2015.01.30 State of Eterna 9am PST

From EteRNA WiKi
Jump to: navigation, search

 

jyesselm: hi nando [9:16 AM]

RedSpah: ha [9:17 AM]

LFP6: Hi Nando, Rhiju, Joe. :) [9:18 AM]

jyesselm: starting [9:18 AM]

jyesselm: https://www. youtube.com/watch?v=jV6ueodnZUk [9:18 AM]

hoglahoo: darn, I can't play audio. will have to watch a replay later [9:19 AM]

RedSpah: rhiju, I think your sound may be off [9:19 AM]

jyesselm: can you not hear? [9:19 AM]

britboy3456: there was a gap as the speaker changed [9:20 AM]

RedSpah: fixed [9:19 AM]

jyesselm: okay good [9:20 AM]

jyesselm: keep me updated [9:20 AM]

jyesselm: sometimes it cuts out for a few seconds [9:21 AM]

jyesselm: but if its longer then that let me know [9:21 AM]

hoglahoo: Is that ElNando's car [9:25 AM]

jyesselm: probably [9:25 AM]

hoglahoo: I call shotgun, pick me up we can carpool to eternacon [9:25 AM]

britboy3456: My sucess: just hit top 1000! [9:31 AM]

Eli Fisker: What I think is what works is that we are not told specifically what to look for, but get free hands to look for what we think could be going on in the data .  [9:37 AM]

LFP6: More structure for those who don't know where to start would be good though [9:37 AM]

britboy3456: well nova was really helpful for me [9:42 AM]

RedSpah: Kappa // [9:49 AM]

Eli Fisker: Congrats to Alumni :) [9:51 AM]

LFP6: Want to know what game Jee is making. :P [9:51 AM]

LFP6: CUSTOM TUTORIALS!!!! [9:51 AM]

Eli Fisker: Oh, great news with the tutorial tools! [9:52 AM]

LFP6: I wanted that. :P [9:52 AM]

LFP6: I suggested it. :P [9:52 AM]

Eli Fisker: LFP, thx :) [9:52 AM]

Eli Fisker: Applause to Team :) [9:53 AM]

jnicol: Yes, I'm here [9:53 AM]

RedSpah: Custom Tutorials - Dis Gun b Good [9:53 AM]

Nando: I'm here too [9:54 AM]

Eli Fisker: Hi EternaBot :) [9:54 AM]

rhiju: hey everyone we're on chat now [9:54 AM]

Adrien Treuille: Hi Everyone! :) [9:54 AM]

britboy3456: Hey! [9:54 AM]

Eli Fisker: Hi Adrien [9:55 AM]

LFP6: Hey Adrian! [9:54 AM]

Nando: thanks Rhiju & Adrien [9:55 AM]

RedSpah: Welcome all devs [9:55 AM]

Adrien Treuille: hi Eli.. Nice to see you.. and LFP6 and Nando [9:55 AM]

ezbot: Thanks for the broadcast! Great info [9:55 AM]

Adrien Treuille: I'll start off the conversation [9:56 AM]

rhiju: thanks for the applause everybody -- am conveying to the team [9:56 AM]

Adrien Treuille: by asking Rhiu [9:56 AM]

Adrien Treuille: what kind of cool stuff do we get to buy if  we win the Feynmann prize? [9:56 AM]

Omei: Feynman prize - here we come! [9:56 AM]

LFP6: lol [9:56 AM]

rhiju: if we get the feynman prize i think we'll get funded to make a new game to win the 2nd feynman prize [9:56 AM]

RedSpah: use Feynman prize to clone Nando [9:56 AM]

rhiju: the 2nd one involves making a robot [9:57 AM]

machinelves: a chat moderator and next buttons in labs! :D [9:57 AM]

LFP6: lol Red [9:56 AM]

rhiju: a nanorobot [9:57 AM]

LFP6: Oooh [9:57 AM]

rhiju: and yes if we make it, eventually the first use should be to clone nando [9:57 AM]

LFP6: That sounds cool [9:57 AM]

Nando: lol [9:57 AM]

Adrien Treuille: With two Nandos we can definitely win more prizes [9:58 AM]

rhiju: http://www.foresight.org/FI/fi_spons.html#GrandPrize [9:58 AM]

LFP6: :) [9:57 AM]

rhiju: there's only two grand challenges [9:58 AM]

Adrien Treuille: ok.. then we'll only clone him once [9:58 AM]

Omei: Surely with two Nando's, we can create some more [9:58 AM]

rhiju: but then we can fund more prizes and clone more nandos [9:58 AM]

rhiju: lol omei [9:58 AM]

Adrien Treuille: after we win both feynman prizes, I suggest we go after no scientific prizes with bounties less than $1M [9:58 AM]

RedSpah: with two Nandos we'll be able to RULE THE WORLD! (of course!)  [9:59 AM]

Adrien Treuille: ...but seriously, I'm interested to hear the community's thoughts about how we will be able to scale Eterna to the massive new scientific throughput [9:59 AM]

Nando: I see two challenges there [9:59 AM]

machinelves: upgrade lab UI & provide community resources [9:59 AM]

Eli Fisker: Better shortcuts, script integrated in lab  [10:00 AM]

Nando: complexity for one [9:59 AM]

RedSpah: complexity is fine [9:59 AM]

Nando: hopefully, the tutorials will help [10:00 AM]

britboy3456: Better tutorials for switches especially [10:00 AM]

