Hi, Here is the log of the chatroom at http://farsides.com/chat/ (or irc://irc.freenode.net/#farsides) for the 2011-09-03.
2011-09-03 00:41:23+0000 mtyaka (mtyaka) is now online 2011-09-03 00:45:05+0000 mtyaka_ (mtyaka_) is now online 2011-09-03 08:57:06+0000 MCMic (MCMic) is now online 2011-09-03 09:28:05+0000 mtyaka (mtyaka) is now online 2011-09-03 09:50:22+0000 antoviaque (antoviaque) is now online 2011-09-03 09:59:12+0000 arbrandes (arbrandes) is now online 2011-09-03 12:52:09+0000 <mtyaka> arbrandes, good morning! 2011-09-03 13:15:19+0000 dachary1 (dachary1) is now online 2011-09-03 14:19:00+0000 <arbrandes> mtyaka, hey dude, sorry, was away 2011-09-03 14:25:43+0000 <mtyaka> arbrandes: hey, no problem at all! sorry to bother you on a weekend :) i just wanted to discuss a couple of things about the vote states... 2011-09-03 14:25:55+0000 <arbrandes> mtyaka, no bother at all, I intend to work myself, ehhehe 2011-09-03 14:26:00+0000 <arbrandes> go ahead 2011-09-03 14:27:20+0000 <mtyaka> i was thinking we might not need to create a separate html div for the vote_anonymous state - as it would be basically the same as vote_voter. what do you think? 2011-09-03 14:27:41+0000 <mtyaka> this is related to task #419 2011-09-03 14:28:27+0000 <arbrandes> If there are no significant differences (and there probably aren' t), I agree. 2011-09-03 14:28:49+0000 * arbrandes thinking 2011-09-03 14:29:26+0000 <arbrandes> Well, you'd have to disable actual voting 2011-09-03 14:29:36+0000 <arbrandes> But depending on how you do it, it could be simple 2011-09-03 14:31:41+0000 <arbrandes> Is there a vote_waiting div? Looking 2011-09-03 14:32:24+0000 <arbrandes> Ok, there's voter_wait 2011-09-03 14:33:08+0000 <arbrandes> Wouldn't it make more sense to have vote_anonymous = vote_wait, instead of vote_voter = vote_wait? 2011-09-03 14:34:20+0000 <arbrandes> Or are we considering making them *all* the same div? 2011-09-03 14:35:15+0000 <mtyaka> hmm... interesting idea... 2011-09-03 14:36:20+0000 <arbrandes> There'd be a few if's in there 2011-09-03 14:36:53+0000 <arbrandes> The anonymous state would be different to what's shown in the video, anyway 2011-09-03 14:37:11+0000 <mtyaka> yeah... it might be a bit more difficult to keep the state of the DOM consistent if we keep reusing the same div 2011-09-03 14:37:37+0000 <mtyaka> ah, yes... there's no video for the anonymous views yet, so i'm only guessing what it is supposed to look like 2011-09-03 14:37:40+0000 <arbrandes> In the player's step 4 (vote_wait), the chosen card is maximized, but in anonymous, it wouldn't. 2011-09-03 14:38:03+0000 <mtyaka> right 2011-09-03 14:38:07+0000 <arbrandes> Otherwise, not much difference, I would think 2011-09-03 14:38:18+0000 <arbrandes> antoviaque, you around? ^ 2011-09-03 14:39:13+0000 <arbrandes> in any case, different div or not, the javascript for the anonymous view should be a lot simpler 2011-09-03 14:39:35+0000 <arbrandes> Just shows the six cards with a modal in the middle 2011-09-03 14:39:53+0000 <arbrandes> (different text for the modal, though) 2011-09-03 14:43:55+0000 <mtyaka> i'll think about it some more. i think we could at least reuse the same div for vote_anonymous and vote_wait, perhaps even vote_voter, but i am not sure about this one yet 2011-09-03 14:46:19+0000 <mtyaka> arbrandes: can I bother you some more? i am having mental blockage with regards to #417 (transition from pick_wait() to vote_voter()) - i thought i understood the problem, but now i am not sure anymore 2011-09-03 14:47:06+0000 <mtyaka> transition to vote_voter is possible from the invitation_pick_wait state.... what other states can transition to vote_voter? 2011-09-03 14:47:36+0000 <mtyaka> another possible transition (when the player doesn't pick a card in time) is from invitation_pick to vote_anonymous 2011-09-03 14:48:12+0000 <arbrandes> you could transition from pick to vote_anonymous, for instance 2011-09-03 14:48:18+0000 <arbrandes> exactly 2011-09-03 14:49:10+0000 <mtyaka> yes... but one you are in vote_voter, the only way you could get there was through invitation_pick_wait, right? 2011-09-03 14:49:18+0000 <mtyaka> s/one/once/ 2011-09-03 14:49:46+0000 <mtyaka> and once you are in vote_anonymous, you could only get there from invitation_pick 2011-09-03 14:49:52+0000 <mtyaka> are there other combinations? 2011-09-03 14:50:38+0000 <arbrandes> Can't think of any, but remember that any state can be accessed *directly* 2011-09-03 14:50:51+0000 <arbrandes> So transitions shouldn't be called then 2011-09-03 14:51:39+0000 <arbrandes> We need a mechanism that takes all these scenarios into account, and preferrably be something malleable 2011-09-03 14:51:50+0000 <arbrandes> What do you suggest? 