Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2015-06-29

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

All times shown according to UTC.

Time Nick Message
12:44 jar286 joined #opentreeoflife
13:19 kcranstn joined #opentreeoflife
13:42 jar286 jimallman, you might be interested in https://github.com/OpenTreeOfLife/reference-taxonomy/pull/148  which has a new doc file for ott 2.9
13:43 jar286 I don’t think meaningful review is really possible… suffice to say it passes the inclusion tests and shows deprecation of less than ten OTUs
13:43 jar286 probably one or two more iterations before releasing 2.9
13:43 jimallman new doc looks nice, thanks jar286!
13:45 jar286 I think it should probably be merged
13:46 jimallman would you like this (and similar versions) to appear on the Progress page? i suppose ideally we would only show taxonomy versions once they’re used in the synthetic tree.
13:46 jar286 we don’t wait to show curation progress for it to show up in the synthetic tree, i don’t see why taxonomy should be different
13:47 jar286 that makes me wonder if we want to track the intermediate taxonomy drafts…  I don’t make separate doc pages for them
13:47 jar286 I don’t think so
13:48 jar286 hang on, draft5 is much better, let me add that to the PR
13:48 jimallman gotcha, that was my question (whether intermediate taxo drafts would become part of the visible “version history”)
13:49 jar286 I don’t think so
13:57 jar286 ok, the PR is now for draft 5. I think it should be merged to master.  But I’ll let you do the review & merge.  There is a small thing, the doc file still says ‘draft 4’, but I’ll fix that later
13:59 jimallman ok, will review shortly. thanks!
14:03 jar286 thanks
14:04 jimallman meanwhile, i’m working on a summary of how the “raw” stats are processed into the current Progress visualization. i’ll add this as an issue for others to review. i’m hoping this will act as a sanity check in case i’m misinterpreting pmidford’s stats or misrepresenting the results.
14:04 jar286 gd
14:42 kcranstn nice new doc file!
15:55 josephwb joined #opentreeoflife
15:55 josephwb jar286 is there any reason to not include ottids in the test files (e.g. monophyly.csv)?
15:56 josephwb since they are permanent, seems silly to look them up
16:07 kcranstn I think mtholder asked for the same thing. Just a sec...
16:08 kcranstn https://github.com/OpenTreeOfLife/germinator/issues/15
16:08 josephwb ok, thanks
16:09 kcranstn if we end up taking requests from community, will need to figure out how to get ottids from the names in their feedback
17:02 jar286 no problem really. we should think about what should go in these files.  maybe a comments column, in addition to the url. and the name should be there in addition to the ott id
17:02 jar286 we still have no policy for ott id stability, btw, so it’s not obvious what these tests mean
17:03 jar286 we would really like to put clade hypotheses in the test files
17:03 jar286 (maybe OTT ids name clade hypotheses)
17:04 kcranstn I think there is difference between how we ask people to describe the tests they want (i.e. their expectations for the tree) and how we code these for use in a script. This file is trying to be both.
17:04 kcranstn or, I hypothesize that this is true
17:09 jar286 that sounds right.
17:10 jar286 we also need to distinguish between regression tests and aspirations.  I’ve put some tests in the file that are things that have been requested, but do not work yet. I think this is useful but it makes the results hard to interpret
17:12 kcranstn there will be lots of things in the synthetic tree that we know won’t pass
17:13 kcranstn this is a way of acknowledging the errors / omissions and hopefully getting trees with the missing data
17:15 jar286 yes, but I’d say both functions are important (regression tests and aspirations). don’t break the past, improve the future
17:16 kcranstn agreed
20:28 kcranstn joined #opentreeoflife

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