2015.10.09 Dev Chat

From Eterna Wiki

( Link back to Dev Chat archive )

On New Switch miRNA Labs, Nando's Educational Blog, Chat Spam Ignore, Puzzle Progression, Tooltips for Chunking Lab Tutorials, SHAPE, & Lab Voting Points


rhiju: hi everyone  [3:01 PM]

Nando: hi Rhiju  [3:01 PM]

Eized: heya c:  [3:01 PM]

Elves: hello!  [3:01 PM]

Nando: @Eized, we'd have to talk about statistical mechanics to explain that  [3:01 PM]

Nando: I did a sort of intro on the subject a long time ago  [3:02 PM]

rhiju: nando, did you end up preparing any slides or videos?  [3:02 PM]

Nando: for the current lab? yes, I wrote a personal blog entry actually  [3:03 PM]

Nando: I'm not sure this can be seen as introductory though... [3:03 PM]

Nando: http://eternawiki.org/wiki/index.php5/User:ElNando888/Blog/Designing_for_the_A_B_lab [3:03 PM]

rhiju: no i meant when you did a intro on stat mech [3:03 PM]

Nando: ah, well, that was a recorded online session, gimme a sec [3:03 PM]

rhiju: cool blog post, though -- am reading! [3:04 PM]

Nando: http://eternawiki.org/wiki/index.php5/User:ElNando888/Online_meetings#Thermodynamics_for_Dummies.2C_by_a_dummy [3:04 PM]

Elves: whoa amazing blog post Nando!! [3:04 PM]

Nando: hmm, the thermodynamics lecture looks broken from here... [3:05 PM]

Nando: maybe the download link still works... [3:06 PM]

rhiju: that's cool. it was using teamviewer? [3:07 PM]

Nando: yes [3:07 PM]

Nando: but I think I messed up the conversion to MP4, somehow [3:08 PM]

Nando: will try to fix it later [3:08 PM]

rhiju: aha. [3:08 PM]

Elves: wow amazing nando treasure trove blog of writeups! http://eternawiki.org/wiki/index.php5/User:ElNando888/Blog [3:08 PM]

rhiju: part of the reason i was asking was that it would be cool to have more such presentations [3:08 PM]

rhiju: johan gave an 'open group meeting' this morning -- very useful stuff. [3:09 PM]

rhiju: but i'd also personally like to hear more presentations like what we had at eternacon [3:09 PM]

Eized: *sigh* I think I'm going to be near-useless this lab, read the blog but my puzzle solving skills don't help [3:09 PM]

rhiju: i wonder if we could have the monthly open group meeting slot occasionally (every other month?) rotate to a player [3:09 PM]

rhiju: it could be devoted to strategies -- e.g., to help folks like Eized (and me, who is a much worse player). [3:10 PM]

Nando: I see one perfect candidate right now: Cody! :D [3:10 PM]

rhiju: :) [3:10 PM]

Nando: hey man :) [3:10 PM]

Elves: another idea that was suggested is that since the current times on Friday are difficult for some players, to have a second day each week for educational and player presentations & walkthroughs [3:10 PM]

rhiju: what are the good times of the week or day? [3:10 PM]

Elves: with the monthly meeting being the once a month presentation from the scientist side [3:11 PM]

Elves: https://getsatisfaction.com/eternagame/topics/hi-here-is-where-you-can-tell-the-devs-what-is-the-best-times-for-you-to-attend-a-dev [3:12 PM]

rhiju: for us its convenient to have that meeting at 9:30am PST, which is our normal group meeting time. [3:12 PM]

bekeep: Hi all, sorry to pop in late [3:12 PM]

rhiju: ah thanks for the getsat link [3:12 PM]

Elves: np! [3:12 PM]

Elves: hi Ben! [3:13 PM]

Elves: so we could do a few things to manage the various timeslots [3:13 PM]

Elves: the monthly presentation could just stand on its own, since we already have a dev chat that day [3:13 PM]

rhiju: the other convenient time is 3pm PST on Fridays (i.e. now) [3:14 PM]

rhiju: @nando, when did you have the online chats/teamviewer, like for your stat mech intro [3:14 PM]

