Developer Chat

SethCooperIRC hello 14:34
Renton Grenoble 14:34
Renton hi seth 14:34
Tamison hi Seth 14:34
auntdeenisland hello Seth 14:34
BletchleyPark Hi Seth 14:35
TheGUmmer hi 14:35
ZiiONIC Hello Seth 14:35
CFC hi 14:35
YoyoParis Jean Legras 1914 14:35
uwezwidtmeyr hi seth 14:35
TheGUmmer Hi Seth 14:35
spmm Hi 14:35
*** anthioniphone joined #global 14:35
*** Renton quit (Read error: Connection reset by peer) 14:35
CFC [hey tam] 14:36
marie_s hi 14:36
thomirc hi seth 14:36
rav3n_pl (5 past) 14:36
SethCooperIRC I hope everyone is doing well 14:36
YoyoParis you know what I'm happy 14:36
*** Dane2010 joined #global 14:36
SethCooperIRC so recently it seems there have been some issues coming up 14:36
*** anneromaine quit (Read error: Connection reset by peer) 14:36
*** Renton joined #global 14:37
CFC really? 14:37
auntdeenisland issues? 14:38
SethCooperIRC well, it seems there's some discussion of evolver thresholds, and cloned accounts 14:38
SethCooperIRC is that correct? 14:38
thomirc Yes 14:38
auntdeenisland yes 14:38
rav3n_pl yap 14:38
anthioniphone Also... 14:38
auntdeenisland script useage is also becoming an issue in different ways 14:38
*** YoyoParis quit (Read error: Connection reset by peer) 14:39
SethCooperIRC okay, so we should discuss evolver threshold, cloned accounts, and script usage, correct? 14:39
TheGUmmer indeed 14:39
Renton Yes Please. 14:39
auntdeenisland yes 14:39
thomirc I would also like to discuss client crashes 14:39
*** krulon quit (Read error: Connection reset by peer) 14:39
Origami314 Oh yes, I crash all the time. 14:40
*** zoran joined #global 14:40
SethCooperIRC okay 14:40
BletchleyPark and client speed, please also clarify what a cloned account is 14:40
auntdeenisland me too - but is way of life on Macs 14:40
anthioniphone <-- Mac user 14:40
SethCooperIRC okay let's start with the evolver threshold 14:40
TheGUmmer 0.001 is to low 14:41
rav3n_pl haha 14:41
auntdeenisland agreed 14:41
SethCooperIRC people think this should be higher than 1 point? 14:41
rav3n_pl yap 14:41
SethCooperIRC is 2 enough? 14:41
Renton yes. its time to make evolving harder 14:41
zoran anyone know if evolvers would complain? 14:41
spmm any wakler will get you 2 14:41
TheGUmmer if i save 100.999 , its evolved at 101.000 14:42
SethCooperIRC right, so 2 would fix that problem 14:42
*** zmf129 quit (Read error: Connection reset by peer) 14:42
rav3n_pl it depends on puzzle imo, when we "always" have over 10k then 2 pt is too low, 5 woulb be better. bun on puzzle where we have top score aobut 600pts... 14:42
gurch then dont save 100.999 :) 14:42
Renton what would 2 points be enough? 14:42
gurch get it to 101.000 yourself then save 14:42
BletchleyPark on harder puzzles 2 is quite hard already 2 is enough for me 14:42
zoran perhaps it should be the ratio of the current energy. 14:42
Origami314 base it on percent from zero? 14:42
Renton it used to be an incentive to attract evolvers to play that game Gurch 14:42
TheGUmmer that doesn't help them group 14:42
TheGUmmer the group 14:43
*** rshin827 joined #global 14:43
spmm people are chasing evolve points not actually evoing the puzzle 14:43
Renton energy would be better, are there breaks that would mean better evolution of puzzles for people to attempt 14:43
CFC Can i ask why what prompted the core devs to come online? 14:43
Tamison Whatever the threshold is set at, the playerbase will adapt 14:43
SethCooperIRC CFC, we just wanted to try to address these issues 14:44
Origami314 I would like to request a red star next to the score instead of a red line through it before it has been evolved. I have dyslexia and numbers are hard enough for me to see. :( 14:44
spvincent I think to evolve you should have to make a meaningful change: local wiggling in itself shouldn't be sufficient. 14:44
ZiiONIC and i'm sure we all appreciate that you came online 14:44
gurch yeah I find the line distracting too 14:44
zoran we can set it to 3 or 5, but the question is at which point will nobody ever want to evolve anylonger. t 14:45
*** rshin827 quit (Read error: Connection reset by peer) 14:45
SethCooperIRC okay changing the line is pretty easy 14:45
Renton I agree spvincent, you have seen this game develop for a long time, 14:45
*** wld333 joined #global 14:45
auntdeenisland Seth & Zoran - what function do you see as evolving serving? 14:45
zoran evolving is highly beneficial to progression of foldit, so the solution must still respect this modality 14:45
gurch I just tied my protein in a knot :( 14:45
SethCooperIRC right, but the threshold is mainly so you can't just open a solution and get credit for it 14:46
thomirc is an evolution a function of the group? I.e. you must evolve beyond any other person in your group? 14:46
thomirc (i know how it is now) 14:46
SethCooperIRC no, it's not necessarily realted to the group score 14:46
SethCooperIRC related 14:46
zoran collective intelligence is always greater than sum of individual efforts. evolving is an important part of this. It also enables novices to be involved in improvement of challenges sooner 14:46
thomirc i'm asking - should the evolution be tied to other evolutions in your group (to prevent evolution trains) 14:47
*** justmissliz joined #global 14:47
auntdeenisland So - your concept of the value of evolving is really at the lower scores, not the higher ones? 14:47