Nando: and then, big data [10:00 AM]

rhiju: hopefully also scripts will help (eli's point) [10:00 AM]

LFP6: A way for users who have absolutely no knowledge to be able to be introduced to enough concepts that they can be helpful without having to go do the work to find resources themselves. [10:00 AM]

Nando: it's been on my todo list, really want to activate global scripting [10:00 AM]

LFP6: Ouch, runon sentence [10:00 AM]

Eli Fisker: Mat suggested recipe's and cookbook integrated like scripts. Just like foldit [10:01 AM]

Eli Fisker: Nando, great :) [10:01 AM]

LFP6: Id anyone follow what I said? :P [10:01 AM]

LFP6: *Did [10:01 AM]

rhiju: @lpf6 ideas for how to make that knowledge obvious? [10:01 AM]

LFP6: Well, a start would be more thurough tutorials [10:01 AM]

RedSpah: more advanced tutorials [10:01 AM]

LFP6: That too [10:01 AM]

jnicol: analysis and display of the results will be a big part of a large pipeline [10:02 AM]

Omei: It's always seemedd that figuring out what the experiments tell us has been more difficult than creating new things to test. [10:02 AM]

RedSpah: current ones make sure (most of the time) that you're able to solve easier challenges - not much more [10:02 AM]

LFP6: But if I need to do something for a lab in order to be useful, I should be introduced to the concepts through the game itself [10:02 AM]

RedSpah: good point Omei [10:02 AM]

rhiju: yes, agreed on tutorials. that's why we're exctied about mike bao's port of a tutorial creator (which he and Nova used for the nova lab) [10:02 AM]

britboy3456: Yes, it was easy to get the hang of the easy things, but the more difficult concepts took a bit of research [10:02 AM]

LFP6: Does this mean that we'll get the new UI and tutorials with it, or no? [10:02 AM]

rhiju: @britboy, good point. [10:03 AM]

Adrien Treuille: I worry that we're going to face an uphill battle scaling Eterna to more and more dat [10:03 AM]

Adrien Treuille: data visualization is hard!! [10:03 AM]

jnicol: agreed adrien [10:03 AM]

RedSpah: as if anything we previously did was any easier [10:03 AM]

Adrien Treuille: I wonder if we couldn't take the tack of integrating external tools [10:03 AM]

machinelves: have you tried .help britboy? there are a lot of good starter activities in there. [10:03 AM]

LinkBot: http://eternawiki.org/wiki/index.php5/HELP [10:03 AM]

Adrien Treuille: so that you could seamlessly move between eterna and other external tools to see the data [10:03 AM]

RedSpah: they'd certainly be useful [10:03 AM]

LFP6: Yes Adrian, would be great [10:03 AM]

machinelves: yes integration into main UI is good [10:04 AM]

rhiju: ok, on big data viewing, question for players [10:04 AM]

Omei: I agree, Adrien [10:04 AM]

LFP6: Now what I'm saying, is that I shouldn't have to manually look stuff up in order to know what I have to do for a lab (other than jsut blindly solving the puzzle). All that should be tought in the game interface [10:04 AM]

hoglahoo: Is there a custom tutorial maker available now? sorry I just got back [10:04 AM]

rhiju: currently players are creating their own visualizations [10:05 AM]

RedSpah: not yet, hai again hoge [10:04 AM]

britboy3456: @lfp6 yes, that's my point, the information is there on the wiki, but not easy to find [10:05 AM]

machinelves: wb [10:05 AM]

machinelves: yes a green help button in the chat window would help [10:05 AM]

Eli Fisker: Yes, which reminds me [10:05 AM]

Eli Fisker: We still have a ton of new players miss the Nova tutorials [10:06 AM]

LFP6: @britboy Yeap. One of the big reasons I want to renovate the Wiki-make catigorization of info a thing, and hopefully make sure we have needed information that might be missing [10:06 AM]

Eli Fisker: Jnicol, thx for what you did to make it more visible. It helped some, but so many still misses them [10:06 AM]

LFP6: Integration of tutorials into the main site will be great once that happens [10:06 AM]

jnicol: little by little :) [10:06 AM]

LFP6: No awy for them to be missed, generally speaking [10:06 AM]

RedSpah: on misc. ideas - support for different languages would be useful [10:06 AM]

LFP6: Good point Red, would open many doors [10:07 AM]

LFP6: Perhaps look for translaters? [10:07 AM]

Nando: I18N is a big investment... [10:07 AM]

LFP6: Translation system might or might not be difficult, depending on how the text info is stored [10:07 AM]

Nando: like any other investment, it has to be justified [10:07 AM]

hoglahoo: what is 118N [10:07 AM]

hoglahoo: I18 [10:08 AM]

hoglahoo: whatever [10:08 AM]

RedSpah: internationalisation [10:08 AM]

Nando: Internationalization [10:08 AM]

jnicol: some players have been asking about transcription services also, for th hearing impaired [10:08 AM]

RedSpah: wat [10:08 AM]

rhiju: lol on spelling internationalization [10:08 AM]

britboy3456: lol [10:08 AM]

Nando: :D [10:08 AM]

RedSpah: afaik we don't even have any audio material bar the dings when you change a base  [10:09 AM]

hoglahoo: I could only assume they're talking about events like the one that just took place [10:09 AM]