Nando: this dates back almost 2 years I think [3:15 PM]

rhiju: ah [3:15 PM]

Elves: 3pm friday would make sense too, then we could consolidate the monthly presentations into the regular friday chat cycle, especially if it frees up the scientists to participate regularly [3:16 PM]

Nando: well, I can no longer make the vids work in the browser, but if you download them and open then with VLC, they work [3:17 PM]

Elves: and then we can choose a different day and time for the alternate meeting where players can give presentations, walkthroughs, lessons, etc. [3:17 PM]

Nando: them* [3:17 PM]

Elves: hahaha oh thank goodness for VLC  [3:17 PM]

rhiju: i'm enjoying the review of http://eternawiki.org/wiki/index.php5/User:ElNando888/Blog/BPS [3:17 PM]

Nando: woah, that's an old one... [3:17 PM]

rhiju: my lab is working on a 3D 'rna puzzle' that involves predicting 3D structure of 2 sequences [3:17 PM]

rhiju: perhaps not coincidentally, the sequences differ by replacing a U-C pair with a G-A pair. [3:18 PM]

bekeep: Yes... I've also been distracted reading Nando's blog posts now... [3:18 PM]

Nando: interesting [3:18 PM]

rhiju: Read this from the blog post: "Now the question is: can a CU pair (CU_33 is presented as "compatible") really replace a GA one? I kinda get the point that the positions and distances of the C1'-C1' doublet is in a very acceptable range (which means that the backbone shouldn't be influenced much), but I have some strong doubts about the stacking interactions..." [3:18 PM]

rhiju: i am wondering the same thing. [3:18 PM]

Elves: yes i too fell in the rabbit hole and only resurfaced several minutes later. thanks Nando! [3:19 PM]

rhiju: there is some evidence that the answer is yes. can talk to you offline about it, nando. =) [3:19 PM]

Nando: would love to hear about it :) [3:19 PM]

rhiju: i have been sworn to secrecy on what the puzzle sequences are, so can't talk on chat -- will PM you [3:19 PM]

rhiju: did everyone hear about /ignore? [3:20 PM]

rhiju: (is that live?) [3:20 PM]

bekeep: I haven't tested it yet [3:20 PM]

Nando: it is [3:20 PM]

bekeep: SWEET [3:20 PM]

rhiju: sweet, just tried it. [3:20 PM]

bekeep: Nice [3:20 PM]

Nando: apparently, not many noticed, until I mentioned it earlier today [3:20 PM]

Elves: is there an unignore? [3:21 PM]

rhiju: might be worth a news post or something. perhaps also a mention at pages that describe chat comments, like http://eternawiki.org/wiki/index.php5/LinkBot [3:21 PM]

bekeep: unignore [3:21 PM]

Elves: someone mentioned earlier but i wasn't sure if it was a joke or serious [3:21 PM]

Nando: @elves: yes http://www.eternagame.org/web/blog/6236223/ [3:22 PM]

Elves: like you type /unignore to work? [3:22 PM]

Elves: ohhh thanks!! [3:22 PM]

bekeep: yep [3:22 PM]

Elves: and someone also mentioned getting a list output of who was ignored so you can take them off, i think hoglahoo. if that is feasible, I second the request [3:22 PM]

Nando: doable... /list-ignores ? [3:23 PM]

bekeep: !$ [3:23 PM]

bekeep: that seems like a reasonable name for the command [3:23 PM]

Nando: or I could make it a /show ignores [3:24 PM]

Nando: with the possibility of showing other stuff in the future... [3:24 PM]

bekeep: makes sense [3:24 PM]

Nando: great, a new item in my todo list :D [3:25 PM]

bekeep: slash-show how awesome nando is [3:25 PM]

bekeep: (how do you escape the slash in chat?) [3:25 PM]

bekeep: "/show" [3:26 PM]

Nando: I don't think it's possible if it's the first character of the message [3:26 PM]