thomirc where multiple people evolve the same solution 0.001? 14:47
*** krulon joined #global 14:47
SethCooperIRC oh i see, it has to be a better score that your whole group before getting credit? 14:47
OmarsTent yep 14:48
rav3n_pl hmm 14:48
CFC that's a beautiful point zoran, but is rapidly becoming redundant, in the current structure of FoldClub 14:48
Tamison That would make it much more challenging, interesting 14:48
zoran evolution train is good. many people evolving single solution by less than 1 pt is not good. 14:48
SethCooperIRC I think that may be too restictive, making it too hard to evolve 14:48
rav3n_pl that way one good player can kill entire group... 14:48
Renton 5 points. 14:49
Origami314 Yes 5 points 14:49
*** uwezwidtmeyr quit (Read error: Connection reset by peer) 14:49
*** JasperD quit (Read error: Connection reset by peer) 14:49
spmm 5 is good for me 14:49
*** justmissliz quit () 14:49
SethCooperIRC it could be cumulative: fist you need one point, next evolution of that solution needs 2, then 3, and so on 14:49
rav3n_pl 0,05% :) for 10k it makes 5 pts 14:49
spmm also discussion of a n open evo - people could put in lower scoring solutions and benefit newbies and soloists 14:49
OmarsTent 1 full point decimal to desimal on each pass person to person 14:50
zoran what if the amount is related to the time maturity of the puzzle. in the beginning you need 10 pts, in the last hour 1 pt is good? 14:50
Tamison percentage would make sense if it is feasible 14:50
Renton Zoran, nice idea ! 14:50
thomirc or the ranking of the person 14:50
*** SuperBEEP joined #global 14:50
*** Savas joined #global 14:50
TheGUmmer i like it 14:50
spmm cumulative is ok too - rank should not count 14:50
SethCooperIRC the difficulty with percentage is that points get harder the higher the score gets... 14:50
Renton remember when we first started folding, 1 point in an hour was hard work but we all compeated for it 14:50
*** jflat06 quit (Remote host closed the connection) 14:51
auntdeenisland personally like the cumulative best 14:51
spvincent I rather like the idea of the threshold being dependent on days till expity also. 14:51
*** jflat06 joined #global 14:51
spvincent *expiry 14:52
CFC pesky hoomans will always find a way around the rules... 14:52
OmarsTent somehow needs to be based on talent 14:52
Tamison That makes for exciting afternoons 14:52
*** jflat06 quit (Remote host closed the connection) 14:52
rav3n_pl it makes sense. week - need 7 pts, 2 days left - 2 pts, last day -1pt. 14:52
BletchleyPark what would be in the nest interest of the solution ? 14:52
zoran CFC: yes, that is the case 14:52
BletchleyPark best 14:52
SethCooperIRC it seems like 5 points had quite a bit of support 14:52
spmm perhaps we can try a few variations for a month 14:53
OmarsTent also somehow prevent conga lines 14:53
spmm surprise us :-) 14:53
CFC since evo is based on points, why not restrict every group to only have an n number of scoring evos? 14:53
SethCooperIRC explain conga line? 14:53
auntdeenisland It seems that the most fun evos - the most exciting - and the most valuable to foldit are the evos that start at the lower solutions 14:53
Renton I like your idea RAV, but im not sure how hard this is to actually inpoliment in the software 14:54
OmarsTent packing the hi scores 14:54
*** jflat06 joined #global 14:54
thomirc what i called evolution trains - many people evolving the same solution slightly in a group (pushing down other players) 14:54
OmarsTent with .001 evo 14:54
*** CanadaDavid quit (Quit: quit) 14:54
*** aendgraend joined #global 14:54
spmm instructions on the ADream wiki 14:54
CFC evry group, dependent on size, has an allocation 14:55
OmarsTent add 1 point for each pass between players 14:55
BletchleyPark or not dependant on size 14:55
Renton hi BP 14:55
thomirc if you DO implement the time based, please also implement the time remaining request (a.k.a. deathclock) 14:55
BletchleyPark huge groups will still push away small groups 14:55
BletchleyPark hi renton 14:55
*** jflat06 quit (Remote host closed the connection) 14:56
auntdeenisland sigh - that culture did prevail in AD for a while - those people are gone from AD - many in AD opposed that whole concept, there were many discussions - soloists very unhappy 14:56
Renton wed like the little groups to get into the evolving action more though too 14:56
OmarsTent also posiable to post with the red line and not add a point not right 14:56
spmm not the end of the world if one puzzle can't be evoed will learn what works as we progress 14:56
rav3n_pl yes, countdown clock is needed 14:56
Renton i think that was the purpose of the All In puzzles 14:56
BletchleyPark what is in the best interest of the solution (science) ? 14:56
*** DKH quit (Quit: quit) 14:57
OmarsTent shareing 14:57
rav3n_pl more exploring :) maybe evo need some exploration instead of "just" plain points? 14:57
thomirc http://fold.it/portal/node/986859 14:57
rav3n_pl additional req for evo? 14:58
Tamison purpose should be to learn from others and expand from there 14:58
zoran ok, we'll start with the simple thing raising the evo to 2 or 5 points. based on what we see, we'll adjust over time 14:58
SethCooperIRC i think we can try increasing the threshold for evolving and see what happens 14:58
SethCooperIRC we can always readjust after a while 14:58
OmarsTent ok 14:58
