Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2014-07-17

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:45 josephwb joined #opentreeoflife
01:57 josephwb joined #opentreeoflife
04:01 josephwb joined #opentreeoflife
11:19 kcranstn joined #opentreeoflife
12:39 josephwb joined #opentreeoflife
13:18 josephwb joined #opentreeoflife
13:19 josephwb u there jimallman?
13:19 jimallman hi josephwb !
13:20 josephwb hu
13:20 josephwb hi
13:20 josephwb did we reach any consensus on the "Biologically correct" issue?
13:21 jimallman i went ahead with some minor revisions last night, but i believe karen is planning a broad review of this.
13:21 josephwb alright
13:21 jimallman unfortunately i broke devtree last night, still trying to sort it out.
13:25 josephwb no rush
13:26 josephwb just some empirical folks seem baffled by the "corect" language
14:05 towodo joined #opentreeoflife
14:08 towodo jimallman, i blew away ot11
14:08 towodo nobody objected to my email saying i’d do so. oops!
14:09 towodo so now devtree will be ot15
14:09 jimallman towodo: no problem. i was just chasing error messages on ot10, glad to have a fresh start.
14:10 jimallman just curious: why the move to a new machine? is it provisioned differently?
14:10 towodo should be ready to go now. give it a try
14:10 towodo it was a mistake
14:10 towodo i meant to get rid of ot3, and flushed ot11 instead.
14:10 towodo my email sent to ggroup aon July 14 evoked no response, it said ot11 and nobody noticed my mistake
14:12 jimallman yep, i see it now. *smacks forehead*
14:12 jimallman i’m getting “invalid request” at http://devtree.opentreeoflife.org/
14:12 towodo try ot15
14:13 towodo ssh into ot15 that is
14:13 towodo you should be able to start deployment. it’s a blank slate
14:13 jimallman ah, ok. will do.
14:14 jimallman yes, i connected via SSH and go the WARNING: POSSIBLE DNS SPOOFING DETECTED messages
14:14 jimallman (RSA host key has changed)
14:15 jimallman anyhoo, i’ll update the development system in deployed-systems, deploy, and commit the changes.
14:16 towodo damn i screwed up again
14:16 towodo there are two ot15s.
14:17 jimallman ok, standing by
14:17 towodo i think you can try ot16, i set that up too for asterales but it’s the same. hang on
14:17 jimallman i thought it was odd that i already had an entry for ot15 in my .ssh/config
14:22 towodo_ joined #opentreeoflife
14:23 towodo_ just added ot16 = 54.190.10.24
14:24 towodo_ jimallman^
14:25 jimallman ok, so this should replace ot11 (new devtree) .. but we’re still using ot10 for devapi, yes?
14:25 towodo_ correct
14:25 jimallman ok, deploying now..
14:26 towodo_ i will get rid of ot3, yes?  I can also go and edit the deployed-servers list, but I’ll create the asterales host first
14:38 jimallman towodo: yes, ot3 is no longer needed.
14:38 towodo tnx.
14:48 towodo jimallman, I’m going to edit opentree-servers.txt, ok?
14:48 towodo refreshing deployed-systems repo now.
14:50 jimallman sounds good. i’ll push up my changes in a moment
14:51 jimallman my changes to development/ have been pushed to origin/master
14:53 towodo ok, thanks
15:01 josephwb joined #opentreeoflife
15:03 jimallman towodo: ot16 (new devtree) is up and happy, or at least showing the same errors as before.  :-/
15:04 jimallman (missing packages on treemachine, i think)
15:04 towodo excellent.
15:04 towodo oh no, maven hell
15:04 towodo try building treemachine locally first
15:04 jimallman yes, i think so (gnu.trove.TLongCollection and others).. ah, ok
15:15 jimallman ah yes, my old friend AetherClassNotFound… upgrading the required maven-shade plugin from 2.0 to 2.1 fixes it.
15:27 jimallman ok, synth-tree view and study curation on devtree are looking good again. onward!
15:33 josephwb towodo: that list of asterales trees was lacking tree ids for most
15:34 towodo right, it’s not a synthesis list. it’s a study list.
15:34 towodo the synthesis list is in gcmdr
15:34 josephwb right
15:34 josephwb so, do you want to add to the synthesis list?
15:34 towodo no
15:35 towodo I want to subtract from both.  there was a really big study that was hogging space & time.
15:35 josephwb ok
15:35 towodo but i don’t know which it is
15:35 josephwb i'll look into it
15:35 towodo i could go through them one by one
15:35 towodo but if you’d look at it that would be fantastic
15:35 josephwb the synthesis list is smaller.
15:35 towodo yes, much.
15:36 towodo so start there.
15:36 josephwb if the big one is on that, it should be easy
15:36 josephwb asynthesis took ~ 2 minutes the other day
15:36 towodo i think it is but not sure.  i just remember trying to build the oti database and a 1.7G RAM machine ran out of memory
15:36 towodo so maybe it’s not  in the syntesis list.
15:37 josephwb checking them now
15:37 towodo oh… I know what might have happened. the test repo which was originally asterales only was expanded at some point to include all the studies
15:38 towodo and maybe i ran the asterales setup against all the studies instead of just the asterales ones
15:38 towodo but in any case, if there’s a monster in either list i don’t want it.