Adrien Treuille: ..and the AWESOME sound when you win :) [10:09 AM]

rhiju: @nando, similarly on how to justify development of data viz tools, what do you thin about integrating external tools [10:09 AM]

RedSpah: transcription: [placed G, placed C, BUBBLESBUBBLESBUBBLESBUBBLES] [10:10 AM]

jnicol: hog, correct [10:10 AM]

LFP6: lol Red [10:10 AM]

RedSpah: oh, that [10:10 AM]

Nando: my impression is that Flash is probably not the best medium for data analysis, so yeah, completely in favor of external tools [10:10 AM]

RedSpah: is Flash the best for anything? [10:10 AM]

LFP6: lol [10:10 AM]

Nando: best would vuz tools that can interface with Flash, so that certain things can still be visualized with that rendering engine [10:11 AM]

Nando: would be* [10:11 AM]

rhiju: the other option for external tools is to create a documeted API to the data and let players create their own viewers [10:11 AM]

LFP6: That would be neat [10:11 AM]

Nando: agreed [10:11 AM]

LFP6: Perhaps make it something with EteRNAScript? IDK [10:11 AM]

Omei: +1 for a general data API [10:12 AM]

LFP6: Well, API would be helpful for ES anyhow [10:12 AM]

hoglahoo: :D [10:12 AM]

machinelves: it would be awesome to have easy access to both spreadsheet data and various graph and pretty data viz tools [10:12 AM]

hoglahoo: Is there any downside to a documented api :P [10:12 AM]

machinelves: lol [10:12 AM]

Omei: None. [10:12 AM]

Nando: yes [10:13 AM]

Nando: it has to be programmed [10:13 AM]

hoglahoo: well, for you and whoever has to do it sure [10:13 AM]

Eli Fisker: Lol [10:13 AM]

hoglahoo: but the rest of us [10:13 AM]

Nando: that means work, for me... [10:13 AM]

hoglahoo: :) [10:13 AM]

RedSpah: thought you'd be used to that by now tbh [10:13 AM]

Nando: and that's definitely a dwonside :P [10:13 AM]

LFP6: How about updating the scripting docs? Don't think that was ever done (even though it was available on the dev server) [10:13 AM]

Adrien Treuille: I'm thinking, like, make experimental data available in open formats for players to make their own visualizsations [10:14 AM]

LFP6: Is there a standard format for such things? [10:14 AM]

Nando: partly already there with RMDB [10:14 AM]

Adrien Treuille: and promote links to those visualizations as first class elements of the data dumps [10:14 AM]

Nando: btw, wanna congratulate Siqi  [10:14 AM]

LFP6: And could there be a tool within EteRNA to create visualizations? [10:14 AM]

Omei: Making the Eterna libraries easily useable outside the internal scripting engine would also help. [10:14 AM]

Nando: he's done a fantastic job on RMDB [10:15 AM]

rhiju: @adrien, yes sound slike a plan [10:15 AM]

rhiju: @lpf6, creating viz in eterna sounds like a difficult dev project [10:15 AM]

rhiju: @nando, will convey to siqi -- that's a big compliment [10:15 AM]

machinelves: i'd like to see the best viz options integrated into the lab UI for players who don't code, so they can more easily see patterns in the data emerge [10:15 AM]

LFP6: @Rhiju Just a though, thinking something like Microsoft reporting tools, IDK. :P [10:16 AM]

rhiju: @elves, good point. sounds like a first step is to document the API & have links to data dumps within game. second step is to listen to player feedback and what is working externally and then consider integrating into the game. [10:17 AM]

LFP6: Definitely think the more that can get integrated into the EteRNA interface and have things flow freely between different sections of the site, the better [10:18 AM]

LFP6: I shouldn't need to learn a whole new UI every time I go to a new section of the site! [10:18 AM]

LFP6: Or go to do some different tasks [10:18 AM]

rhiju: @omei,by eterna libraries do you mean the text files with the sequences? [10:18 AM]

LFP6: But each thing obviously having their own components so it can work, just keeping it unisolated [10:19 AM]

Omei: No, I was thinking about the JS code, e.g. EternaUtils [10:19 AM]

Nando: on the other hand LFP6, you can't expect EteRNA interfaces and routines to rival Matlab or R [10:19 AM]

rhiju: @lpf6, on updating the scripting docs -- that seems critical if we integrate scripting into the game [10:20 AM]

Omei: Right now, they are bound up with the game in a way that makes them difficult to use from outside. [10:20 AM]

Adrien Treuille: right @nando.. that's why we should integrate rather than compete :) [10:20 AM]

Nando: @adrien: yep :) [10:20 AM]

rhiju: @omei, agreed [10:20 AM]

LFP6: @Omei: Yep. That's an issue I really have. I can't do things from the game. The game is limited, and my only real frame of reference, so doing anything outside of it is very hard [10:21 AM]

Omei: I think there is a place for both in-gamer and external tools.   [10:22 AM]

Omei: Making APIs that can be used by both would really help [10:22 AM]

rhiju: on viewing data & secondary structures, siqi is in 'phase II' of RMDB upgrade, which will include upgrading the sequence browser and viewer. [10:23 AM]

Omei: APIs for RNA-specific code, as well as date [10:23 AM]

rhiju: if those tools are acceptable, we might ask him to integrate into eterna [10:23 AM]

LFP6: Quick thought [10:23 AM]

