01:38:19 <Gretell> Unstable branch on crawl.develz.org updated to: 0.25-a0-937-gcfae464cf3 (34)
02:10:13 <alexjurkiewicz> is Sequell meant to be really slow for normal-ish !lg queries?
02:10:19 <alexjurkiewicz> i thought
02:10:24 <alexjurkiewicz> just !lm was bad
02:10:42 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.25-a0-937-gcfae464cf3
02:20:54 <gammafunk> Sequell is having issues, yeah
04:07:58 <Kelbell> Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-3054-g1a545eca1e
04:52:38 -!- Raichvent_ is now known as Raichvent
04:57:24 -!- Raichvent_ is now known as Raichvent
05:02:03 <Kelbell> Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-2667-gd8c717c28d
07:54:27 <flt_> hi all
07:54:34 <flt_> is one of the devs present?
07:55:37 <flt_> basically i want to file a bug report to mantis, but for that i need to be able to login first
07:56:01 <flt_> who can create an account for me?
08:01:00 <ebering> flt_: your best bet is to file one on github
08:01:09 <ebering> (I am a dev but I don't have the permissions to create mantis accounts)
08:08:09 <flt_> you mean, as a pull request an then mark it somehow as bug report?
08:09:27 <flt_> yes sir, can do :)
08:11:42 <alexjurkiewicz> there's an "issues" tab
08:11:49 <alexjurkiewicz> next to "pull requests"
08:54:39 <advil> flt_: https://github.com/crawl/crawl/issues/new/choose
08:54:55 <advil> oh maybe you already did it
09:03:06 <aidanh> re ttyrecs, maybe it makes sense to ask the player after their game finishes whether they want to save it
09:03:17 <aidanh> defaulting to 'no' of course
09:04:53 <aidanh> that won't help solve the existing problem, but it will at least stop it from getting any worse
09:06:39 <flt_> yes, i did, thanks for the link anyway :)
09:06:42 <flt_> yes, i did, thanks for the link anyway :)
09:14:28 <MikeHollisJr> is there a better question than here to ask about sequell issues (installation, not public sequell)
09:14:54 <aidanh> a better place, you mean? here is probably a good start
09:17:24 <MikeHollisJr> okay well I've got sequell running, and for whatever reason its not retrieving info from the learn database, I can set entries but when I try to retrieve them, I get a message that the subprocess exploded, and in the console it complains about not being able to load re2.so despite having been installed .
09:19:26 <MikeHollisJr> I'm getting the relevant lines from the sequell log now
09:21:46 <advil> well, to be clear, there probably isn't a better place but there's also not necessarily a lot of on-the-ground experience running Sequell from people who are active in here these days
09:22:01 <MikeHollisJr> https://pastebin.com/A6iazFiq
09:22:07 <MikeHollisJr> that's fair enough
09:22:15 <advil> I think amalloy_ got it running some time last year just as a test?
09:22:54 <MikeHollisJr> I'm pretty sure this isn't sequell so much as Re2 being stupid, problem is I know nothing about how it's actually supposed to work let alone what it's doing
09:23:51 <advil> although every time I use ruby these days I seem to spend 90% of my time debugging version issues like this, so yeah maybe it won't require specialized knowledge
09:24:53 <MikeHollisJr> any help would be appreciated,
09:24:53 * aidanh did not realize that sequell was written in ruby
09:25:40 <advil> heh it's 2/3 ruby, 1/3 perl
09:26:00 <MikeHollisJr> at least it's dependancies seem to be
09:26:02 <aidanh> classic
09:27:58 <advil> MikeHollisJr this is a very sketchy guess but maybe you have re2 installed correctly but not relative to the bundler setup for sequell?
09:28:57 <advil> have you checked if that .so is somewhere around `/usr/local/rvm/gems/ruby-2.7.0/bundler/gems/re2-7184c40f16d8/lib` (not sure where it's supposed to be?)
09:29:36 <MikeHollisJr> I can
09:30:49 <Cheibriados> 03Aidan Holm02 07[remove-old-monsters] * 0.25-a0-938-g31db3c4: Re-sync axed / removed monsters 10(10 days ago, 2 files, 9+ 0-) 13https://github.com/crawl/crawl/commit/31db3c4590f7
09:30:49 <Cheibriados> 03Aidan Holm02 07[remove-old-monsters] * 0.25-a0-939-g743b897: Add some basic tests 10(10 days ago, 2 files, 23+ 0-) 13https://github.com/crawl/crawl/commit/743b897fbabe
09:30:49 <Cheibriados> 03Aidan Holm02 07[remove-old-monsters] * 0.25-a0-940-gf7e96df: Remove dead monster code 10(10 days ago, 8 files, 4+ 43-) 13https://github.com/crawl/crawl/commit/f7e96df330d3
09:30:49 <Cheibriados> 03Aidan Holm02 07[remove-old-monsters] * 0.25-a0-941-g2248984: Show names of removed monsters 10(2 days ago, 2 files, 113+ 107-) 13https://github.com/crawl/crawl/commit/2248984e4469
09:31:07 <MikeHollisJr> it's actually not, a symlink might do the trick correct?
09:31:50 <advil> not sure, that could indicate that there are going to be other issues too
09:32:15 <MikeHollisJr> actually
09:32:18 <MikeHollisJr> that was all it needed
09:32:44 <MikeHollisJr> no sub process explosions and I'm getting the correct information from the test entry :D
09:33:12 <MikeHollisJr> thank you so much, I figured it was something that silly
09:36:03 <advil> my next suggestion was going to be literally "reset your ruby install" so I'm glad that worked :-P
09:36:34 <advil> (that's seriously what I've ended up doing for jekyll in the past)
09:36:53 <MikeHollisJr> me too because my followup question would have been how do I do that, and I don't want to bug you guys to that level
09:37:10 <Cheibriados> 03Quipyowert202 {Aidan Holm} 07* 0.25-a0-938-gad7ffe8: Fix resource leak if allocating PNG structs fails. 10(32 hours ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/ad7ffe8aff14
09:38:26 <MikeHollisJr> I do have one cerebot question though, (maybe it's sequell) I can't seem to add to the learndb via cerebot, it complains about not being able to add to the db via proxy, any idea if that's a settting somewhere or requires code modification
09:52:15 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.25-a0-938-gad7ffe8aff (34)
09:56:18 <MikeHollisJr> nevermind I think I  may have found the documentation for that part
09:59:18 <MikeHollisJr> Yup I'm good, thank you! have a great day everyone
10:01:39 <ebering> Hi everyone, in preparing for the 0.25 tournament we're looking at overhauling the tournament scoring system. Here is the current draft of what we're thinking of. I would appreciate both general impressions and specific feedback on the point values in the "banner score" section.
10:01:44 <ebering> https://github.com/crawl/crawl/wiki/New-Tournament-Scoring
10:01:49 <ebering> oops, I only meant to paste the link
10:02:03 <ebering> I've carved the sketch into a draft, some of the flavor text is still missing
10:02:13 <ebering> but I'll circulate this in player forums and see what comes back
10:02:33 <ebering> Once thesse finals are graded I'll have time to think about implementation
10:03:14 <advil> are you thinking to build on the csdc scripts instead of tournament scripts?
10:03:32 <ebering> possibly; the latter already do clans but I understand the former better
10:03:57 <advil> the tournament scripts aren't too hard
10:04:45 <advil> actually I would probably say start with them, they have a bunch of auditing stuff that is fairly carefully implemented
10:04:45 <advil> plus a lot of work that I think elliptic did on the scoring
10:04:45 <ebering> yeah ok
10:04:47 <ebering> I just haven't put in the work to read them well
10:04:56 <ebering> they seemed overpowered for what I wanted to do for csdc
10:04:58 <advil> are the csdc scripts based on the general scoring scripts?