15:38 josephwb got it
15:38 josephwb synthesis: 1 min 39 sec
15:38 josephwb tree sizes coming
15:40 towodo could just look at file sizes… a little bit of python
15:40 josephwb none huge, biggest has 632 taxa
15:41 josephwb that is the big Soltis Angiosperm tree
15:44 josephwb pg_2539: 632
15:44 josephwb pg_715: 39
15:44 josephwb pg_329: 37
15:44 josephwb pg_9: 122
15:44 josephwb pg_709: 68
15:44 josephwb pg_41: 63
15:44 josephwb pg_82: 68
15:44 josephwb pg_932: 209
15:46 josephwb study 9 has a weird url (it lacks the "pg_" prefix). fixed that in the gcmdr script. entire analysis (building tax DB, downloading studies, loading, synthesis) now takes 57 sec on my machine towodo
15:47 josephwb erg. wait. broke. fixing now
15:48 towodo ok, maybe the problem was not what I thought, it only just occurred to me (see above) that it might have a difference source.   thanks for looking, the problem may be fixed by recovering the small repo.
15:50 josephwb re files sizes: some studies have one tree, some have many, so not perfect
15:56 josephwb towodo so study 9 is breaking things for some reason. commented out for now. hopefully good for you to proceed?
15:56 towodo yes.  thanks, will wait for mark to respond
16:06 jimallman josephwb: fyi - we’ve been treating the lack of pg_ prefix in study 9 as a one-time anomaly, with no ill effects. if that’s causing the problem, mark will definitely want to know.
16:06 josephwb jimallman: is taxon mapping broke?
16:07 jimallman hm, it does seem to be broken (all attempts fail) on devtree.. cody might be replacing a database though.
16:07 josephwb for study 1944, i am trying to map the only remaining unmapped taxon ("Fauria")
16:07 josephwb oh, ok
16:08 josephwb i was checking against ott2.8draft5. taxon is not present in ott2.6 (hence the failure)
16:08 josephwb well, present in ott2.6, but "barren" (so excluded)
16:08 josephwb READ: not your problem
16:09 jimallman gotcha. yeah, my mapping failures here were caused by a too-specific mapping context (“Land plants”).. it’s working fine.
16:15 josephwb towodo: do we have any plans for purging existing duplicated studies?
16:16 towodo get dois for all studies, then write a script?
16:24 josephwb right. some are good, some not.
16:25 josephwb shouldn't oti be able to generate a list?
16:26 jimallman yes, entering a DOI in the study list should show duplicates
16:26 josephwb i just fluked upon pg_1944 as a duplicate for pg_9
16:27 jimallman apologies, i’m trying to automate these tests for duplication (on study creation/import, and when assiging a DOI), but waiting on an oti fix.
16:29 josephwb towodo: using 1944 as a alternate to pg_9 works for the asterales example. not sure why… possible that the unprefixed PG studies were not processed fully by Mark's scripts?
16:29 josephwb ok, lunch
16:30 towodo thanks
17:01 towodo jimallman, I’m going to remove the taxomachine .jar file to force a build.  only touching taxomachine on ot10
17:01 jimallman towodo: no prob, thanks for the heads-up
17:20 towodo getting lunch now.
17:35 josephwb regarding deleting duplicate studies: should we not communicate that information back to PG?
17:39 jimallman josephwb: seems like it, yes. i think the medium-term goal is to sync pg’s db with phylesystem, right?
17:44 josephwb i thin kso
17:44 josephwb think so
18:31 kcranstn joined #opentreeoflife
18:56 travis-ci joined #opentreeoflife
18:56 travis-ci [travis-ci] OpenTreeOfLife/phylesystem-api#570 (master - 97ddd24 : Jim Allman): The build passed.
18:56 travis-ci [travis-ci] Change view : https://github.com/OpenTreeOfLife/phylesystem-api/compare/778a9db34709...97ddd2461930
18:56 travis-ci [travis-ci] Build details : http://travis-ci.org/OpenTreeOfLife/phylesystem-api/builds/30200349
18:56 travis-ci left #opentreeoflife
19:32 towodo joined #opentreeoflife
19:58 josephwb joined #opentreeoflife
20:02 josephwb joined #opentreeoflife
20:02 kcranstn @jimallman - do you know anything about the status of all of the cards on the “Supporting terser new-NexSON” list in trello?
20:02 kcranstn https://trello.com/b/BZHcxPN2/software
20:03 * jimallman is looking now...
20:05 jimallman kcranstn: sorry, i don’t think i understand this card. i thought perhaps it was about HoneyBadgerFish 1.2, but it doesn’t look like it to me. unless the first checklist item means that Phylografter should understand how to read and generate HBF 1.2..?
20:05 kcranstn look to the extreme right of the board - there is a whole list of cards
20:06 kcranstn which I was going to summarize into a single card, unless these are all completed
20:30 jimallman kcranstn: gotcha, looking there now…
20:35 jimallman kcranstn: I see, these *are* about HBF 1.2 .. I’ve updated the card about the curation tool’s JS, which still uses HBF 1.0. i’m afraid Mark will need to confirm which of the others are complete, but I *believe* he wrapped up most of these tasks. it’s likely that oti and treemachine are still working with older formats.
20:35 jimallman since we can transform between these formats quickly and without data loss, i believe we decided that it’s not a big problem if some tools are slow to adopt HBF 1.2.
20:36 kcranstn ok, I’ll just combine all of them into a single card and tag mark. Thanks!
21:03 towodo joined #opentreeoflife
22:53 kcranstn joined #opentreeoflife
22:57 kcranstn joined #opentreeoflife
22:57 jimallman joined #opentreeoflife
22:57 scrollback joined #opentreeoflife

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary