Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2015-05-13

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

All times shown according to UTC.

Time Nick Message
00:45 codiferous joined #opentreeoflife
00:49 codiferous joined #opentreeoflife
01:09 kcranstn joined #opentreeoflife
01:10 josephwb joined #opentreeoflife
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
01:51 jar286 joined #opentreeoflife
03:20 kcranstn joined #opentreeoflife
08:37 mtholder joined #opentreeoflife
09:55 josephwb joined #opentreeoflife
10:37 ilbot3 joined #opentreeoflife
10:37 Topic for #opentreeoflife is now Open Tree Of Life | opentreeoflife.org | github.com/opentreeoflife | http://irclog.perlgeek.de/opentreeoflife/today
11:36 jar286 joined #opentreeoflife
11:54 jar286 joined #opentreeoflife
12:25 mtholder joined #opentreeoflife
13:09 kcranstn joined #opentreeoflife
13:44 kcranstn joined #opentreeoflife
13:52 mtholder joined #opentreeoflife
14:36 kcranstn joined #opentreeoflife
15:24 jar286 joined #opentreeoflife
15:42 kcranstn_ joined #opentreeoflife
15:44 kcranstn joined #opentreeoflife
15:47 mtholder joined #opentreeoflife
15:49 jar286 joined #opentreeoflife
16:19 jar286 joined #opentreeoflife
16:22 jar286 joined #opentreeoflife
17:14 kcranstn joined #opentreeoflife
19:00 kcranstn joined #opentreeoflife
19:03 kcranstn thanks for making those changes to the doc, jar286
19:03 kcranstn going through them now
19:05 jar286 sure thing
19:08 kcranstn what’s the opposition to the sentence “These properties (stored in the node structure) allow us to easily compare ingroups and outgroups of nodes without needing to traverse the graph. “ re: ingroup and outgroup properties
19:08 kcranstn I found it helpful to know this
19:08 kcranstn otherwise, defining these as “properties” was confusing (why not just refer to the ingroup and the outgroup)
19:11 jar286 the opposition is that there is no *other* way to compare ingroups and outgroups
19:12 kcranstn ?
19:12 kcranstn storing vs collecting via traversal
19:12 jar286 there is no way to collect via traversal.  the information isn’t in the graph
19:13 kcranstn because there are no edges
19:13 jar286 more specifically, the ingroup can be found by tracing the edges to the ends, but the outgroups can’t be, because the outgroup ‘property’ is a subset of the unreachable tips
19:13 kcranstn ok
19:14 jar286 maybe the word ‘property’ is misleading?
19:15 jar286 hmm, the ingroup property isn’t the same as what’s reachable, anyhow.
19:15 jar286 I take back what I said
19:15 jar286 you can reach lots of tips that aren’t in the ingroup
19:15 kcranstn again, this is implementation details leaking through into descriptino
19:16 kcranstn description
19:16 jar286 that was my objection to the sentence
19:16 jar286 I don’t really like the word ‘property’, maybe it should be excised.  can just say that every node has an ingroup and an outgroup?
19:17 jar286 but I don’t think that matters much
19:18 kcranstn I’d be ok with s/[ingroup|outgroup] property/[ingroup|outgroup]/
19:22 kcranstn I suppose the real question is whether it is important to know that the ingroup and outgroup are stored as a property of the node vs being able to get that list from the graph
19:24 kcranstn “The tree alignment graph (TAG) is a multigraph whose edges are labeled with tree identifiers."
19:24 kcranstn can I replace ‘tree identifiers'
19:24 jar286 well, maybe, since I remember wondering about this at one time, but if nothing is said, the reader can’t assume they’re redundant with the graph, and that’s fine
19:24 jar286 labeled w’ith tree indexes’? ‘trees’?
19:25 kcranstn one label per edge?
19:25 kcranstn that was a question, not a suggested replacement text
19:25 jar286 it’s important to know that the label identifies (or is) a tree, I think.
19:26 kcranstn what about OTT edges?
19:26 jar286 OTT is a tree… but I suppose one could say ‘taxonomy or input tree index / identifier’
19:27 kcranstn but one id per edge, or multiple?
19:27 jar286 one id per edge. it’s a multigraph, many parallel eges.
19:28 kcranstn can we simple say “labeled with the source tree or taxonomy"
19:28 jar286 yes, that’s fine, but someone changed this when I wrote it
19:28 jar286 (and we’re saying ‘input tree’ not ‘source tree’ for some reason)
19:29 kcranstn The tree alignment graph (TAG) is a multigraph whose edges are labeled with the source information (input tree or taxonomy)
19:29 jar286 that’s fine by me
20:20 kcranstn Looking at the first paragraph of “Deriving the TAG"
20:24 jar286 I need to get my wife’s car from the mechanic, leaving in a few minutes
20:24 kcranstn ok
20:26 jar286 I’m beginning to get a sense of how it all works. not sure whether to raise the question of why the TAG is what it is again (first principles analysis intended to draw useful insights out of the designers’ heads)
20:26 kcranstn I’ll be around later tonight, and then intermittently on irc tomorrow and friday
20:26 kcranstn I like your changes, though
20:27 jar286 someone has to be decision maker for all of those change suggestions, might as well be you (hah)
20:27 kcranstn I am starting with accepting all of the grammatical changes to clean things up
22:55 jar286 joined #opentreeoflife
23:24 jar joined #opentreeoflife

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