Nando: for instance [3:26 PM]

bekeep: ah /ah [3:26 PM]

bekeep: got it [3:26 PM]

Elves: thanks nando :) :) and no rush, just an idea  [3:26 PM]

Elves: it could be /listignore [3:27 PM]

Jennifer Pearl: hi [3:27 PM]

Elves: hi Jen! [3:27 PM]

Nando: hi Jenn [3:27 PM]

rhiju: hi jen [3:27 PM]

Elves: Nando this post you made on miRNA & switches is cool too http://eternawiki.org/wiki/index.php5/User:ElNando888/Blog/New_switches [3:30 PM]

bekeep: Hi Jen [3:30 PM]

Nando: @elves: I worked hard on that one, yes :) [3:30 PM]

Elves: :) :) :) thanks for your efforts to explain things [3:31 PM]

Elves: especially in pictures !! [3:32 PM]

rhiju: based on eized's comments earlier, it seems like we should start thinking of 'starter' puzzles for the labs that will newer players 'glide in' to A/B  [3:32 PM]

Elves: that would be a huge help! [3:33 PM]

Elves: are there puzzles like these labs in the puzzle progression Ben? [3:33 PM]

Elves: for training up to it [3:33 PM]

rhiju: kind of like nando's link to the 'dimer tutorial puzzle' in the blog entry above (http://eternawiki.org/wiki/index.php5/User:ElNando888/Blog/New_switches) [3:34 PM]

Elves: http://nando.eternadev.org/web/puzzle/3378672/ [3:35 PM]

Nando: since puzzlemaker can't create these puzzles, I guess I would have to publish a few simpler ones... [3:35 PM]

bekeep: We've talked about making a lab progression but haven't made it yet.  The current puzzles only go up to the MS2 /flourescence paradigm [3:35 PM]

Elves: ahh so puzzlemaker doesn't do dimers? so the tutorial above is some fancy magic juju on your dev server? :D [3:36 PM]

Nando: there's a lot of magic on the dev server, yes [3:37 PM]

Elves: @Ben it would be ideal if the puzzle progression ends with the player being equipped to begin the current labs [3:37 PM]

Elves: hehehe [3:37 PM]

rhiju: the alternative is to include the 'trainer puzzles' in the lab project itself. [3:37 PM]

bekeep: @Elves Right - that is the goal [3:37 PM]

Elves: @rhiju that sounds like a great idea regardless [3:38 PM]

rhiju: that way if labs change, we'll have customized progressions for each. [3:38 PM]

Elves: i know it will always be an evolving landscape [3:38 PM]

bekeep: "trainer puzzles" is what we've talked about before [3:38 PM]

bekeep: then we could fold trainer puzzles into the progression [3:38 PM]

Elves: but there is definitely a wall sized hurdle for new players and those who had a break from labs in trying to participate in current labs [3:38 PM]

bekeep: (and we would have tested the puzzles well before they would be folded in) [3:38 PM]

bekeep: "Tear down that wall!" [3:39 PM]

Elves: hehehe [3:39 PM]

Elves: where are we at with the NOVA style tooltips as popups for guided tutorial puzzle solving? [3:39 PM]

Elves: was that something now available through scripting or not there yet? [3:39 PM]

Nando: text boxes are available in tutorial scripting [3:40 PM]

Elves: i think from my experience doing a wlakthrough of the previous lab was that many of us are new to terminology and procedure of solving itself. so a guided walkthrough puzzle might help break the pieces down... [3:40 PM]

...into smaller chunks

Elves: awesome!! thanks Nando [3:40 PM]

Nando: it's just that we're following a general tutorial designing rule that prescribes to use as little text as possible [3:40 PM]

Elves: does that include a stepped / gated progression through those text boxes? like next - next or disappear after completion of x task? [3:41 PM]

Nando: I think so [3:41 PM]

Nando: I mean, both are possible [3:42 PM]

Elves: yeah that is ideal i agree. less text more pictures. :D but the complexity is pretty high so it can help to read 10-15 words in each tooltip while you basically get a live walkthrough explaining what to do,... [3:43 PM]