Omei: Where do I find out about what siqi is doing? [10:23 AM]

Nando: http://rmdb.stanford.edu/repository/ [10:24 AM]

LFP6: When doing stuff with tutorials, would it be possible to teach everything that is currently housed in the Wiki, forums, lab descriptions, etc. in a way that makes sense going from that info to the game interface? [10:24 AM]

LFP6: I want to learn how I ndeed to change what I'm doing in the game interface. Currently I see the basis of the UI for that, and science explanations in the description, but nothing that tells be in the interface (or elsewhere) how to approach a solve differentlu [10:25 AM]

LFP6: *defferently [10:25 AM]

LFP6: *differently -_- [10:25 AM]

rhiju: for siqi's planning, http://rmdb.stanford.edu/repository/help/history [10:25 AM]

Hyperium12: my chat thing never seems to update... [10:25 AM]

rhiju: and i'm getting him on chat now [10:25 AM]

Hyperium12: odd.. [10:25 AM]

rhiju: @hyperium12, i think that might be an issue with flash version [10:26 AM]

rhiju: are you using chrome? [10:26 AM]

Nando: teaching is a non-trivial task, and "everything" is pretty ambitious LFP6 [10:26 AM]

LFP6: Trying to clarify my houghts [10:26 AM]

Hyperium12: yes rhiju [10:26 AM]

LFP6: @Nando: Just seems rather ambiguous to me right now :D [10:26 AM]

rhiju: @LPF6 i think its posisble to have 'everything' in a tutorial if we ask players to create those, and link to them from the wiki [10:26 AM]

rhiju: or for each lab project, we have a link to a tutorial puzzle designed by the puzzle admin [10:27 AM]

LFP6: But does it make sense what I mean whenI'm talking about relatability to the interace and changing what I'm doing in it? [10:27 AM]

LFP6: Just want to make sure I'm being clear [10:27 AM]

rhiju: @hyperium12, you may have to go to prefs and select to use an upgraded flash rather than chrome's builtin [10:28 AM]

Nando: the one thing I don't like in that approach LFP6, is that it creates a bias [10:28 AM]

Nando: and we may be interested in creative solutions never thought of before [10:28 AM]

Nando: remember the last round with the oligos? [10:29 AM]

LFP6: I'm really going for creating a smoother transition between the conceptual and the game [10:29 AM]

LFP6: Which is currrently pretty statically binary [10:29 AM]

rhiju: hey all, i've got to go -- can someone record a chat log? [10:30 AM]

Nando: I tried to give an example of a technique to solve a problem, but I was conflicted while doing it [10:30 AM]

hoglahoo: I feel so dumb I don't even understand the problem [10:30 AM]

hoglahoo: bye rhiju [10:30 AM]

Omei: Tx, Rhiju. [10:30 AM]

Eli Fisker: Bye Rhiju [10:30 AM]

Nando: cya Rhiju, I think machinelves will have it, worst case I'll save mine [10:30 AM]

rhiju: thanks! [10:30 AM]

Adrien Treuille: I have to go too [10:30 AM]

Adrien Treuille: it was really fun chatting with you all [10:30 AM]

Eli Fisker: Hogla, don't feel dumb. Its a problem that doesn't have a solution yet [10:31 AM]

machinelves: bye everyone, thanks! [10:31 AM]

Nando: bye Adrein, congrats to your bro [10:31 AM]

rhiju: will see some of you later today (3pm PST). [10:31 AM]

jnicol: thanks rhiju, adrien and everyone! [10:31 AM]

Eli Fisker: Bye Adrien [10:31 AM]

Adrien Treuille: and @rhiju -- thank you for charting this super exciting scientific course for Eterna's future!! [10:31 AM]

Eli Fisker: Bye John [10:31 AM]

Adrien Treuille: bye!!! [10:31 AM]

hoglahoo: Has current player participation level been mentioned yet? [10:31 AM]

Eli Fisker: No [10:32 AM]

hoglahoo: I'm kinda worried these new tools might be kinda lonely without newer folks being brought into the game [10:32 AM]

machinelves: scaling up to 100k / month lab designs also requires more participation, would help to increase player retention [10:32 AM]

hoglahoo: At least compared to when eterna was getting writeups in wired and nova [10:32 AM]

LFP6: I feel like all these tools will alinatate the new players because they'd have no clue what to do with them or how it has anything to do with the game, as we are game players. :P [10:32 AM]

hoglahoo: I'm all for the new stuff [10:33 AM]

hoglahoo: hoping if we build it they will come [10:33 AM]

LFP6: That's my problem! There's apperantly stuff to do, but I don't know what exactly I should be doing. :P [10:33 AM]

Eli Fisker: Hogla and Elf, I agree. We need to be more [10:34 AM]

hoglahoo: Weren't you working on a wiki UI or something [10:34 AM]

machinelves: lfp6 labs is a good place to start, that needs more participation [10:34 AM]

LFP6: One big issue I was worried about was finding, but apperantly that will be hopefully dealt with? :P [10:34 AM]

Omei: LFP6, I think that's a good point -- working to prevent a growing gap between the game and the science. [10:34 AM]

LFP6: @Hog: yes, I'm talking about science/game [10:34 AM]

machinelves: requires gamification of the work aspects  [10:35 AM]