*** ericman314 joined #global 14:58
zoran moving on to next topic 14:58
SethCooperIRC okay next topic is... 14:58
Renton zoran seth, can you guys see from there the efforts people make in evolving other peoples puzzles then going to their own and making better adaptations to their puzzles? (ie evolving - looking at other peoples puzzles then adapting their own* 14:59
Renton *and making it work better* 14:59
*** kinglewi quit (Read error: Connection reset by peer) 14:59
BletchleyPark copying evo to solo you mean ? 14:59
SethCooperIRC yes we have seen that evolving other player solutions is useful 14:59
SethCooperIRC so, cloned accounts 14:59
zoran Renton: we'll do some data analysis to find out for sure 14:59
Renton well, loading a high ranking solo evolving it for team then going back to solo and then making that higher 14:59
Renton THanks Zoran. 14:59
*** infjamc joined #global 15:00
BletchleyPark ok, thanks for clarifying 15:00
auntdeenisland yes, cloned accounts.... 15:00
SethCooperIRC people are making multiple accounts and using to their advantage? 15:00
CFC cloned accounts benefit no-one, they just cause trouble. They always did 15:00
zoran can someone explain the way cloned accounts are used now? 15:00
auntdeenisland right now - that I am aware of - there are cloned accounts playing both on a team & solo 15:01
zoran and how do people benefit from them? 15:01
auntdeenisland and some outright multiple clones on some teams 15:01
TheGUmmer crowding out people at the top 15:01
TheGUmmer of the ranking 15:02
*** bog joined #global 15:02
auntdeenisland perhaps spmm can explain, he had something in feedback 15:02
SethCooperIRC so if the evolver threshold is raised will that help this too? 15:02
auntdeenisland no 15:02
zoran so they don't improve their own score, just lower others ranking? 15:02
CFC yes 15:02
auntdeenisland yes 15:02
TheGUmmer bingo 15:02
Renton running four clones by the same high ranking / experienced user will knock out four llegitimate players 15:02
BletchleyPark if they rank 10 on a solo, their clones will rank near that 15:02
spmm I have one to experiment if I have a high scoring solution - because puzzles do noy reset score 15:02
BletchleyPark and kick others down the ranks,, correct ? 15:02
zoran amazing that people would waste time with that 15:03
auntdeenisland the problem seems to be that they either don't understand or don't wish to learn the limitations of the foldit tools 15:03
TheGUmmer they run scripts, no time wasted 15:03
Renton yes, we have tried to explain that, although its interesting to hear those cloned persons reasoning. 15:03
auntdeenisland but some are just blatant and crowding others out 15:03
*** sixuns quit (Remote host closed the connection) 15:03
spmm ok why when I rest puzzle is my highest score still showing - it is very hard to work past 15:03
zoran is cloning bad for evo scores or also individual? 15:04
Renton i think at times clones are using other groups best solutions to fuel their own personal puzzles in another team also. 15:04
auntdeenisland the only reason that you should allow more than one account per ip is a one computer household with either husband/wife accounts or parent/child 15:04
SethCooperIRC how do they get the best solution from another group? 15:04
OmarsTent 1 clone for playing 1 clone for irc 15:04
*** ericman314 quit (Remote host closed the connection) 15:04
Renton they can print image, or see structure 15:05
TheGUmmer snap pictures of it 15:05
auntdeenisland other than one irc clone 15:05
BletchleyPark how can one tell if its a clone ? 15:05
SethCooperIRC how do they get the picture? 15:05
CFC it's the glittering prize of a point based game, zoran. Some do not realise the futility behind it 15:05
OmarsTent multi clones bad for everybody 15:05
*** bog quit (Read error: Connection reset by peer) 15:05
TheGUmmer the little camera icon beside send 15:05
CFC careful omar 15:05
gurch auntdeenisland: so people on shared/dynamic IP networks shouldn't be able to play? 15:05
Renton printscreen i would imagine, im not 100% sure although Brick or B-2 may be able to explain. 15:05
marie_s I really dont care 15:06
OmarsTent sorry 15:06
SethCooperIRC right, but how do tehy get the structure to take a picture? 15:06
auntdeenisland gurch - there will always need to be exceptions - those can be okayed by foldit 15:06
ZiiONIC clones affect both evo and soloist scores 15:06
Renton one player, two account, one is in team A the other Team B 15:06
Origami314 Two or more accounts in different groups 15:06
TheGUmmer log on one account, take picture, log on other acount , fold to picture 15:06
SethCooperIRC i see 15:06
SethCooperIRC even if we limit by ip I don't think it would be too hard to fake multiple ips 15:07
OmarsTent same with scripts 15:07
TheGUmmer or thats what i would do ;|) 15:07
zoran do groups known of members that are part of other groups? 15:07
*** smanteufel quit (Read error: Connection reset by peer) 15:07
auntdeenisland yes 15:07
ZiiONIC yep 15:07
CFC you really want to know? 15:07
spmm name and shame 15:07
zoran can they not be booted from the group? 15:07
*** jflat06 joined #global 15:07
gurch is this actually a problem 15:07
rav3n_pl just use tor/vidalia to get about 1000ips.... 15:08
auntdeenisland oh gee, zoran - that worked real well for us, lately in team AD.... 15:08