10:05:03 <ebering> yes
10:05:05 <ebering> well
10:05:15 <ebering> the "new scoring" written by zxc, chequers, and kramin
10:06:45 <ebering> could you pin the reddit thread I just made
10:08:32 <advil> tournament scripts were forked off scoring at some point too I think, so the overall structure is not so different
10:08:44 <advil> oh
10:08:52 <advil> new scoring was very different
10:11:35 <ebering> yeah
10:15:27 <Cheibriados> 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-939-g47ca1c5: Allow any floor to be annotated with ! 10(2 weeks ago, 7 files, 194+ 53-) 13https://github.com/crawl/crawl/commit/47ca1c57af0f
10:15:27 <Cheibriados> 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-940-gc4e05f1: Fix whitespace after merge conflict 10(2 days ago, 2 files, 1+ 3-) 13https://github.com/crawl/crawl/commit/c4e05f18608f
10:15:27 <Cheibriados> 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-941-gfd5b02c: Unbrace. 10(2 days ago, 1 file, 0+ 2-) 13https://github.com/crawl/crawl/commit/fd5b02ce9275
10:15:27 <Cheibriados> 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-942-gc9daeeb: simplify code, and add space to prompt 10(2 days ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/c9daeeb06943
10:15:27 <Cheibriados> 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-943-ga7de50f: Implement reviewer comments 10(2 minutes ago, 2 files, 8+ 16-) 13https://github.com/crawl/crawl/commit/a7de50f892e1
10:43:50 <MikeHollisJr> I am so close, I can't get cerebot to identify on freenode, once I nail that I've got it, is there any information related to the configuration options for authentication for cerebot?
10:44:25 <MikeHollisJr> I've tried the ones in the file in all sorts of combinations, and I've added the mask to nickserv, it just is being stubborn
10:44:37 <MikeHollisJr> and unfortunately cerebot doesn't give a lot of hints as to what is going on
10:46:31 <advil> that I would ask gammfunk (who is not in channel right now)
10:47:28 <MikeHollisJr> okay will do, I'll keep an eye out for them
10:47:31 <MikeHollisJr> thank you
10:48:20 <MikeHollisJr> cd ..
10:48:20 <MikeHollisJr> l
10:48:20 <MikeHollisJr> oops sorry
10:48:20 <advil> haha
11:06:44 <elliptic> ebering: a couple comments on the tourney overhaul plan:
11:10:07 <elliptic> first, I suspect the banner point system is much more complicated than is really ideal for something that is only meant to be a small part of the overall scoring
11:10:11 <elliptic> i.e. you are spending way too many "complexity points" in that part of the rules
11:10:43 <elliptic> I would switch to something where tier 1/2/3 each have a fixed point value
11:10:45 <MikeHollisJr> ugh, cerebot issue? forgot to verify the email for nickserv registration lol talk about simple
11:11:28 <elliptic> and just not worry about some banners being similar to other ranked categories or being harder than others or whatever
11:12:45 <elliptic> partly I think the system you have has way more little details than necessary, and partly I feel like "banner points" should be tied to the actual cosmetic banners appearing on player pages
11:13:08 <elliptic> so you don't have weird things where one player has more/shinier banners but another player has more points
11:14:19 <ebering> hm. I want the banner category to be a way for players who win maybe once to see their ranking go up and down and have smaller scale competitions (advil brought this up), and a coarser banner scoring system will mean these players are all in a "close last tie"
11:14:52 <elliptic> right, so that brings me to my second comment
11:15:33 <elliptic> which is that I strongly think that more of the overall ranking categories should be accessible without needing to win games
11:15:42 <advil> I would support that :-)
11:17:07 <elliptic> at the moment you have 9 categories listed and 6 of them require a win, while 2 of them are specifically about score and the last one is banners, which I don't really view as something that solves this
11:17:39 <elliptic> (also I suspect yo uare wrong about a coarser banner scoring system making a big difference, since there are a decent number of relatively easy banners that don't require winning)
11:18:25 <elliptic> anyway I would add some categories for incremental achievements like # of distinct uniques killed
11:18:50 <elliptic> or # of distinct locations entered
11:19:34 <ebering> how about "uniques + pghosts" and an ash style "locations entered + (weighted) runes collected" category