Nando: and you can have multiple boxes with varying lifetimes, if I recall correctly [3:43 PM]

...since I often forget what the description said after opening the puzzle itself

bekeep: Actually the playtesting process is pretty humbling... we might be adding a (*small*) amount of text in or at least better orientation [3:43 PM]

Elves: ooo cool! [3:43 PM]

Elves: the other thing the tooltips do is to create chunking of the process into achievable steps [3:44 PM]

Nando: "as little as possible" is not necessarily zero, of course :) [3:44 PM]

bekeep: So I'm learning... : ) [3:44 PM]

Elves: whereas the sense of opening a lab usually is of wide open time commitment and unknown level of difficulty. getting to the next tooltip is far more digestible goal [3:44 PM]

Elves: we kind of do this when players make puzzles that are smaller pieces of larger puzzles [3:45 PM]

bekeep: Yes, breaking the lab into doable chunks would be valuable in increasing participation, I think [3:45 PM]

Elves: but they are still separated. so this would allow the whole project to exist in one session, but chunking it [3:45 PM]

Elves: this is a very very basic explanation of that principle http://www.dashe.com/blog/learning/chunking-memory-retention/ [3:46 PM]

Elves: there is another bit of research on taking breaks during tasks to increase retention as well [3:47 PM]

Astromon: hi sry im late (again today) i have a question [3:47 PM]

bekeep: Hi Astromon [3:47 PM]

Elves: hi Astro! [3:47 PM]

Astromon: when i have a design (or anyone) i wonder is it best to make it change shape for the mod of it to score dif [3:48 PM]

Astromon: like i can replace an a for a u for instaeance [3:48 PM]

Astromon: but it doesnt change the shape [3:48 PM]

Astromon: which i9s better more than likely is my question [3:49 PM]

Astromon: with 100 i can do many dif ways but i was just wondering you guys opinion on it [3:50 PM]

Astromon: whenever your done with what you were discussing o9f coarce  :0 [3:50 PM]

Astromon: :)* [3:50 PM]

Elves: it's a good question Astro - whether to mod or make unique design you mean? [3:51 PM]

bekeep: @astro I'm not sure if we know the answer to this question. [3:51 PM]

Elves: i think both are relevant approaches, and we can better answer it with more data, so I would say do both! [3:51 PM]

bekeep: Or do you mean what mods that don't change the shape are better? [3:51 PM]

Astromon: well both would be mods [3:51 PM]

Astromon: but one would actualy change the shape [3:51 PM]

bekeep: ah.... [3:52 PM]

Astromon: was wondering which would be best [3:52 PM]

bekeep: one mod changes the shape and one mod does not, but both shapes are permissible under the lab constraints? [3:52 PM]

Astromon: maybe sometimes one way is best and sometimes the other ? [3:52 PM]

Elves: oh I see. well we want to know the shape that comes back from the lab. whereas the shape you see on the computer is a prediction. so i would still try both [3:52 PM]

Astromon: haha [3:52 PM]

bekeep: Barring input from Nando or Rhiju, I would say try both and see what happens [3:52 PM]

Astromon: oH [3:52 PM]

bekeep: : ) [3:53 PM]

Astromon: Okay tyhnaks! [3:53 PM]

Astromon: thanks* [3:53 PM]

Elves: like for example you trade out a U for a G. And the shape on the computer doesn't seem to change. but that is an algorithmic prediction. not necessarily accurate. [3:53 PM]

Astromon: oH Okay [3:53 PM]

Elves: we want to know how to update the algorithm. so it helps to challenge it to return results that either validate the prediction or show where it fails [3:53 PM]

Astromon: it may actualy change shape we just dont know it [3:53 PM]

Astromon: hmm cool [3:54 PM]

Elves: exactly, we don't know until we get lab data back. well - Nando or Rhiju can say - I don't know that we get SHAPE data back anymore, or any way to validate the precise shape. [3:54 PM]

Nando: SHAPE labs will come back in the future [3:55 PM]

bekeep: These labs don't use SHAPE data, but later labs next year will [3:55 PM]