gurch i'm looking through the top 100 for puzzle 406 and I don't see any evidence of cloned accounts 15:08
OmarsTent score rank problum 15:08
*** msingh quit (Read error: Connection reset by peer) 15:08
Lazarus-uk it is not a problem for me, it is just unfair for people below trying to climb the ranks 15:08
CFC ...but this was always the way... now that AD have fractured, and others have jumped on the bandwagon, it's now becoming a flame issue 15:08
Renton thanks cfc, lets work on the solutoins, we know the problems 15:09
zoran if the group is open, you can't complain about people taking things away from the group. 15:09
spmm how do you identify them? 15:09
rav3n_pl only "good" thing to use clone for is irc - make second account to use it for gruop chat. 15:09
auntdeenisland AD has not fractured - we had 3 people leave - 2 voluntarily, One booted 15:09
Renton Zoran, do we actually own the recipes rights as a group? 15:09
CFC ...yes, open groups have always been open to abuse 15:09
BletchleyPark @renton i'd say any individual owns the copyright to his work 15:09
*** oxyi joined #global 15:10
Renton the nature of groups is to evolve too, and splits are not always bad. 15:10
spmm we are liding int oanother issue 15:10
OmarsTent dont have open groups? 15:10
*** ATLADAS quit (Read error: Connection reset by peer) 15:10
*** znagy quit (Remote host closed the connection) 15:10
spmm sliding* 15:10
gurch is there a disclaimer in the foldit eula that says "all ur recipe are belong to us"? 15:10
*** wld333 quit (Remote host closed the connection) 15:10
gurch if not then yes, its yours 15:10
Renton @BP ide like to imagine that. In this way, countrie law can be used to protect it as IP. 15:11
zoran can we discuss cloning for soloist score: any suggestions of how to address this? 15:11
rav3n_pl every recipe is sent do server anyway ;] 15:11
spmm why! no value to science just bragging rights for a person or team 15:11
*** ATLADAS joined #global 15:11
TheGUmmer get rid of group recipes 15:11
Renton i made one suggestoin once that should a team member of a group press the camera button, their solo goes to evolver play. *i think that sounded right.* 15:11
zoran technically everything created by the game is in the open domain so that no individual can take advantage for only their personal benefit 15:11
SethCooperIRC but it seems that the main issue is that people are joining open groups and taking things to other groups? 15:12
OmarsTent make all scripts public after 3 months better for the science 15:12
auntdeenisland the problem is the foldit structure - you set the team options - you allow scripts to be designated to be shared within group only. In a perfect world, no malcontent could wwalk away with 50 team recipes & share with new people - but within the structure, that can happen 15:12
auntdeenisland and did 15:12
gurch auntdeenisland: how can that not happen 15:13
TheGUmmer aslong as you can copy and paste in the script window there is no way to stop that 15:13
gurch auntdeen: if you can use recipes, you can copy them 15:13
SethCooperIRC right 15:13
zoran auntdeenisland can you describe what happened? 15:13
spmm Seconded GU - collaborative scripts - with points for scripters perhaps 15:13
Renton we all want to share our work with people in groups so we can teach them more one on one, when people just take them, that people using it has no idea why or what circumstances this should or shouldnt be used. 15:13
gurch points for scripters? 15:14
rav3n_pl yess! :) 15:14
TheGUmmer lol 15:14
*** bregnier joined #global 15:14
auntdeenisland we threw out the worst of the clone people - he went to another team - then uploaded 50 of our scripts for that team 15:14
gurch so instead of evolver trains we have people repeatedly changing 1 comment in a script :p 15:14
spmm why noyt points for scripters - 15:14
SethCooperIRC it seems like a solution would be to not have an open group, right? 15:14
Lazarus-uk why? 15:14
zoran we will eventually have scripter ranking. 15:15
gurch fundamental problem: people on the internet are not trustworthy 15:15
OmarsTent or release scripts after a set time 15:15
Renton well, we just ask those members not to take stuff that doesnt belong to them, but our team. 15:15
auntdeenisland Zoran - we are trying to track down scriptwriters to upload in global - but feel that we can't without their permission - their permission was to share in group 15:15
rav3n_pl and cookbook should clean itself from group recipes if player exits group. but in first, cookbook should be unbugged ;] 15:15
CFC yes, aot, open groups are open to abuse. 15:15
zoran i know i'm repeating myself, but i don't see any grounds for compalaint that something is taken away from the open group. that's the whole intent of the open group 15:15
Renton thousands of hours go into the scripts 15:16
Tamison Close the group, close the loophole 15:16
*** bregnier quit (Read error: Connection reset by peer) 15:16
SethCooperIRC right, even if we cleaned the cookbook, they can still copy/paste 15:16
TheGUmmer you can still copy all the scripts in your cookbook before you leave or are booted 15:16
OmarsTent yes clear team scripts from cookbook 15:16
auntdeenisland then - zoran - it would be good to note that on team options page - that sharing with group not an real option 15:16
TheGUmmer theat doesn't solve it 15:16
Renton wed like to have some kind of backup from you guys there to support us when we have or are being abused ( having our scripts abused* 15:16
Tamison My impression is that the most prlific script writers share globally 15:16
auntdeenisland the best writers do share most recipes 15:17
Renton correct Tam 15:17
OmarsTent yes thanks 15:17
TheGUmmer but keep best for the team? 15:17
Renton its good to have team spirit and team scripts to temp other new players to join us because we put the effort into learning them and how they work 15:17
auntdeenisland hoever - if novice writer, then wants to polish work first, so shares with team first - even experienced 15:17
zoran we can make it clear that any open group means that all findings of that group are available to everyone including groups. 15:18
OmarsTent yes 15:18
zoran if you want to protect a recipe but share with select few, make a private group 15:18
auntdeenisland or some feel that it was a collaboration with team embers that led to the recipe - so share only with team 15:18
SethCooperIRC I don't really see a technical solution to this other than closed group 15:18
spmm why have group scripts except for development - against the grain of for science 15:18
spmm soloist can also have scripts 15:19
Renton hmmm. is that the best solution Zoran, and what should be the protection if we then get the same thing happening via that, what can we do to protect it going from a private group? 15:19
*** jlowdermilk joined #global 15:19
zoran we just send a paper on the huge beneficial effect of shared macros to the conference, btw 15:19
auntdeenisland so where does a budding scriptwriter go for help with debugging? 15:19
TheGUmmer nothing 15:19
CFC the more intriguing issue from zoran, is the notion of how he feels that Foldit should shift back towards a human element... Scripting is too entrenched to be removed. How to push it back towards 'what can you do?' rather than the Fold@home 'what can you run 15:19
spmm collaborative competitive scripting 15:19
rav3n_pl in most cases I use group as debugging/testing "tools" (on offense i hope:) then release it for public. but we have some few group ideas discussed nad worked on it internally, not intended to be open. 15:19
*** lazystar quit (Ping timeout: 180 seconds) 15:19
marie_s he ask to rav (I do) 15:20
Renton ahh ok Zoran, hehe, thanks for that update. How does the paper look? can we get a draft copy please 15:20
TheGUmmer CFC - exactly like my feedback post 15:20
SethCooperIRC btw - is this the issue with scripting that we wanted to discuss from earlier? 15:20
SethCooperIRC just wanted to make sure we got to that as well 15:20
CFC hey GU... feeling like an 'old-timer' yet? 15:20
Renton *lol* 15:21
TheGUmmer little bit, i like scripting, i have a collection of miy own shared with the group 15:21
... missing... 15:21
auntdeenisland perhaps the player code could cover group scripts - or if not, then perhaps time for foldit to remove that designation 15:29
Renton you happy with that AD ? 15:29
rav3n_pl true 15:29
spmm yes identify the abusers not the people trying to learn 15:29
CFC ....trying to learn? 15:30
SethCooperIRC okay so we will basically rely on the players to identify who is cloning accounts 15:30
Renton <--- im trying again 15:30
*** eikem4 quit (Read error: Connection reset by peer) 15:30
Renton yes. 15:30
Origami314 Yes 15:30
infjamc Seth: some account names are very obvious, though 15:30
SethCooperIRC do you already have some idea of who these acounts are? 15:31
+++ Judecca has given op to SethCooperIRC 15:31
+++ Judecca has given op to zoran 15:31
*** Karel joined #global 15:31
Judecca i can always check my logs if anyone wants me to look into names coming from the same ip 15:31
SethCooperIRC okay 15:31
Renton that would be great Jude, ty. 15:31
*** pauldunn quit (Read error: Connection reset by peer) 15:31
spmm If yopu are going to boot my second account would you please respond to the feedback post which explains why I have it 15:32
zoran we should remove the obvious ones right away. pls send all suspected accounts to seth. we will also post news so that others know as well. if players can update the wiki as well it would probably help. 15:32
*** jflat06_irc joined #global 15:32
Renton i'll hand myself in. 15:32
SethCooperIRC yes, please send me a message on the site with further information 15:32
infjamc Now the tricky question... how to maintain due process 15:32
zoran spmm does our second account have scores very similar to your first account? 15:32
Judecca spmm: yours was for helping new players on the beginner puzzles right? 15:32
*** jflat06 quit (Remote host closed the connection) 15:33
infjamc We don't want to go to the extreme of assuming guilty until proven innocent either 15:33
spmm no Z and J - its when I get a high score and need to start again 15:33
Judecca oh yea. maybe that was someone else 15:33
TheGUmmer maybe a email warning 15:34
zoran the list will trigger our investigation, not outright removal 15:34
CFC ...LFB? 15:34
infjamc But imagine this: 15:34
TheGUmmer omg where 15:34
infjamc Players A and B have very similar playing styles, but A performs much better than B 15:34
infjamc Suppose that B hates A so much and wants to get rid of A's account... 15:34
Judecca that high score thing goes back to there being no way to reset or ignore the absolute best on an account. 15:34