11:20:20 <Cheibriados> 03Aidan Holm02 07* 0.25-a0-939-g1e880d3: Fix show_map() sometimes falsely choosing target (#12250) 10(2 minutes ago, 2 files, 15+ 0-) 13https://github.com/crawl/crawl/commit/1e880d33e3b7
11:22:02 <elliptic> is pghost frequency/difficulty tied at all to the server people play on? I don't remember exactly what the resolution to that was
11:22:22 <advil> frequency is not, difficulty may be
11:23:20 <advil> that at least should be a lot more subtle if so than it used to be, but I'm not sure anyone's ever really looked closely since the reform
11:23:54 <elliptic> anyway in general I'd err on the side of making more things overall categories
11:24:02 <elliptic> rather than having them just be banner points
11:24:21 <elliptic> this isn't just for "easier" achievements, but also stuff like zigs
11:26:52 <elliptic> looking at the banners, I think most of the things that you have listed as smoothly increasing banner points could/should be changed to overall ranked categories
11:28:50 <elliptic> that's 6 I guess, ABCEXY
11:37:01 <elliptic> might still be good to brainstorm more categories that start to differentiate players well before they get a win, but then there would at least be 6: combo high scores, best high score, banner score, locations/runes, ****** piety, killing uniques
11:38:33 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.25-a0-939-g1e880d33e3 (34)
11:40:04 <elliptic> this doesn't address that issue, but "low XL win" is another category that could be good to have
11:44:06 <elliptic> I guess my third comment is that I think that it's probably best to have every ranking category still matter for clan scoring, with separate clan leaderboards (and computing achievements that can be split over multiple games/accounts on the clan level, e.g. combo high scores rank clans based on the sum of the # of combo high scores rather than the max
11:44:31 <elliptic> oh, I just noticed that you aren't having actual clan rankings for each category
11:45:53 <elliptic> I would basically just use rankings for everything, so each ranking category has separate leaderboards for individual and clan
11:46:34 <elliptic> and then use some sort of weighted power mean (e.g. harmonic mean) for overall clan ranking just like for overall individual ranking
11:49:01 <elliptic> one of the main appeals to me of this sort of overhaul is having actual leaderboards for everything so that people can see "my clan is 19th for combo high scores but only 97th for nchoices won" and such
11:49:43 <elliptic> rather than just showing the top 3-5 players in each category and an overall ranking
11:52:23 <ebering> yeah, I was planning to rank the clans in each category and combine, where the clan rating function per category can be more interesting than min()
13:35:50 <ebering> elliptic: I've updated things to incorporate your suggestions
13:36:10 <ebering> erm. I must push
13:53:12 <Henzell> Orph (L15 DrFE) Crash caused by signal #15: Terminated (D:14)
13:54:16 <Henzell> sullage (L9 BaFi) Crash caused by signal #15: Terminated (D:6)
13:57:16 <advil> !crashlog
13:58:49 <Sequell> 90s limit exceeded: killed !crashlog
14:26:54 -!- Tiobot is now known as Guest60115
16:07:54 <Bcadren> !tell floraline gorgo's current bcadrencrawl save is broken to the point of crashing while trying to load. I don't think I can fix it (at least not quickly) and they asked to just have it deleted. Can you remove the file?
16:08:11 <Bcadren> Sequell dead?
16:10:23 <Sequell> Bcadren: OK, I'll let floraline know.
16:22:00 <ebering> gammafunk: elliptic suggested that differentiating banner tier points has too much complexity overhead and that its better to just have them all level
16:22:42 <gammafunk> ebering: yeah, I read through that scrollback, and I need to read through it again actually
16:22:46 <ebering> it's ok that they have different difficulties, since eventually to move up in the banner score category a player will need to attempt the more challenging banners, and the relative rank is all that matters for the overall contribution
16:23:13 <gammafunk> I can definitely get behind the simplification idea
16:23:34 <gammafunk> I suppose one sort of downside is that we can expect people to say a lot of "it's weird that this banner which is harder is worth the same points"
16:23:46 <gammafunk> but hopefully we can have a nice concise explanation on the rules page to link to
16:23:53 <gammafunk> including a brief bit about why what you said is ok
17:10:58 <floraline> !tell bcadren done
17:11:00 <Sequell> floraline: You have 1 message. Use !messages to read it.
17:11:12 <Bcadren> Thanks.
17:11:13 <Sequell> floraline: OK, I'll let bcadren know.