LFP6: @Machine: But when I get there, what exactly am I supposed to be doing? I can solve the puzzles like I would any other, but that won't help anyone [10:35 AM]

machinelves: otherwise it seems like work [10:35 AM]

LFP6: Yes! [10:35 AM]

Eli Fisker: LFP, there are lab guides, that can help give you ideas for how to go about the task [10:36 AM]

machinelves: lfp6 i think you point out what is a roadblock for many new: labs UI could hold our hand while walking through steps, and chunk them into smaller pieces [10:36 AM]

LFP6: I personally feel that all the work has nothing to do with the game at all, and I have no clue where I would get started on it, what exactly I would be doing, etc. [10:36 AM]

LFP6: @Eli: Link please? [10:36 AM]

machinelves: Eli's lab is an intro to labs, that is the best place to start currently [10:36 AM]

britboy3456: re: getting more players, social media other than just facebook could help (e.g. reddit, twitter) [10:36 AM]

LFP6: (Btw, shouldn't have to ask for one!!!) [10:36 AM]

Omei: I wonder if we could make puzzles that actually use prior lab results [10:36 AM]

Eli Fisker: http://eternawiki.org/wiki/index.php5/Player-Created_Guides#Lab_Guides [10:37 AM]

machinelves: i think we should fix infrastructure before bringing people in, or the retention wont be there anyway [10:36 AM]

LFP6: Machine, definitely think that chunking would be the best way to go. Lab tutorials like we've been asking for. :) [10:37 AM]

machinelves: yes, agree lfp. check out Eli's lab, it is the closest to a tutorial we have [10:37 AM]

britboy3456: Oh, I found a twitter, but it hasn't been used since 2013 [10:37 AM]

Eli Fisker: LFP, here is a good place to start: [10:38 AM]

Eli Fisker: http://eterna.cmu.edu/web/lab/5525856/ [10:38 AM]

Eli Fisker: Check that big green Lab manual [10:38 AM]

hoglahoo: Omei: what do you mean? [10:38 AM]

hoglahoo: Eli created a series of puzzles something along those lines, if I understand you right [10:38 AM]

Eli Fisker: Yes. Moment [10:38 AM]

LFP6: Can you tell me one thing? What is the point of the lab you made (as in, what should be designed)? Is this the same as current labs, or are they looking for something more? [10:39 AM]

Eli Fisker: Coming here:  [10:39 AM]

Eli Fisker: http://eterna.cmu.edu/web/playerpuzzles/?skip=72&size=24&sort=date&search=Lab%20tutorial [10:39 AM]

Eli Fisker: The point of this lab is to solve the shape [10:39 AM]

Eli Fisker: And learn watching past lab data [10:39 AM]

LFP6: Is that true with the latest puzzles that have been presented? [10:39 AM]

Eli Fisker: To make better designs [10:39 AM]

Omei: ust beginning to formulate.  But suppose there was game that involved making choices between sequences that were actually synthesized.  The goal would be to find the best scoring ones. [10:39 AM]

hoglahoo: oh, I see [10:40 AM]

LFP6: It was my impression that with latest labs there is more to them than just solving the shape. And if I want to make my own lab, I would think that I would be investigating more than solving a shape (and how would I decide what shape to solve anyhow?) [10:40 AM]

hoglahoo: that could be useful when there are some hard lines between what is known to work and what is known not to [10:41 AM]

LFP6: @Omei: perhaps that would be a good introduction to lab! [10:41 AM]

Eli Fisker: Omei, I relaly like that line of thought [10:41 AM]

Eli Fisker: So a lab simulation [10:41 AM]

Eli Fisker: Run on past lab data [10:41 AM]

Eli Fisker: Giving scores based on designing [10:41 AM]

LFP6: Do you udnerstand my question Eli? [10:42 AM]

machinelves: omei i think that may have been floated as an alternate voting method [10:42 AM]

machinelves: actually [10:42 AM]

machinelves: really interesting idea [10:42 AM]

Omei: Yes, some general game puzzle that could be run on any set of real data. [10:42 AM]

Eli Fisker: LFP, sorry, I'm unsure which of your replies you are reffering to [10:42 AM]

Eli Fisker: Omei, I would love that [10:42 AM]

Eli Fisker: Would be a great way to learn [10:42 AM]

Eli Fisker: in a gamified way [10:42 AM]

LFP6: @Eli " It was my impression that with latest labs there is more to them than just solving the shape. And if I want to make my own lab, I would think that I would be investigating more than solving a shape (and how would I decide what shape to solve anyhow?)" [10:42 AM]

Eli Fisker: I was asking for it a long time back, around when the first lab tutorial was made by devs [10:43 AM]

Eli Fisker: But back then it was too complicated, and we didn't had much data [10:43 AM]

Nando: @Omei: just confirming: similar to what was done with Brent's theophylline lab, right? [10:43 AM]

machinelves: ahh thanks Eli couldn't remember where i saw :D [10:43 AM]

Omei: Nado, remind me what that was. [10:43 AM]

Eli Fisker: LFP, yes, a lab is like asking a question. Can be more than just solving shape [10:44 AM]

Eli Fisker: np Elf [10:44 AM]

Nando: Brent had experimental data for 1024 possible combinations in a theophylline+hammerhead -related sequence [10:44 AM]

Nando: players submitted sequences, but they all had already been tested [10:45 AM]