Judecca short of deleting the solution file 15:35
zoran can player B play with player A ip address? 15:35
spmm that you J ;-) 15:35
infjamc When the Email warning due to a false positive is sent to B, B could claim (falsely) that it was indeed a clone and get A removed 15:35
BletchleyPark theoretically yes 15:35
spmm thank you * 15:35
*** iluvthek joined #global 15:35
TheGUmmer if the ip are the same or simimilar 15:36
*** iluvthek quit (Read error: Connection reset by peer) 15:36
SethCooperIRC well let's start with the potentially cloned accounts and go from there 15:36
TheGUmmer hear hear 15:36
zoran moving on to scripts 15:36
*** jflat06 joined #global 15:36
TheGUmmer 15:37
SethCooperIRC have we already discussed that? 15:37
SethCooperIRC or is it something different than before? 15:37
*** iluvthek joined #global 15:37
CFC ...I'm still curious whether seth and zoran will confirm that the last phase of FoldCentral research is focussed on scripts, and how we use 'em 15:37
TheGUmmer is foldit about the scripts or the scienece 15:37
SethCooperIRC both 15:37
SethCooperIRC the scripts are part of the science 15:38
Tamison The macros are just a faster way to solve the science 15:38
Lazarus-uk who said they were faster? 15:38
zoran ii've recently recruited tlaloc to add many more hooks in the code for scripting purposes, as well as have scripts towards diversification of the score more on that soon. 15:38
Judecca i think having a few script-free puzzles would provide some interesting date. from there i think you can see how its effecting things and if its something that should be further encouraged. 15:38
Tamison cool! 15:38
*** jflat06 quit (Remote host closed the connection) 15:38
zoran scripts are very useful for bootstrapping novices as well which is a huge benefit in the long run 15:38
CFC excellent... I've always wanted to play my game in his style... never had the balls to admit it 15:39
Judecca if it helps the science of things. i think some new players can learn from seeing the scripts in action 15:39
TheGUmmer I use them to foold 24/7 15:39
infjamc Well, it would be interesting if there is another measure for the likelihood that a structure is correct beyond the energy score 15:39
infjamc ...because one problem with scripts is that you could leave it running for hours, only to gain a single point 15:40
zoran CFC we'll send the first paper on scripts. one exciting outcome recently is that blue fuse type scripts are practically identical to the unpublished state of the art method used currently. we're yet to report on this. 15:40
Judecca now thats interesting 15:40
TheGUmmer sweet 15:40
*** Karel quit (Read error: Connection reset by peer) 15:40
auntdeenisland very 15:40
*** Judecca is now known as Judecca|vicodin 15:40
*** Ferdinandvbg joined #global 15:41
rav3n_pl cool 15:41
zoran any problem in term of protecting scripts in private groups? 15:41
spvincent II would like to see a mechanism foe deleting scripts. There's too much dead wood" in the global script lists. Old versions of popular scripts:: scripts that weren't much use anyway but someone thought they'd be useful, etc. 15:41
Renton Please send Tlloc and family our love ! 15:42
auntdeenisland zoran - what do you mean by private groups? 15:42
marie_s yes I want to dlete my old scripts and cant 15:42
zoran groups that are not AD 15:42
*** oxyi quit (Read error: Connection reset by peer) 15:42
SethCooperIRC you should be able to remove old scripts that you have shared... 15:42
Renton are (were we) we the only open group around? 15:42
rav3n_pl zoran: thy should be protected somehow. i`m not sure that clinet have ability to block run of group script when player is outside thet group. 15:42
auntdeenisland lol - you set 4 designations - open, moderated, invitation & closed 15:43
OmarsTent lots of groups have no script writers and depend on puplic scripts 15:43
zoran we'll add script deletion, but you can't remove scrips that have evolved from yours, naturally 15:43
*** jflat06 joined #global 15:43
Judecca|vicodin uh, might want to look into the invite system for groups. i think moderated or closed groups may not be able to use invites. or at least that might have been a problem in the past 15:43
auntdeenisland zoran - unless you have a closed group - it can happen to anyone under the current structure 15:43
TheGUmmer aslong as you can cut and paste a script to notepad it doesn't matter what the group tye is 15:43
TheGUmmer type 15:43
SethCooperIRC the group type does matter though 15:44
OmarsTent yes 15:44
SethCooperIRC in an open group anyone can come in and see scripts 15:44
TheGUmmer if some leaves they can take the scripts with them 15:44
*** Ferdinandvbg quit (Read error: Connection reset by peer) 15:44
SethCooperIRC but in a non-open group you needed to invite them first 15:44
TheGUmmer but things happen and they can still leave with the scripts 15:45
auntdeenisland no - under structure you now have - only a closed group that you can't get out of - will prevent theft 15:45
spmm why can't we share scripts? 15:45
auntdeenisland so all except closed could have this problem 15:45
SethCooperIRC right, someone can still leave, but you had to have invited them first 15:45
TheGUmmer but if you can't leave you can start a new account and still upload the saved script 15:46