17:11:14 <Bcadren> !messages
17:11:20 <Sequell> (1/1) floraline said (7s ago): done
18:22:42 <Lantell> Unstable branch on underhound.eu updated to: 0.25-a0-939-g1e880d33e3 (34)
18:59:05 <Cheibriados> New branch created: pull/1386 (1 commit) 13https://github.com/crawl/crawl/pull/1386
18:59:05 <Cheibriados> 03Nikolai Lavsky02 07https://github.com/crawl/crawl/pull/1386 * 0.25-a0-940-gcec8aca: Update the options guide 10(21 minutes ago, 1 file, 61+ 61-) 13https://github.com/crawl/crawl/commit/cec8aca52d58
19:08:49 <alexjurkiewicz> ebering: the points fo combo high scores are 1 for a combo highscore and 5 for a species or background highscore. I wonder if the latter should be much higher given there are 20+ of each species/background
19:10:16 <alexjurkiewicz> "The first 8 players to win a given Nemelex' choice combo earn a point in this category" <-- For players who can't win reliably, nem choice is basically a pointless part of the tournament, which is a shame as there should always be an incentive to play fun random combos. I wonder if there is some way to change scoring to encourage everyone to play the combos?
19:12:08 <gammafunk> well, the banners still exist for nem choice, even if you don't win
19:12:32 <alexjurkiewicz> right, but most banners would be easier achieved with non-nemchoice combos
19:12:51 <gammafunk> I don't understand what that means
19:12:58 <gammafunk> the banners for nemelex choice still exist, is what I meant
19:13:05 <gammafunk> so you get the banners for e.g. rune with nchoice
19:13:07 <gammafunk> or whatever those are
19:13:33 <gammafunk> I: Reach experience level 9 with a Nemelex' choice combo.
19:13:33 <gammafunk> II: Get a rune with a Nemelex' choice combo.
19:13:41 <alexjurkiewicz> ah right. Well true! I forgot about those
19:13:53 <alexjurkiewicz> ebering: "3 points per distinct rune of Zot discovered" <-- "collected" rather than discovered?
19:16:22 <alexjurkiewicz> ebering: the "Ziggurat Diving" description could probably do without the second sentence "Exiting a Ziggurat from the lowest floor counts as ..." because this clarification only applies to one tier. The Tier II text could read "exit a Ziggurat from the 10th or deeper floor" (if dying on zig:10 doesn't count??)
19:17:20 <alexjurkiewicz> maybe I'll write this feedback down, there are too many messages
19:20:09 <gammafunk> yeah, I'm not sure where suggestions are being collected; I sent him a formatted discord message
19:28:35 <ebering> I'm collating them as they come in from various channels
19:28:37 <ebering> there's a tavern thread
19:28:44 <ebering> and a reddit thread
19:55:06 <ebering> alexjurkiewicz: re zig diving, I mean to count megazig 1 floor 1 as zig:29
19:55:29 <ebering> as a category
20:05:12 <alexjurkiewicz> oh, nice idea
20:06:12 <alexjurkiewicz> about harmonic mean. I wonder if it makes sense to have *any* overall ranking. It might be interesting to haev each cateogry be its own thing. If someone wants to spend all tournament reaching zig level 300, cool! With a harmonic mean overall score, this would be pointless
20:06:40 <alexjurkiewicz> some people want to "win the whole tournament" though, so this idea would be bad for them
20:08:25 <ebering> part of the nice thing about the harmonic mean is that getting first in a category is really strongly weighted
20:08:45 <ebering> keep in mind that in a 2-category hm rating system (1, ∞) is tied with (2,2)
20:35:05 <gammafunk> unicode?! in my irc channel?!?!
20:37:56 <due> I swear I logged into this account last year
20:38:00 <due> but they've deleted it :|
20:38:32 <gammafunk> yeah, freenode did a purge of logins some time back; I forget if they do rolling purges now
20:39:03 <gammafunk> we can certainly give your new account voice
20:39:08 <due> well
20:39:25 <due> sure but that means I've lost like... my cosmetic mask and stuff
20:39:27 <due> jeez
20:40:27 <due> frustrating because it was definitely less than a year ago
20:46:37 <due> Oh wait, I still have access to the account I did all my commits from
21:07:11 <thurin> the purge was based on no usage in 120 days, not a year.
21:18:51 <due> yeah but
21:18:56 <due> they'd never done a purge before sooo
21:20:09 <gammafunk> I guess they wanted to reclaim the years and years buildup of dead accounts
21:27:38 <thurin> they normally happen every 4 years or so.  speaking from experience of having lost accounts before.
22:02:08 <alexjurkiewicz> your cosmetic mask!!!1
22:05:16 <due> eh