LFP6: So if it is, how do I know what I'm supposed to do? I could read the description, but that generally doesn't tell me how to change how I use the interface. Similarly, I would have no idea how I would create such research questions myself [10:45 AM]

Eli Fisker: So something like a near totally tested lab set, would be a good start for a simulation [10:45 AM]

Omei: I see.  I was thinking more of an interactive puzzle. [10:45 AM]

machinelves: lfp try the tutorial lab, and maybe you will learn a few steps. not all, but a start [10:46 AM]

Omei: But then, I don't remember what was done before. [10:46 AM]

Eli Fisker: And just different kind of shapes picked [10:46 AM]

LFP6: Well, if we were on our game,w e would use lab data to improve our model. :P Being able to actually see in 3D the interactions as opposed to jsut bonded or not would be helpful though [10:46 AM]

Eli Fisker: as they behave differently depending on shape [10:46 AM]

Eli Fisker: Omei, ok.  [10:46 AM]

Eli Fisker: Problem with our current puzzle simulation is that is not giving good enough feedback [10:47 AM]

LFP6: I'm just presenting the diea that I shouldn't have to search for info-it should be given to me and be compeltley compatable with what I already know about the interface [10:47 AM]

Eli Fisker: for what nature actually would accept [10:47 AM]

Eli Fisker: or want [10:47 AM]

LFP6: Yep [10:47 AM]

Eli Fisker: So puzzles are like training for ever in a bad simulation [10:47 AM]

machinelves: lfp i agree on in-main UI integration of info, and stepped interactive tutorial walkthrus, that's a matter of development [10:48 AM]

Nando: if you guys can wait just a couple weeks longer, Mike, Rhiju and I will be working on integrating these tutorial functions around second half of February [10:49 AM]

LFP6: It got to the point that when I finished challenges, I could no longer have *much* of anything I could move on to that was grind-ish and on-par with my current knowledge level (plaeyr puzzles to some effect, but I need to gain some skill for that :P) [10:49 AM]

Nando: can't say when exactly a first will be available, but it shouldn't take too long [10:50 AM]

machinelves: yes Eli I noticed that is a big confusion for new players moving to labs, is unlearning energy model tricks that don't work in real life [10:50 AM]

machinelves: nando, awesome! :D [10:50 AM]

Nando: maybe that's an artifact of the puzzles, not the labs [10:51 AM]

machinelves: labs use puzzle UI, could they get tutorial functionality as well? [10:51 AM]

Eli Fisker: Nando, big thx! [10:51 AM]

Nando: maybe it should be made more clear in the puzzles areas that these are only in silico predictions [10:51 AM]

Eli Fisker: And I gree, its an effect of the many puzzles [10:51 AM]

Nando: @elves: I would think so [10:51 AM]

Nando: I haven't seen the engine yet [10:52 AM]

Omei: Possiblty: Using the MS2 data, player starts with a randomly selected solution and sees the switch graph for it, along with some alternative designs that are close (or not) to the current one.  Player selects one of the choices and sees its results, along with a new set of alternayives.  Player navigates through the designs in this way trying to find the best one. [10:52 AM]

machinelves: yes that is a good point, not a clear differentiation between algorithmic & real world modelling [10:52 AM]

machinelves: in puzzle ui [10:52 AM]

hoglahoo: Nando: I don't think that will matter much. We're still being trained a certain way regardless of the warning/note\ [10:52 AM]

machinelves: haha [10:52 AM]

hoglahoo: It's like you try to tell people rank and points don't matter [10:52 AM]

hoglahoo: then you work toward increasing it anyway for 3 years [10:53 AM]

machinelves: actually i said rank and points do matter, which is why fmnsanity succeeded [10:53 AM]

Eli Fisker: Good observation, Hogla [10:53 AM]

machinelves: and i was rank 10 on my second day and have been dropping in rank ever since [10:53 AM]

Nando: I would have to disagree that it was a success, but that's an argument I don't want to start now [10:53 AM]

machinelves: by succeeded i mean produced a large quantity of submissions due to point incentivization [10:54 AM]

Eli Fisker: Omei, I like your idea [10:54 AM]

Eli Fisker: It reminds me of a method of voting we were testing a long time back [10:55 AM]

Nando: @hoglahoo: maybe the over-training comes from the fact that there are way too many puzzles? and not enough time spent in labs [10:55 AM]

Eli Fisker: We were given already scored design (but score not showed) , and made to pick between which we thought best [10:55 AM]

Nando: which makes Omei's idea even better [10:55 AM]

LFP6: I do think a comparrison thing would be cool, at least in tutorial (seeing what things work, don't, etc. and why) [10:55 AM]

Nando: we could give players things to do, even between rounds [10:55 AM]

hoglahoo: Nando: if not solely from that fact, surely partly to some degree [10:55 AM]

LFP6: From the perspective of a new plaer, challenges are the most important thing, then labs, then player puzzles [10:56 AM]

LFP6: Not really true [10:56 AM]

hoglahoo: I spent more effort in lab when I only had a couple dozen puzzles left to solve and could only create three daily [10:56 AM]

LFP6: Because challenges are apperantly not used [10:56 AM]

LFP6: @Nando: Brings up a good point. For me personally, I want to be handheld more than I am now about what specific things I should be doing next [10:56 AM]

LFP6: Challenges are great because solve 'em all, but... [10:57 AM]