Elves: it's more about whether it binds or does not bind to a miRNA which favors a general shape, that may or may not be precisely the same as the shape predicted [3:55 PM]

Nando: as for Astromon's dilemma, I agree with what was said, try both, if you have enough slots [3:55 PM]

Astromon: i was thinking changing shape is good because it may be an esier switch in that shape er another [3:55 PM]

Nando: gotta prepare for the next meeting, going afk, see ya folks [3:56 PM]

Elves: bye Nando! [3:56 PM]

Elves: thanks!! [3:56 PM]

Astromon: seeya [3:56 PM]

Astromon: im going to try the 3 state r2 tonight  [3:58 PM]

Astromon: im exited [3:58 PM]

Elves: nice! [3:58 PM]

Astromon: my company hates when i design rna haha [3:59 PM]

Astromon: so i turn on good music for everyone and cook good food that always helps<. [3:59 PM]

Jennifer Pearl: i totally forgot that there was a chat just now [4:00 PM]

Astromon: did you guys have a nice dev chat [4:00 PM]

Jennifer Pearl: I ws coding and totally spaced ou ton new music  [4:01 PM]

Astromon: lol [4:00 PM]

Astromon: pass that this way :P [4:01 PM]

Elves: ok i don't have my research references handy on chunking atm, but it's a common principle for eLearning and instructional design. here's another basic intro to it http://blog.convergencetraining.com/chunk-training-m [4:01 PM]


Elves: lol link broken http://blog.convergencetraining.com/chunk-training-materials [4:01 PM]

Jennifer Pearl: when is the next cht? [4:01 PM]

Jennifer Pearl: it sounded like nando said there would be a later one [4:01 PM]

Astromon: i was thinking on the days you have the morning chat you could take break on dev chat latyer that day [4:02 PM]

rhiju: go astromon! will look for lab solutions from you  [4:02 PM]

Astromon: but i love 2 chats personaly [4:02 PM]

rhiju: i have to go -- thanks all for being at the chat [4:02 PM]

Jennifer Pearl: hi rhiju did you get my message [4:02 PM]

Astromon: Thanks Rhiju! [4:02 PM]

Elves: thanks Rhiju! [4:03 PM]

Jennifer Pearl: bye rhiju [4:03 PM]

Jennifer Pearl: thanks even though I wasnt here [4:03 PM]

Elves: I don't think it was decided what days to have the meetings. it would be good for everyone to chime in on time and day options in this forum post and we can decide accordingly: https://getsatisfaction.com/eternagame [4:03 PM]


Elves: omg [4:03 PM]

Elves: https://getsatisfaction.com/eternagame/topics/hi-here-is-where-you-can-tell-the-devs-what-is-the-best-times-for-you-to-attend-a-dev [4:03 PM]

Elves: lol Jen. thanks to the wiki we all can attend through the magic of archival time travel!! :D [4:04 PM]

Jennifer Pearl: yay [4:04 PM]

Astromon: Yay! [4:04 PM]

DragonairOmega: I am waiting for my scores on my RNA, I did not vote for any of my own designs. For future reference, was this a mistake? [4:07 PM]

Astromon: prob not [4:08 PM]

Elves: hi dragon! no worries, there are many labs. you can vote for your own designs, but the lab is able to synth most designs I think, so voting is not always necessary [4:08 PM]

Elves: voting selects which designs get synthesized if there are more submissions than slots [4:08 PM]

Astromon: i voted for 1 for you but unless to many slots are filled up (rare they will be tyested [4:09 PM]

Astromon: tested* [4:09 PM]

Elves: so you can look at the number of slots available for the particular lab, and see how many submissions there are. if more submissions than slots, go ahead and vote! [4:09 PM]

Elves: also perhaps in the future votes for high scoring designs are rewarded with points, but i don't know if that is plugged in at the moment [4:09 PM]

Astromon: great idea Elves i like that alot [4:10 PM]

Elves: :D me too! we can add it to the wish list. in fact i'll update the chat log to include this part. :) [4:11 PM]

Astromon: yes indeedy! [4:11 PM]