auntdeenisland we had no idea that the malcontent was such a - um - malcontent 15:46
zoran seems like anything that is really highly useful is shared widely. Correct? 15:46
TheGUmmer no! 15:46
OmarsTent hopefully 15:46
spmm solists may have scripts also 15:46
TheGUmmer each team keeps the best for themselves 15:46
gurch i have a script im not sharing with anyone 15:46
OmarsTent yep 15:46
gurch it sucks though :) 15:46
CFC well, I guess I've booked my place at the end of the bar... that's a stunning admission, I'll no longer shout quite so much. 15:46
Renton gurch wanna share it? 15:47
Renton perhaps it works best on puzzles that other people make 15:47
*** aendgraend quit (Read error: Connection reset by peer) 15:47
Renton *grin* 15:47
*** jcarrizales31 joined #global 15:47
gurch it rebuilds then shakes then wiggles 15:47
gurch and keeps going round in a loop 15:47
gurch forever 15:47
*** HRMChatzibarbakarapapadopoulos quit (Quit: quit) 15:48
gurch eventually it finds something good. sometimes. mostly not though :) 15:48
infjamc auntdeen: even then theft cannot be prevented. Just create another account... 15:48
TheGUmmer like i said 15:48
*** Savas quit (Read error: Connection reset by peer) 15:48
auntdeenisland inflamc - I agree 15:48
CFC ...since the dremaers have found the way, I salute you and wish the luck. 15:48
Renton ty CFC ! 15:48
OmarsTent close if you can open if you cant control 15:48
auntdeenisland my point is that it is not possible for any group to protect scripts 15:48
SethCooperIRC yes, i think that's it 15:49
SethCooperIRC once someone is in your group they have access to things 15:49
zoran so the problem is once you take someone in the group, reneging later always means they'll walk away with scripts? 15:49
*** thom001 quit (Read error: Connection reset by peer) 15:49
spmm but no one team is ever at the top all time - imagine how much better we could all be if scripts were shared 15:49
*** SuperBEEP quit (Read error: Connection reset by peer) 15:49
*** DanielFoshag joined #global 15:49
TheGUmmer zoran - not always 15:49
OmarsTent yes 15:49
auntdeenisland so - there is no such thing as group scripts in practicality under the current structure 15:49
SethCooperIRC it means they can 15:50
Judecca|vicodin it'll be hard to implement all the protections everyone wants for their script 15:50
infjamc auntdeen: or to put it bluntly... 15:50
TheGUmmer share them all 15:50
infjamc All scripts are practically public... if you can reverse engineer them 15:50
Judecca|vicodin maybe something like the current format for solutions/pdb obscurity, for protected group solutions they can be a binary blob that can't just be read by a text editor. identify keys' etc. if you leave those groups the client can no longer read em. 15:50
OmarsTent thats the best for the science 15:50
auntdeenisland yes, inflamc - you are right 15:50
thomirc I think it's like MIT's Open CourseWare: The value is not in the scripts themselves 15:50
rav3n_pl k, all scripts shared, but then site should be changed to manage huge anmount of scripts - it need be like a tree structure, not as list like it is now. cookbook need that change to... asap ;] 15:50
Judecca|vicodin access control lists, giving read/write.view access. sounds like a pain.. 15:51
*** DanielFoshag quit (Remote host closed the connection) 15:51
*** iluvthek quit (Read error: Connection reset by peer) 15:51
spvincent A big part of scripting is knowing which script to run given the state of the protein. 15:51
TheGUmmer all open , easy 15:51
BletchleyPark i agree with spvincent 15:51
zoran we will instantiate the scoring for scripters. hopefully that will incentivize sharing. but people should be allow to protect scripts. 15:51
spmm share share share 15:51
*** thom001 joined #global 15:51
*** martinf joined #global 15:51
marie_s A big part of foldit is work between scripts 15:51
OmarsTent most teams dont have script writers 15:51
infjamc Those with better hardware would still be at an advantage, though 15:52
Sidran Points should be given for all kind of sharing. in a way that you cannot feel used if you shared something. 15:52
zoran yes 15:52
auntdeenisland zoran - it would be good for that to happen - simply to encourage sriptwriters to collaborate & debug 15:52
infjamc Any way to prevent newbies from being discouraged because their computers don't have the 8 cores to run a superscript effectively? 15:52
rav3n_pl yes zoran, some ideas are just fail, not all fails need to be public ;] add option to edit (delete or "reshare form server") option to site 15:52
marie_s yes but very difficult, script are made with others scripts,good scripts can give few points ... 15:53
gurch they can they just have to leave it running overnight like me 15:53
Renton woohoo thanks for spending some time with us today Zoran, Seth ! 15:53
Judecca|vicodin in one of the early videos of scripting there was a console where one could just type commands and they would happen.. where did that go?. seems nice for debugging and learning 15:53
Sidran zoraneee ;) 15:53
*** jcarrizales31 quit (Read error: Connection reset by peer) 15:53
zoran incidentally, scripting is also one of the culprits of not enough diversification. are any of you here part of the diversification challenge beta testing? 15:53
rav3n_pl yes! i want console back too! :) 15:53