LFP6: Again, unuseful. Though that's what you're directed to [10:57 AM]

machinelves: players are obviously hitting a roadblock in transitioning to labs, so chunking lab steps into smaller pieces that are each rewarded would increase participation [10:57 AM]

LFP6: Yep [10:57 AM]

LFP6: Gradual increase to new concepts, as opposed to it being so chunky [10:58 AM]

machinelves: i think there is a chasm between the gamified puzzles and the work of labs [10:58 AM]

Nando: I'm not sure I can imagine how to chunk that part [10:58 AM]

LFP6: New tutorials for features will help... Woudl they be able to be promoted to being official tutorials from player puzzles? Would be important IMO [10:58 AM]

LFP6: @Machine: Yep. Very little gamification with labs. Foldit is nice because you get straight score you're looking to improve (above others') [10:59 AM]

machinelves: lfp that is a good idea, tutorial walkthru can add chunking nodes, each popup can be a step [10:59 AM]

machinelves: with opportunity for learning and reward on completion [10:59 AM]

LFP6: ^^ [10:59 AM]

machinelves: then save progress, so you can quit in the middle and not lose your place [11:00 AM]

machinelves: so that the time investment per session is less [11:00 AM]

LFP6: Key there: Make more reward and completion stuff aimed towards what they player 'should be doing' [11:00 AM]

LFP6: Save feature would be amazing [11:00 AM]

LFP6: *the player [11:01 AM]

Nando: what players should be doing has been clearly established: win the Feynmann Prize, and use the funds to clone me :P [11:01 AM]

Omei: What';s plan B? [11:02 AM]

Nando: lol :D [11:02 AM]

LFP6: And how do I do that? You need to be direct to me as opposed to giving me an abstract concept with no idea how to implement. :P [11:02 AM]

Nando: no LFP6, I don't need to [11:03 AM]

LFP6: The completion/reward aspect can really help in that regard [11:03 AM]

LFP6: @Nando: Then how will I be able to do it? [11:03 AM]

LFP6: How will I know what to do? [11:03 AM]

Nando: the goals are stated, that solves the what [11:04 AM]

Nando: the how, you will learn to look up for the information yourself, if necessary [11:04 AM]

LFP6: Stop right there [11:04 AM]

Nando: that's part of learning what it is to do research [11:04 AM]

LFP6: You're assuming too much of the palyer, the average one which will not want to do extra work unless it's finding cheats or hints [11:04 AM]

LFP6: NOT learning conepts [11:05 AM]

LFP6: If you expect a player to learn a concept themself, good luck! [11:05 AM]

machinelves: yes if we were being paid or getting a degree we would work [11:05 AM]

Nando: I don't expect gamers to get interest in the labs [11:05 AM]

machinelves: but as a game, the whole concept is fun [11:05 AM]

LFP6: Especially if you're halfway into whatever you're dealing with [11:05 AM]

machinelves: i think the whole point is to gamify science [11:05 AM]

machinelves: that requires UI dev [11:05 AM]

Nando: http://static1.squarespace.com/static/522b7349e4b0d5218db16308/t/54a2d427e4b0a7ae51d268f8/1419957287423/Ryan+Murphy+-+Facilitating+Citizen+Science+Through+Gamification+-+Introduction+and+Methods_v3.pdf [11:06 AM]

machinelves: LFP6, for the time being, here are Eli's lab tutorials in puzzle form, not what you are asking for I know, but it does help intro some key lab concepts http://eterna.cmu.edu/web/playerpuzzles/?skip=48&size=24&sort=date&search=%5Blab%20tutorial%5D [11:06 AM]

LFP6: Nando.... This is a game. Citizen Science Game. THE WHOLE POINT OF THIS IS TO BE A GAME WHERE YOU CAN DO THE SCIENCE WITHOUT BEING A SCIENTIST [11:06 AM]

Nando: task gamification, or game taskification [11:06 AM]

Nando: there are different approaches and that's delicate [11:06 AM]

Nando: @LFP6: it may not be possible to apply that concept everywhere [11:07 AM]

Nando: read what I just posted to get an idea [11:07 AM]

LFP6: Then EteRNA is NOT the right palce to do whatever that is [11:07 AM]

machinelves: thanks for the paper, and interesting. but i think that game taskification is tantamount to slave labor, imho [11:08 AM]

hoglahoo: lol [11:08 AM]

machinelves: the platform premise as it is understood by players is gamification of science [11:08 AM]

Nando: make EteRNA into a pure game, and you will lose all the participants who see themselves as amateur scientists and not as gamers [11:08 AM]

Omei: I don't think we should emphasize the distinction between the game and science aspects.  [11:08 AM]

machinelves: crowdsourcing and scalability of work at a low cost is the holy grail of... industry, not science or gaming [11:09 AM]

LFP6: What Omei said' [11:09 AM]

Omei: Better to find the integration. [11:09 AM]

machinelves: never said to remove science altogether [11:09 AM]

machinelves: part of the incentive is that it is for a good cause [11:09 AM]

machinelves: which motivates, but only so far [11:09 AM]

LFP6: Nando, ther eshould be no difference between science and game. As a gamer, I should be able to do science, and know I am, and be able to do it legit [11:09 AM]

LFP6: But still being a gamer [11:09 AM]

Nando: once you ask me to guide your hand in every aspect of the space, there is no science anymore, because I have biased all your possible responses [11:10 AM]

machinelves: actually it is just good UX what LFP6 asks for [11:10 AM]

machinelves: and it does require dev, as the paper cites [11:10 AM]

