Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2014-08-21

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

All times shown according to UTC.

Time Nick Message
00:27 jimallman joined #opentreeoflife
01:13 kcranstn joined #opentreeoflife
01:22 jimallman hi kcranstn! i can push the new taxomachine and opentree (just the quick OTU mapping fix) to production, if you like.
01:22 kcranstn yes, please
01:24 jimallman ok, i should probably update my code to recognize Cody’s new arguments in the mapping service. will test that locally before attempting a push.
01:47 ilbot3 joined #opentreeoflife
01:47 Topic for #opentreeoflife is now Open Tree Of Life | opentreeoflife.org | github.com/opentreeoflife | http://irclog.perlgeek.de/opentreeoflife/today
02:19 jimallman Tested on devtree (looking good) in modified Cavia. I’ve updated production taxomachine from branch master, with a standing pull request to merge in these recent changes to the proper Cavia branch: https://github.com/OpenTreeOfLife/taxomachine/pull/58
02:20 jimallman I’ve also pushed the new Cavia branch of opentree (webapps) that has the cherry-picked stricter OTU mapping. So far, so good!
02:22 kcranstn joined #opentreeoflife
10:56 kcranstn joined #opentreeoflife
11:51 kcranstn joined #opentreeoflife
11:58 kcranstn jimallman - you there?
13:01 jimallman kcranstn: here now
13:01 kcranstn just about to leave to walk into nescent. will ping you when I get there.
13:04 jimallman ok, talk to you then
13:29 josephwb joined #opentreeoflife
13:46 kcranstn joined #opentreeoflife
15:54 towodo joined #opentreeoflife
15:54 towodo I don’t understand why colloquy sometimes automatically rejoins room, and sometimes doesn’t
16:00 josephwb towodo: Adium for mac works fine
16:01 josephwb using Pidgin on Linux at the moment
16:01 jimallman towodo: it generally works for me, if the room was open when i last quit. though it can take a minute to re-establish contact.
16:01 towodo but adium is so ugly, and has an annoying logo
16:02 towodo ‘when you last quit’ - maybe that’s the problem, I never quit it, I just shut my laptop.
16:03 kcranstn I use colloquy on mac
16:04 towodo right, I’m using colloquy on mac, and often it doesn’t reestablish the session after I close & reopen the laptop
16:05 josephwb that's annoying
16:06 towodo meaning I miss all the excitement
16:12 jimallman and then when you try to catch up with the latest chatter, the IRC log tricks you with its (apparent) use of Greenwich Mean Time… lots of today’s messages end up in “yesterday”
16:12 jimallman (log is still very handy, though)
16:49 jimallman josephwb: i’ve pushed a minor change to production (Cavia). as we’ve discussed, this ignores the is_perfect_match property and simply applies any *single* match that is returned. (zero matches, or multiple matches, are treated as a failed mapping.) please take a look and let me know if this will do as a short-term fix.
16:50 josephwb will do when i get a minute
16:50 josephwb :-D
17:42 kcranstn joined #opentreeoflife
19:35 josephwb hey jimallman. finally found some time to do some testing
19:35 josephwb found the first synonym i tried
19:35 josephwb but then i unmapped a good taxon, and it could not be remapped
19:35 jimallman ! which study + taxon?
19:36 josephwb http://devtree.opentreeoflife.org/curator/study/edit/pg_2628?tab=otu-mapping
19:36 josephwb Sarcophaga giganta
19:36 josephwb i have it open on "edit" is that a problem fo ryou to view?
19:37 jimallman nope
19:37 josephwb Same for Sarcophaga setinervis
19:37 jimallman it works for me (un-mapping, then re-mapping it)
19:37 josephwb good taxon
19:37 josephwb context = Insects?
19:38 jimallman nope, i was on All life
19:38 josephwb i am only finding the synonyms!
19:38 josephwb nothing for good taxa
19:38 josephwb *trying all life*
19:39 josephwb works with all life, but not insects
19:39 josephwb is this a cody problem?
19:39 josephwb weird that it finds synonyms under the specific context, but not the good direct matches
19:40 jimallman i see the reason it’s failing, but not the underlying cause (yet)
19:41 jimallman in each case, we’re getting two results (matches), which we treat as failed mapping
19:41 jimallman (i can help you see this in the debug tools. what’s your web browser?)
19:43 jimallman odd, it looks like one match is the normal/expected one, and the other is marked as a synonym
19:44 josephwb yes, cody and i are getting that too
19:44 jimallman all other properties are identical (maybe this is what Cody’s working on next, proper responses for synonyms)
19:44 josephwb right
19:45 josephwb we can get around this by getting unique hits only
19:45 josephwb but weird that it works under one context, but not the other
19:46 josephwb cody is building a DB at the moment. when he is done we can do some testing.
19:51 jimallman ok, i’ve added a ticket in the meantime: https://github.com/OpenTreeOfLife/taxomachine/issues/62
20:01 josephwb sounds like we have a solution
20:37 codiferous joined #opentreeoflife
20:37 codiferous jimallman, are you there?
20:37 jimallman codiferous: howdy
20:37 codiferous hi there
20:38 codiferous i'm thinking about removing the is_homonym property from things
20:38 codiferous we don't have a clear definition of homonyms
20:38 codiferous but i'm wondering if you use it for anything?
20:39 jimallman meaning, in which context?
20:39 jimallman no, i’m pretty sure i never check this (grep’ing now)
20:39 codiferous yes, or whether synonyms should also be considered homonyms
20:39 jimallman oh, good one.
20:39 codiferous and since jonathan added the unique name field a while ago, i'm not sure we need to wade into the muddy waters of homonym definitions
20:39 jimallman yeah, is_homonym never appears in the webapp code
20:40 codiferous ok
20:41 codiferous then i will just make it go away. we can revisit later if we want it back for some reason
20:41 josephwb towodo: have you thought more about having a permanent url for the most recent version of ott?
20:42 towodo no. you want to look up how to enable .htaccess in apache?
20:43 towodo we need to change the vhost file for files.opentreeoflife.org to enable .htaccess
20:43 towodo then create a .htaccess with the redirect in it
20:43 towodo not too hard, just a hurdle
20:44 towodo but I’m still not sure of the use case.  what is ‘most recent’ for one purpose may not be the right ‘most recent’ for another.
20:44 towodo taxonomies have to be vetted for particular purposes, and so on.
20:47 josephwb aren't people only ever going to want the most recent "release" version?
20:48 towodo No.  e.g. experimental syntheses should be built with the most recent experimental taxonomy.
20:48 josephwb then, a url for the most recent "bleeding edge" taxonomy too?
20:48 towodo production and development could be at different taxonomy releases
20:49 josephwb yes
20:50 towodo It’s easy to imagine that production could even be behind the current “release”
20:51 towodo I think what you want is services that give the URLs of the “release” and “bleeding edge” versions
20:51 towodo maybe
20:51 towodo I don’t know.
20:58 josephwb alright. i thought this would be a convenience for users, but it seems clear you are resistant to the idea. i'll give it up.

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