themselves - at least until played in a game against another school. That gave them an advantage for a cetain time. But that doesn't mean that they were the best.

The most famous game in the history of Go is (at least from the anime "Hikaru no Go" on) the game with the so called "ear redding move" http://senseis.xmp.net/?EarReddeningMove between 17year old Shusaku and old Grandmaster Gennan in 1846. (read it now!)
The "taisha joseki" mentioned was very new at that time. It was Gennans special script you could say. Shusaku played it wrong, but in the end the youngster won. That because of a move that can't be expressed in a script. That was pure human genius.

Today the Pro's games have evolved beyond that of this time and it is still improving. And Go is one of the most friendliest Games in the world - because of the sharing.
And also, winning is not the main point in Go. It's the fun you have to play against a players that is as strong as you or better (and Go has an easy way to handicap, so even players with quite different strenght can play together).

Concerning the ranks: Go has a very clear ranking system - but that is mostly a tool for a) comparing the strength of two players to find appropriate players and b) to make you happy if you go up the list. For most players it has not much to do with "I am better, **** you!".

And by the way: Every good human wins every time against a Computer in Go. Whatever the reason...

[quote]If so, then the apparent prevalence of largely computer crunched proteins calls any such claim into question.[/quote]
If you could get one million people to play foldit 24/7 then it would be so.

[quote]What I long for is a Manual Folding Master Class, where the Master Manual Folders may pass on the benefits of their knowledge to the "new guard," in apprenticeship fashion. Now that the game has advanced and progressed so far down the scripting path, these manual skills may not be transmissible/preservable by any other means.[/quote]
100% A teaching ladder would be a great thing. And very important, because I too fear that newbies aren't playing because the have nothing to learn from.

[quote]The tricky part, though, is that tips from manual playing seem to be more difficult to teach-- with a script, all one has to do is to share it and have someone else re-download it. [/quote]
100% here, too.

One last point: Cognitive Science has shown: giving is more rewarding then keeping for yourself ;)

itskimo's picture
User offline. Last seen 7 years 12 weeks ago. Offline
Joined: 08/10/2010
Groups: foldeRNA
have open evos for people without groups so everyone can evo

:)

Joined: 11/14/2009
Groups: None
do share them!

Over last few months we spent large amounts of time analyzing the patterns of recipe use by folders. We discovered many interesting things and drew conclusions. One of the conclusions we came to is that overall folders don't share enough. We want you to share more. To that end we want to provide an incentive for recipe authors to share their creations despite the potential for losing their competitive edge. We haven't figured out any specifics yet. So at this point, it would be interesting to find out your ideas about how we can encourage authors to share more.

B_2's picture
User offline. Last seen 4 years 25 weeks ago. Offline
Joined: 11/29/2008
Groups: None
That's an easy question to answer

ALL recipes should be freely available. Eliminate the Group "private" recipes.

As it is now, only a few teams participate in Evolving, so maybe the whole Evolver thing shoudl just go away. The Groups can still exist, but compete on the basis of the individual player scores.

As it is now, there are many players whose only contribution is to wiggle up a fraction of point here and there, and get highly ranked for that minimal effort.

infjamc's picture
User offline. Last seen 2 years 45 weeks ago. Offline
Joined: 02/20/2009
Groups: Contenders
Re: B_2, itskimo, and ilya.makedon

Regarding the "players are highly ranked for minimal effort issue": I've discussed this before the following feedback posts.

http://fold.it/portal/node/986077 (mentions the issue of picking up a cheap evolution from a XXX.8 or XXX.9 structure)
http://fold.it/portal/node/986220 (the calculation for evolver scores are incorrect, in the sense that the denominator takes in the number of soloist players rather than evolver players)

----------------------------------------

As for the idea of making evolver solutions available to everyone: this has been tried before in the form of "all hands" puzzles, where the top solutions that are sufficiently dissimilar in structure are automatically uploaded every 2 hours or so. What the developers of Foldit have found, however, is that these puzzles actually *discourages* having a variety of solutions because most people would simply focus on only the top solution. Hence, as counter-intuitive as this might sound, having many different teams limit themselves to evolving only their own structures is actually better from a scientific standpoint because a more diverse set of structures would result.

==> Given this, a better idea might actually be setting caps to the group sizes... because the marginal benefit to adding an extra group member will approach zero at some point.

----------------------------------------

Finally, regarding the issue of encouraging sharing: here's an idea.

One possible way to give players credit for sharing their recipes/scripts is to award more points (which could be a category separate from global soloist and evolver points). The difficult part, though, is enforcement-- what does that stop people from creating a script that consists of only global wiggle and global shake, which will often generate a lot of points at the very beginning when given a bad starting structure? To protect against such abuse, a possible workaround might be awarding points to the automated methods only if they prove pivotal-- for example, if they raise a solution's score from the 75th percentile to the 99th percentile. This requires extreme amounts of extra work, though, as it will require tracking (and analyzing!) score vs. time for every single player; plus, what's "significant" is subjective. But there's always the option of manually awarding achievements to individuals and/or teams after the fact for scientific breakthroughs made possible by automated methods-- for example, achieving an RMSD below 2 angstroms relative to the native structure.

Joined: 02/24/2011
Groups: None
cooperation and competition


Elevator summary: diverse strategies make the search better.

the point of the game to me is to create a hybrid parallel search of the structure space. Cooperation and competition are thematic ways of characterizing different search strategies. Parallelism is the point. If too much correlation between different approaches then you get herding. herding isn't evil but it doesn't span the space very well - it makes search threads cluster.

The only thing Mao is much good for is the quote "Let a thousand flowers bloom."

Joined: 02/24/2011
Groups: None
a further thought


Elevator pitch summary: the open source model of sharing might be helpful.

one might adopt a practice similar to open source licensing - like gnu - for example that permit anyone to use certain materials but that requires such uses to inherit the open source character too. Under that model anything I develop solo is subject to my discretion in the matter of publishing, but anything based on someone else's published work would need to be published too. A drawback might be tweak spam though.

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