Nando: do you understand that we do need people thinking for themselves, creating, solving unsolved problems, or differently [11:10 AM]

LFP6: There is a difference between you telling me how to act in the game and how to paly it, if that makes any sense [11:10 AM]

Nando: we can't give all the answers, sometimes even when we have some of them [11:11 AM]

machinelves: yes i do understand that, i also understand people spend a lot of energy doing those things with work, home, and family, and when they play a game they want easy, fun, rewarding [11:11 AM]

LFP6: THinking for youself does not require not being told how to use an interface or be left to yor own devices to find knowledge of concepts [11:11 AM]

machinelves: so even when learning is required, that learning is easier when incentivized and calibrated to the user's ability [11:11 AM]

LFP6: Btw Nando, will definitely read the paper, looks interesting :D [11:11 AM]

Omei: Did anybody else notice it seems to be getting stuffy in here? [11:12 AM]

hoglahoo: That's probably the large platter of greasy, cheesy texmex I polished off a few minutes ago [11:12 AM]

Nando: :D [11:13 AM]

LFP6: You need to be given directives to know you're doing a good task, though not neccessarily that you're creating the corret solution or even in the right way [11:13 AM]

Omei: Maybe we should all go out for a drink. [11:13 AM]

machinelves: * opens window * [11:13 AM]

Eli Fisker: :) [11:13 AM]

LFP6: I'll go get some ice. :P [11:13 AM]

Nando: @machinelves: there are various theories about how people best learn... [11:13 AM]

Eli Fisker: Will find some tea [11:13 AM]

hoglahoo: You can have a pepsi, LFP6 [11:13 AM]

LFP6: Machine, it's cold outside! [11:13 AM]

machinelves: hahaha [11:13 AM]

LFP6: OMG though, the snow and cold today from school were brutal. O_o [11:13 AM]

Nando: and certain things may work well with some, and not at all with others [11:14 AM]

Omei: Snow?  I didn't notice any. [11:14 AM]

Eli Fisker: Hehe [11:14 AM]

Omei: The pool was a little chilly, I admit. [11:14 AM]

LFP6: lol [11:14 AM]

LFP6: Ice fishing? :P [11:14 AM]

LFP6: Wait, chlorine would kill the fish. :P [11:14 AM]

Omei: lol [11:15 AM]

Omei: Well, I'm going to go make coffee.  Should I make one for anybody else? [11:16 AM]

machinelves: thank you kindly, yes please [11:16 AM]

Eli Fisker: Warm and pleasant offer. Apprechiated Omei [11:17 AM]

Omei: You like it black, I hope?  I have a feeling I'll end up drinking it myself [11:17 AM]

Eli Fisker: Hehe, yep [11:17 AM]

machinelves: hehe as you like it, is perfect [11:17 AM]

LFP6: How about hot chocolate? Would love some of that 'bout now. [11:17 AM]

Omei: Super [11:17 AM]

LFP6: lol Omei [11:17 AM]

machinelves: good idea lfp [11:18 AM]

lroppy: is the open group meeting over? [11:18 AM]

machinelves: lroppy presentation ended, and we were just chatting [11:19 AM]

lroppy: ah [11:19 AM]

machinelves: and hi :) [11:19 AM]

lroppy: hi [11:19 AM]

Eli Fisker: Yes, now it is coffee or whatever is your favorite drink meeting  [11:19 AM]

lroppy: I just have a few mins b4 nxt wrk mtg [11:19 AM]

hoglahoo: There's supposed to be a reboot of the meeting later today [11:19 AM]

lroppy: 5:15 for us, yes [11:19 AM]

lroppy: of 5pm [11:19 AM]

lroppy: or [11:20 AM]

lroppy: I guess [11:20 AM]

machinelves: welcome to the warm beverages meeting, and wish you happy work meeting next [11:20 AM]

LFP6: 6:00 here, I'll be back at Robotics. :( [11:20 AM]

Eli Fisker: :) [11:20 AM]

Eli Fisker: And sorry, LFP [11:20 AM]

LFP6: Wish I could make it! [11:20 AM]

Eli Fisker: But sounds like you will have fun programming [11:21 AM]

lroppy: beverages, like revenge, best served cold [11:20 AM]

machinelves: cool robotics meeting [11:21 AM]

LFP6: nope, electronics. :) [11:21 AM]

lroppy: catch folks later, perhaps [11:21 AM]

Eli Fisker: Programming with components ;) [11:21 AM]

LFP6: Just hooking up stuff for Programming to use, really. :P [11:22 AM]

Eli Fisker: :) [11:23 AM]

machinelves: bye lroppy! [11:23 AM]

machinelves: time for me to go as well [11:24 AM]

machinelves: see you all later [11:24 AM]

LFP6: Bye, good talking! [11:24 AM]

Eli Fisker: Bye Elf, see you [11:24 AM]

Eli Fisker: And I should go too [11:24 AM]

machinelves: bye LFP6 & Eli! good to see you too :) [11:24 AM]

LFP6: Bye Eli, enjoy. :) [11:24 AM]

Eli Fisker: Thx, LFP :) [11:25 AM]

Eli Fisker: Thx Elf too [11:25 AM]

machinelves: :) [11:25 AM] 

Personal tools
Main page
Introduction to the Game