spmm yes the contact map 15:54
marie_s the diversification reward unstable solutions at the limit of the conditions 15:54
thomirc zoran: The exploration? 15:54
auntdeenisland just started exploring that - enjoying it 15:54
infjamc zoran: I'll admit that I beat it with another giant pi helix 15:54
infjamc There has to be ways to prevent that 15:54
TheGUmmer tried it, don't quite understand it yet 15:54
Renton i think i made the first recipe. ide be worth mega points 15:54
zoran infamc: not likely in new version 15:54
Judecca|vicodin i was looking into those exploration/contact puzzles. nice idea. some really weird unstable proteins can get a good score. 15:54
*** monkry quit (Read error: Connection reset by peer) 15:54
rav3n_pl because they are far from "start fold" 15:55
infjamc Judecca: another problem is that wiggling at a clashing importance of 0.01 increases your score as well 15:55
Judecca|vicodin really changes the game, and will lead to some interesting scripts. once theres some good hooks for credit best/recent credit best 15:55
gurch except that presumably you don't actually want weird/unstable proteins for science purposes? 15:55
zoran marie_s: not once the score threshold is high 15:55
rav3n_pl also noticed, that sidechain position is not couned anyway in exploration. but tis is good :) 15:55
TheGUmmer got to go, thanks seth and zoran 15:56
marie_s zoran, the simple way is to be next the limit of conditions because no way to reset recent bst 15:56
rav3n_pl and yes, we finally need acces to scores from lua (recen best, credit best, abs best). Tlaloc posted it looong time ago... 15:56
marie_s to difficult to find very different and risky, gest lost 15:57
*** katling quit (Remote host closed the connection) 15:58
marie_s in the 2 modes, we have this problem of unstabiity of best scores 15:58
*** ATLADAS quit (Read error: Connection reset by peer) 15:58
*** Dolichwier joined #global 15:58
*** gurch quit (Quit: quit) 15:59
zoran we will introduce recent credit best. would that help? 15:59
thomirc In both client and LUA: yes, that will help 15:59
spmm aslo have to go thank you for your time 15:59
Renton thanks for input spmm 16:00
marie_s help yes but not resolve. In normal, you try very different, you will pass agreat time with condition not fullfill and have no restore best of any kind 16:00
marie_s so you stay at the limit of the conditions and then unstable solutions are best because scripts will find just enough good to the condiion and different but with no meaning because unstable 16:02
infjamc A possible solution would be automatically creating multiple autosaves 16:04
zoran it will have many unstable solutions, surely, but it will also be more likely to find the stable one no. once we do the clustering of diversified solutions, we will pick exemplars from which to start standard puzzles 16:04
*** anthioniphone quit (Remote host closed the connection) 16:04
infjamc ...so that a user may go back to the #2 credit best in case the top solution is unstable 16:04
zoran infjamc: history timeline? 16:04
Renton jude remember me saying bluefuse was the greawtest protein script 16:05
infjamc In a sense, yes, but remember that you're limited to 100 undos 16:05
thomirc it might be good to have a large number of save slots available (say 100) for any purpose 16:05
marie_s too short, I use scripts also on exploration 16:05
spvincent Is it necessarily bad that a protein be unstable and perhaps flip readily between shapes? 16:05
zoran we can make it selectable by user. only the issue of your memory on the machine 16:06
marie_s perhaps we can choose our best to share on this puzzle so honnest players will choose only after wiggle 16:06
thomirc selectable by user is good 16:06
auntdeenisland I need to leave - Seth & Zoran - thank you both for instigating the chat - I hope that some other time you will come back to discuss more (hopefully minor issues) - it was great to have everyone discuss! 16:07
zoran there's no way of telling if something is unstable until much later. 16:07
zoran feel free to call out these chats any time a big issue arises. we'll be there 16:07
Origami314 Can I make a suggestion on the tutorials? 16:07
thomirc can we talk about crashes before you go? 16:07
Renton when are they happening Thom? 16:08
Renton have you made a feedback entry yet? 16:08
Judecca|vicodin only the issue of your memory(leaks) on the machine :P 16:08
thomirc to start...I've identified over 20 separate crashes 16:09
thomirc *types of crashes 16:09
thomirc so far 16:09
zoran let's setup another time for these. jflatt is here and he'll continue esp on tutorials and crashes. i've got to have dinner with my daughters... 16:09
CFC thank you zoran. 16:09
thomirc thanks for your time 16:10
BletchleyPark your presence is much appreciated 16:10
BletchleyPark enjoy dinner 16:10
SethCooperIRC I have to leave as well 16:10
*** zoran quit (Quit: http://www.mibbit.com ajax IRC Client) 16:10
SethCooperIRC thanks for chatting 16:10
Get Started: Download
  Windows    OSX    Linux  
Windows
(7/8/10)
OSX
(10.7 or later)
Linux
(64-bit)

Are you new to Foldit? Click here.

Are you a student? Click here.

Are you an educator? Click here.
Search
Only search fold.it
Recommend Foldit
User login
Soloists
Evolvers
Groups
Topics
Top New Users
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