2011-09-03 14:54:09+0000 <mtyaka> i was thinking how to implement your suggestion A) in #417 and probably the easiest way would be to set metadata on root in notify_active, since we are already calling that in every state - there would be very little changes that we would need to make 2011-09-03 14:56:02+0000 <arbrandes> Ok, so we save the active state there, and success() returning from poll() would check it, is that it? 2011-09-03 14:56:44+0000 <mtyaka> yes, i guess that's all there is to it 2011-09-03 14:56:49+0000 <arbrandes> (I believe only poll() should worry about transitions, as no other "asynchronous" transition seems to be possible) 2011-09-03 14:57:52+0000 <arbrandes> Unless there's some race condition somewhere, such as send_game() returning a different state than expected... but then, I say screw it for now :) 2011-09-03 14:58:18+0000 <arbrandes> The worse that'll happen is that the correct state will just pop in, instead of transitioning in. 2011-09-03 14:58:20+0000 <mtyaka> hehehe... i agree :) 2011-09-03 14:59:42+0000 <arbrandes> Ok, so poll() sets the state, poll().success() does something with it. 2011-09-03 14:59:59+0000 <arbrandes> Actually 2011-09-03 15:00:04+0000 <arbrandes> notify_active() sets the state 2011-09-03 15:00:07+0000 <arbrandes> sorry 2011-09-03 15:03:06+0000 <mtyaka> yes, notify_active sets the state... and then either poll().success() could check the current state, or we could examine the state in the state callback (in our case vote_voter) before calling notify_active with the new state 2011-09-03 15:03:48+0000 <arbrandes> Wait, notify_active should not be called more than once per state 2011-09-03 15:04:13+0000 <arbrandes> Ah, ok 2011-09-03 15:04:17+0000 <mtyaka> no, i didn't mean to suggest that... 2011-09-03 15:04:34+0000 <arbrandes> I see 2011-09-03 15:04:57+0000 <mtyaka> i just mean to say that before we call notify_active with the new state, the old state will still be present in the root metadata 2011-09-03 15:05:42+0000 <arbrandes> Ok, let's take the more complicated case, pick -> vote_anonymous 2011-09-03 15:06:09+0000 <antoviaque> hi! 2011-09-03 15:06:13+0000 * antoviaque reading :) 2011-09-03 15:06:19+0000 <arbrandes> hey antoviaque :) 2011-09-03 15:06:21+0000 <mtyaka> hi antoviaque :) 2011-09-03 15:07:24+0000 <arbrandes> insite vote_anonymous, you'd check if the previous state was "pick", and then call the transition function from there? 2011-09-03 15:07:33+0000 <arbrandes> *inside 2011-09-03 15:09:17+0000 <mtyaka> yes, the first thing you'd do inside vote_anonymous, would be to check the previous state, and then, depending on what the previous state was, proceed accordingly 2011-09-03 15:09:29+0000 <antoviaque> Yep, for the anonymous it's pretty similar - any help needed on this? 2011-09-03 15:09:57+0000 <arbrandes> antoviaque, well, more validation really, since there's no video, we're playing by ear 2011-09-03 15:10:47+0000 <antoviaque> sure - which parts would be useful to validate? 2011-09-03 15:11:51+0000 <arbrandes> mtyaka, sounds simple enough, and since we don't have that many transitions possible anyway, a more abstract approach may be over-engineering. 2011-09-03 15:12:02+0000 <antoviaque> It's basically the player view, minus the vote interaction 2011-09-03 15:12:35+0000 <antoviaque> and the illustration for results 2011-09-03 15:13:31+0000 <arbrandes> antoviaque, in this case, just two points: the player goes straight from picking to waiting, but with no card maximized, just a modal saying something. 2011-09-03 15:14:04+0000 <antoviaque> yep 2011-09-03 15:14:37+0000 <antoviaque> the text of the modal will be specific, true - it can be just "Players are trying to figure out which card is the card story" 2011-09-03 15:15:25+0000 <arbrandes> Ok, cool 2011-09-03 15:15:53+0000 <mtyaka> what about the transition from pick to vote_anonymous? it probably isn't going to be very smooth... do we care? 2011-09-03 15:16:20+0000 <arbrandes> In previous conversations with antoviaque, I believe we agreed that it's ok 2011-09-03 15:16:27+0000 <arbrandes> (ok to not be smooth, I mean) 2011-09-03 15:16:38+0000 <mtyaka> on the second thought, the player should probably know what happened 2011-09-03 15:16:48+0000 <mtyaka> how do we inform him he was too late? 2011-09-03 15:17:12+0000 <antoviaque> yes, it can't be worse than what is happening right now in the game - where the screen just becomes all blank : ) 2011-09-03 15:17:40+0000 <antoviaque> brb - going to grab some groceries 2011-09-03 16:06:52+0000 <antoviaque> back 2011-09-03 17:35:34+0000 MCMic (MCMic) is now online 2011-09-03 17:52:27+0000 arbrandes (arbrandes) is now online 2011-09-03 18:10:30+0000 MCMic (MCMic) is now online _______________________________________________ Farsides mailing list - [email protected] Wiki: http://farsides.com/ List: http://farsides.com/ml/ Forum: http://farsides.com/forum/ Ideas: http://farsides.com/ideas/ Chat: http://farsides.com/chat/
